<![CDATA[The Roundup]]>

<![CDATA[The Roundup]]>https://tsgconsumer.substack.comhttps://substackcdn.com/image/fetch/$s_!4vOK!,w_256,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F8e126c4b-4287-47c1-b3ce-b900d7f6f542_1068x1068.pngThe Rounduphttps://tsgconsumer.substack.comSubstackWed, 02 Sep 2026 23:18:42 GMT<![CDATA[The Roundup by TSG Consumer - Issue #32]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-8a6https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-8a6Tue, 04 Aug 2026 16:55:32 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

Why Are Berries Everywhere, in Every Season? Driscoll’s (Julia Moskin, The NYT): If you’re a regular grocery shopper, it’s hard to miss the proliferation of Driscoll’s berries, with the brand front and center in produce aisles across the country. What may be surprising is that the $7B company doesn’t actually farm. Instead, it focuses on breeding and testing new varietals across strawberries, blueberries, raspberries, and blackberries, then licenses those varieties to a network of independent farmers who plant and cultivate them at their own expense. According to the company, the farmers keep 75-80% of revenue, with Driscoll’s pocketing the remainder for its genetics, marketing, and distribution services. Structurally, Driscoll’s looks more like an R&D lab and patent licensor than a traditional, scaled farming operation like Tanimura, or a hybrid model like Taylor Farms. With that structure, Driscoll’s has transformed one of the most commoditized categories in the grocery store into one of the most valuable branded businesses in food, with Circana ranking Driscoll’s as the second-highest-earning brand in U.S. supermarkets (behind Coca-Cola).
The story highlights the power of a patent-driven flywheel. Driscoll’s can invest more in developing better berries than any individual farm could justify. Better fruit earns stronger shelf placement and margins, attracting additional growers into the network, which funds the next round of R&D. Each cycle expands both berry quality and distribution reach. Driscoll’s even became the first foreign berry company allowed to operate in China, where blueberry production surpassed the U.S. in 2025, helped by cheap agricultural technologies such as pesticide robots and anti-mold lighting, which would cost seven to eight times more in the U.S. Not every grower wants in, though. The article profiles a third‑generation farmer in Western New York who recently pulled 95 acres of apple trees to plant blueberries, but he intends to sell them himself. He’d rather trust his own read on what tastes good than farm to an outside company’s rules, and his target consumer isn’t a spec sheet, it’s his wife, who decides what to feed their kids and what to put in the cart.
How High can Red Bull Fly? (The Economist): A recent article from The Economist is a reminder of just how idiosyncratic Red Bull is. Dietrich Mateschitz built the brand in the 1980s by adapting Krating Daeng, a Thai energy tonic, for European tastes, and the Yoovidhya family, heirs to the original tonic’s creator, still own 51% of the business. Red Bull has never gone public. When Mateschitz died in 2022, some expected the company to fall as fast as it had risen, since he left no clear successor, and his son Mark inherited the remaining 49% but has no operating role. Instead, an exceptionally low-profile three-person board has run the company to great effect ever since.
The strategy that Mateschitz pioneered is still very much in place. Red Bull spends ~30% of sales on marketing, more than $3B last year, compared with ~10% at Coca-Cola and 7.5% at Monster. It continues to largely bypass traditional advertising in favor of owned content through Red Bull Media House, which is reportedly Austria’s largest media company after the state broadcaster. The approach has helped Red Bull stay well ahead in the category, with 28% global market share compared to Monster’s 17%. That said, competitive and regulatory pressures are beginning to build. Officials in Brussels are investigating whether Red Bull pressured retailers to sideline rivals, and Celsius, a sugar-free U.S. energy drink positioned as the healthier option, has already climbed to the number three spot in the U.S. with 7% market share
Rise of Delivery Apps Takes a Bite Out of Sales for Big Pizza (Haley Zimmerman, FT): This is one of those analyses that feels obvious with the benefit of hindsight, but the Financial Times lays out how delivery apps dismantled the two advantages that historically separated the big pizza chains from everyone else. The first was logistics: Domino’s, Pizza Hut, and Papa Johns spent decades building proprietary delivery networks that local pizzerias and most other cuisines could not match. The second was technology. All three launched mobile ordering apps by the early 2010s, long before most independents had a seamless digital storefront. Uber Eats and DoorDash, whose penetration surged throughout 2020, effectively handed both capabilities to every restaurant that wanted them. Any neighborhood restaurant now sits alongside Domino’s on the same app, and a burger or Thai order travels about as well as a pizza. That shift helps explain why delivery sales at U.S. quick-service pizza chains fell from ~$20B in 2021 to a low of $16.5B in 2023. Domino’s, which held onto its proprietary delivery network longer than its peers, has recently credited third-party platforms with helping stabilize volume, largely because the apps provide access to higher-income customers. Even with the benefit of hindsight, it is difficult to argue that the major pizza chains could have responded differently. Sometimes technology doesn’t create a new advantage so much as eliminate an existing one.
Patagonia’s Philosopher-King (Nick Paumgarten, The New Yorker): It’s impossible to do justice to Patagonia’s story in a few paragraphs, but we wanted to highlight a New Yorker profile from 2016 that does a remarkable job chronicling the history of the company and its enigmatic founder, Yvon Chouinard. One pivotal moment stands out. In 1991, Patagonia had expanded to the brink of bankruptcy and was forced to lay off 20% of its workforce. Chouinard concluded that the company had been “growing for the sake of growth” and responded by imposing a set of constraints that would come to define both the business and the brand for decades to come.
On his blog, Commoncog, Cedric Chin files Patagonia under the category of counter-positioning. Whether Chouinard intends it or not, the environmental and charitable commitments he embedded in the business created a form of differentiation that competitors struggle to replicate without sacrificing their own economics. Every dollar Patagonia directs toward conservation efforts is also a dollar that reinforces the brand’s positioning. While those choices undoubtedly carried short‑term costs, the company’s anti‑consumption stance does not appear to have constrained its growth. With FY2025 revenue of roughly $1.5B, Chouinard ultimately built the business he wanted, the way he wanted.
IM8 Takes $1B From General Catalyst’s ‘Customer Value Fund’ (Julie Bort, Techcrunch): Prenetics (NASDAQ: PRE) went public in 2022 via a biotech SPAC and has spent the past year remaking itself into a pure-play consumer company, divesting diagnostics and genomics assets along the way. The reason for the pivot is IM8, a wholly owned subsidiary co-founded with David Beckham that sells a subscription-based longevity supplement containing açai extract, Coenzyme Q10, and other compounds linked to healthy aging. IM8 launched in December 2024 and generated ~$92M in revenue in 2025, up 480% YoY, reaching a run rate of ~$120M by December. Management has since raised 2026 guidance to $210-$220M. IM8 reports an LTV/CAC ratio near 4.8x, and CEO Danny Yeung claims every dollar spent acquiring customers has generated $1.44 of gross profit.
which takes a very different approach from traditional growth equity. Rather than purchasing ownership stakes, CVF provides non-dilutive capital that is repaid through a capped revenue share, financing up to 70% of IM8’s customer acquisition costs. General Catalyst earns what it calls “reference income,” defined as revenue from those acquired customers multiplied by a predetermined gross-margin assumption, until the fund recovers its investment and reaches an agreed-upon return cap. After that point, all future revenue from those customer cohorts flows back to Prenetics.
In effect, General Catalyst is underwriting future marketing spend based on retention curves and margin assumptions rather than taking equity risk in the business. The firm used a similar structure with Grammarly in 2025. For those interested in the mechanics and rationale behind the model, General Catalyst outlines its thinking in this essay.

Quick Hits:

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #31]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-07chttps://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-07cFri, 17 Jul 2026 12:55:07 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

‘Expectations on Every Calorie:’ How PepsiCo’s CEO Is Navigating GLP-1s and Inflation (Laura Cooper, WSJ): PepsiCo CEO Ramon Laguarta told the WSJ that snack volume grew 1% in the first half of 2026, reversing several years of declines following price cuts on brands like Doritos and Lay’s earlier this year. Laguarta said restoring consumption is the company’s top priority and noted that affordability has never been more important for consumers.

Of note was the commentary on how PepsiCo is positioning around GLP-1 adoption. Rather than treating appetite suppression as a volume headwind to fight, Laguarta described the shift as an opportunity to repackage around evolving consumer preferences. GLP-1 households are increasingly seeking hydration, fiber, and protein, while still wanting “their little pleasures” in more controlled portions. He pointed to 100- and 120-calorie multipacks as an early success story. Smaller packs typically carry a higher price per ounce, which suggests portion control is also how PepsiCo recoups some of the margin it has given up on headline price.
How Avocados Stopped Being Seasonal (Rhishi Pethe, SFTW): Americans now eat almost 9 pounds of avocados per person annually, up from roughly 1 pound in the mid-1990s, and Rhishi Pethe argues that little of that growth was organic. A pest quarantine dating to 1914 kept Mexican avocados out of the U.S. for over 80 years until NAFTA opened the market. Once imports surged, growers and importers won congressional approval for a checkoff program in 2000 that levies a 2.5-cent tax on every pound to fund generic category marketing. The industry's marketing arm, Avocados from Mexico, then tied demand to the Super Bowl, an event that conveniently coincides with peak harvest in Michoacán. The strategy was clearly successful, with roughly 20% of annual avocado sales now concentrated around Super Bowl weekend. About 300M pounds were imported in the four weeks before this year's game.
Today, Mexico supplies ~80% of avocados consumed in the U.S. and enjoys tariff-free access under the United States-Mexico-Canada Agreement ("USMCA"), which replaced NAFTA in 2020. Exporters from Peru, Chile, and Colombia, by contrast, face a 10% surcharge. At the USMCA's mandatory six-year review on July 1, the U.S. declined to renew the agreement, so it now faces annual reviews and will expire in 2036 unless all three countries agree to extend it. Current trade rules remain unchanged, but the industry has already seen how quickly disruptions can affect prices. After assailants attacked USDA inspectors at a roadblock in Michoacán in 2024, imports were paused for one week and avocado carton prices rose 40%.
He Earns $33 an Hour as a Costco Cashier. Now He’s a Millionaire (Sarah Nassauer, WSJ): The WSJ profiled Tony Barzar, a Costco cashier of more than four decades who earns $33 an hour, has accumulated over $1M in his 401(k), and has repeatedly declined promotions to supervisor. According to CFO Gary Millerchip, Tony is not an outlier, and “many thousands” of Costco’s U.S. hourly employees now have retirement accounts exceeding $1M. The outcome reflects a compensation philosophy dating back to the Price Club era that views frontline pay as an investment rather than an expense. For Costco, that investment appears to pay off in retention and operational performance. Turnover after one year is ~7% and shrinkage runs at ~0.2% of sales, both a fraction of industry averages. The company recently raised its hourly pay ceiling, expanded bonuses, and added an additional week of vacation for employees with more than 30 years of service.
McKinsey estimates that replacing a frontline retail worker costs roughly $10K, and its 2023 analysis of more than 100 retailers found that companies in the top quartile of employee satisfaction were more than twice as likely to rank in the top quartile of customer satisfaction. Costco saw that correlation early. Experienced cashiers check out shoppers faster, train new hires, and carry institutional knowledge that is difficult to replace in a high-turnover workforce. At this point, Costco’s labor model is a real moat and an input cost most rivals continue to treat as a cost to minimize.
The Retailer Amazon Couldn’t Kill (Ryan Henderson, Fiscal.ai): On July 5, 2017, O’Reilly Automotive fell 20% in a single day, extending its decline to more than 40%, after Q2 2017 comparable sales of 1.5% missed guidance of 3-5% and appeared to confirm investor fears that Amazon’s expanding direct relationships with parts manufacturers would upend brick‑and‑mortar auto retail. In a recent blog post, Henderson revisits the episode and argues investors underestimated the importance of physical distribution in categories where availability and speed matter more than price. The typical buyer is a mechanic with a car in the shop who needs a part ASAP to complete a repair and move on to the next job. The competitive advantage was physical density. At the time, 32 distribution centers, 399 hub stores, and a company-owned delivery fleet provided professional customers with same-day access to 63K SKUs, with more than 95% of stores receiving multiple deliveries each day.
Repair shops order frequently and predictably, allowing O'Reilly to carry a large inventory base that is largely financed by suppliers. As a result, free cash flow has exceeded net income over the past decade, a rarity in retail, where growth often consumes cash. Management has consistently reinvested in the network first and directed excess cash toward buybacks, reducing shares outstanding by more than 60% since 2012. That means the 51% increase in average store revenue since 2017 has compounded on top of a steadily shrinking share count.
Bloomberg reported on July 2 that O'Reilly has submitted a $10B cash bid for Genuine Parts' NAPA automotive business, which Genuine Parts had been preparing to spin off as a standalone company in 2027. If completed, the deal would be O'Reilly's largest acquisition since its ~$1B purchase of CSK Auto in 2008. The proposed acquisition reflects a broader conviction that density remains the defining competitive advantage in automotive aftermarket distribution, which is a view that looks increasingly validated nearly a decade after investors first worried Amazon would disrupt the category.
Nike just can’t do it anymore (Amanda Chicago Lewis, The Economist): The Economist’s 1843 magazine profiles Nike’s attempt to recapture cultural relevance under Elliott Hill. Their diagnosis essentially boils down to this: Nike became too big to be cool, and its attempts to stay apolitical following the Kaepernick backlash left the brand culturally constrained. One executive described the company as spending the year avoiding any position on social issues, while another employee characterized the internal mindset as wanting to stand out without offending, driven by fear. As of last week, the stock was down >60% over the past five years, following roughly ten quarters of flat to declining revenue.
The cultural critique is directionally right, but it feels downstream of a more fundamental strategic misstep. As we covered in Issue #6, Nike’s return to Amazon after a six-year absence signaled a reversal of its push toward DTC. That strategy pulled product out of wholesale channels under the assumption that direct sales would drive better margins. In practice, MoffettNathanson found the opposite once advertising (~8% of revenue) and logistics (~12%) costs were fully accounted for. Unfortunately for Nike, there was a bigger cost that didn’t show up directly in the P&L. They gave up shelf space that had historically functioned as a powerful customer acquisition engine. Where consumers once encountered Nike passively and repeatedly in retail environments, they now had to actively seek it out online. That shift created an opening for newer, more culturally resonant brands to capture attention at the point of discovery. As a general rule, incumbents work to reinforce their distribution moat over time. Nike, by contrast, appears to have spent the past five years eroding their own.

Subscribe now

Quick Hits:

From Sportico: If you were wondering, Fox paid $485M for the rights to broadcast the World Cup. On hydration breaks alone, which have added ~260 seconds of ad time per match, Fox generated ~$200M in ad sales just during the group stage and is on track to bring in $450M by the end of the tournament.
From The New Consumer: Dan Frommer’s 2026 Mid-Year Consumer Trends report with Coefficient Capital centers on the rise of the "Optimizers," the roughly one-third of Americans who are simultaneously pursuing better health, longer lives, and greater enjoyment. The report tracks how that mindset is pulling peptides, protein, and AI into everyday consumption.
From The WSJ: A quick look at Nestle’s efforts to balance affordability with consumer demand for value. Smaller and larger size pack formats have been the answer so far, reinforcing a theme we revisit often in this newsletter: it’s increasingly difficult to compete in the middle.
From Bloomberg: A profile of the steady decline in places designed for in-person socializing across the U.S. In 2003, Americans spent ~285 hours socializing and communicating per year, but by 2025, that had fallen to ~212 hours, a 25% decline.
From The New Yorker: Fiber is increasingly being positioned as the new protein, with creators attracting large audiences by teaching followers how to “fibermaxx.”
In hindsight, Smucker’s $5B acquisition of Hostess in 2024 appears poorly timed. While overall U.S. snack sales are down only ~4% over the past four years, sweet snack sales have fallen roughly 17%. Twinkies and similar products have struggled amid GLP‑1 adoption and a broader consumer shift away from heavily processed foods.

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #30]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-78ehttps://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-78eWed, 01 Jul 2026 21:26:36 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

Protein-in-Everything Craze Has a Problem: Not Enough Whey Protein (Ameera McKee, WSJ): Whey protein concentrate has nearly tripled over the past year, reaching a record $13 per pound, with the more refined isolate up ~50%. As we have all observed, protein has shifted from a point of differentiation to a baseline label requirement over the last two years, driven by GLP-1 users and broader health-conscious consumers. Walk through any grocery store and you’ll see brands packing it into cereal, ice cream, popcorn, and even water. Not surprisingly, price pressure is most detrimental for brands built around the ingredient. Protein now accounts for more than half of total costs at Rogue Protein Snacks, and Legendary Foods expects to raise prices for the second time in two years.
Whey itself is a byproduct of cheese, which historically was so low‑value that producers once paid farmers to dispose of it. Converting that raw liquid into concentrate or isolate requires multimillion-dollar ultrafiltration systems and industrial drying capacity. U.S. dairy processors have committed $11B through 2028 to expand that capacity. FrieslandCampina, which acquired processor Wisconsin Whey Protein this year, expects unmet demand to persist for the next several years, partly because producing more whey inherently means producing more cheese than the market can absorb. Some processors have even begun running cheese production primarily to harvest whey, selling the resulting curds into an already soft market. The traditionally cheaper substitute, milk protein isolate, is now following the same trajectory, with prices rising rapidly as demand shifts downstream.
The EV Fleet Transition (Hannah Ritchie, By the Numbers): IEA 2025 data, summarized by Hannah Ritchie, shows that one in four new cars sold globally last year was electric, up from one in five in 2024. Adoption continues to accelerate in leading markets: China crossed 50%, Denmark is above 70%, and Norway is now effectively full electric. The U.S. remains the outlier, stalled near ~10% share at roughly 1.5M units, flat YoY. On this trajectory, 2017 appears likely to stand as the peak year for global internal combustion engine (ICE) sales. EVs may represent a growing share of new sales, but the global car parc turns over slowly. Even in China, where more than half of new sales are electric, only about 14% of vehicles on the road are EVs. Globally, EV penetration remains just ~4-5% of a 1.5B vehicle base. A car purchased today typically stays on the road for 12-15 years, meaning each year's sales replace only a small fraction of the installed fleet. Ritchie's model projects China's vehicle base reaching 50% electric around 2032 and 90% in the early 2040s, even under an aggressive adoption curve. The ICE aftermarket (parts, maintenance, and fuel) governed by the slow turnover of the existing fleet rather than headline EV sales growth. Despite accelerating adoption, there is still a long runway before markets like the U.S. approach meaningful EV penetration at the fleet level.
How Edikted Built a Teen Sensation on Crop Tops and Miniskirts (Chavie Lieber, WSJ): The Wall Street Journal profiled Edikted, founded in 2021, whose founder describes the company as “more like a data company than a fashion company.” Software tracks sales in real time, third-party factories in China produce initial batches of ~500 units, and only the bestsellers are reordered. All in all, Edikted is a high-velocity model: ~300 styles per month, design-to-launch cycles of two to three weeks, and an incredible $460M in gross sales in 2025, all built on just $9M of equity, all while profitable.
Those economics are driven by the underlying inventory model. Traditional fashion commits to large, seasonal orders months in advance, effectively betting on their seasonal forecasts; dodgy guesses become inventory markdowns and excess inventory, a persistent drag on margins. Test-and-repeat flips that risk profile. Edikted commits only to small initial runs, reacts to real demand, and doubles down on the winners, which reduces both the cash tied up in inventory and the cost of being wrong. This playbook isn’t new, but it continues to scale. Zara built the first and largest version decades ago through short lead times, frequent drops, and decentralized ordering policies. Shein industrialized it with on-demand micro-batching. Edikted is the latest brand to reach scale on that structure.
The Core Disney Strategy Behind Its Empire, According to the Hosts of Acquired (David Rosenthal & Ben Gilbert, WSJ): The Acquired hosts frame Disney not as a movie studio, but as a flywheel company. Its core profit engines sit in physical consumer categories. Animation creates characters with unusually durable economics - no talent leverage, no aging curves, no renegotiations. Wide theatrical distribution turns those characters into cultural touchstones, and then merchandise, licensing, and parks convert that affinity into high-margin revenue streams that can last for decades. In 2025, Disney generated more than twice as much operating income from Experiences as from Entertainment, on ~$94B in revenue. License Global ranks Disney as the world’s largest licensor, with an estimated $63B in licensed retail sales in 2024, more than one-fifth of the $308B market and nearly double the next largest brand owner. The rest of the industry has spent years copying Disney’s monetization strategy, treating IP as inventory to extend across as many product categories as possible. Lego is a very clear example. Its 1999 Star Wars line was the company’s first licensed property and remains a cornerstone 25 years later, following the blueprint for subsequent tie-ups with Harry Potter, Marvel, and Nintendo. Disney defined how IP gets monetized at scale. In doing so, it positioned itself to capture value even as others adopt the same strategy, collecting royalties whenever another brand reaches for the same playbook.
Has AI Already Killed How-To Nonfiction? Sales Trends, My Personal Data, and What It Might Mean for the Future (Tim Ferriss): Tim Ferriss recently shared domestic print sales data for his five-book catalog, which had long behaved like an annuity but has seen a sharp revenue decline since ChatGPT’s launch in November 2022. Ferriss is not alone here. Publishers Weekly reported self-help units down ~26% YoY in Q1, the steepest decline of any nonfiction subcategory, and Seth Godin, commenting on Ferriss’ post, noted he is seeing similar trends. The category appears particularly exposed because of the nature of the product itself. At a basic level, much of how-to nonfiction functions as a lookup table for achieving discrete outcomes. The underlying advice is often not proprietary enough to fall outside an LLM’s training set, and increasingly the advantage of modern AI is personalization plus the ability to tailor that guidance instantly to a user’s specific context. Ad-supported publishers have already felt the impact through lost traffic to Google’s AI overviews. Ferriss’ data suggests the substitution effect is moving beyond free, search-driven content and into premium content that consumers previously sought out and paid for.

Quick Hits:

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #29]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-4b5https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-4b5Wed, 17 Jun 2026 13:13:57 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

Subscribe now

The Roundup:

Mammoth Brands’ Ambition to become the next CPG giant (Amelia Lucas & Melissa Repko, CNBC): Mammoth Brands began in 2013 as Harry’s, founded by Andy Katz-Mayfield and Warby Parker co-founder Jeff Raider to address the frustration of $20 replacement razor blades. The brand launched online, reached Target shelves by 2016, and nearly sold to Schick owner Edgewell in 2020 for $1.4B before the FTC blocked the deal. Coming out of that failed transaction, the company accelerated its multi-brand strategy: launching Flamingo in 2018, acquiring Lume in 2021 (with sales more than doubling within two years, per the company), introducing Mando in 2022, rebranding to Mammoth in April 2025, and acquiring Coterie in October at a reported valuation above $1B. The group generated $835M in revenue and nearly $100M in adjusted EBITDA in 2024, compounding at more than 20% annually over five years. Bloomberg reports Mammoth is now weighing an IPO as soon as the second half of this year. Katz-Mayfield told CNBC the goal is a modern CPG company “like if Procter & Gamble and Unilever were getting built today.”
Why is this possible now? As we’ve mentioned before, incumbent moats in CPG have been steadily eroding over the past decade, driven by the combination of co-manufacturing and digital marketing. Today, there is effectively a co-man for almost any product formulation, while platforms like Meta and YouTube allows brands to reach consumers with unprecedented precision. Put simply, barriers to entry have come down. The data supports it. From Bain’s 2026 Insurgent Brands report:
This year’s cohort of 113 insurgent brands continues to outperform at an accelerating pace. While the overall market grew less than 2% in 2025, insurgents captured about 36% of market growth in tracked channels across all FMCG categories in 2025.
The impact spans across categories. Insurgents drove 25% of category growth in food and 13% in nonalcoholic beverages. In beauty and personal care, they captured nearly all the sector’s growth.
Importantly, this outperformance was almost entirely driven by volume expansion. Insurgents grew volumes by approximately 55% year over year in a market where overall volumes were largely flat, demonstrating that their growth reflects real consumer demand, not just price inflation.
Amazon’s Identity Graph (Eric Seufert & Adam Epstein, Mobile Dev Memo): Amazon recently announced that its authenticated identity graph now deterministically reaches 90% of U.S. households with verified, logged-in data. “Deterministic” means real identities tied to actual accounts rather than statistical models. More importantly, those identities are linked to purchase behavior: what households buy, when they buy it, and at what price, across a platform that accounts for ~50% of U.S. e-commerce. Competitors can approximate pieces of this. Google has logged-in identity through YouTube but limited transaction data; The Trade Desk has neither proprietary identity nor proprietary inventory. No one else combines verified identity with transaction-level commerce data at Amazon’s scale.
Amazon's advertising services segment generated just over $70B in LTM revenue as of Q1’26 and continues to grow ~20% annually. Sponsored listings, product detail page placements, and promotions already cover every inch of Amazon.com, and sellers are approaching the limit of what they can pay in CPCs without impairing margins. The next leg of growth has to come from the demand-side platform (DSP), particularly in connected TV, where major buyers allocate budgets across Disney, Netflix, Paramount, and Prime Video. Presumably, Amazon's pitch to those buyers is that effective reach and frequency management requires a single, deduplicated view of the household, and its identity graph is the highest-signal version of that view available. It then layers in a structural advantage with Prime Video, whose inventory is only available through its DSP, while reportedly incentivizing sales teams to push non-Amazon inventory as well. That posture underscores the ambition to not just to capture spend tied to Prime Video, but to intermediate an entire brand’s TV budget.
Predicting AI Job Exposure (Benedict Evans): In a well-argued article, Benedict Evans makes the case that the growing cottage industry of scoring jobs by “AI exposure” is largely misguided. Looking back 50 years, accounting should have been one of the most automatable professions in the U.S.: a century of calculating machines, mainframes, spreadsheets, and ERP systems were built around it. By the late 1970s, there are anecdotes of CPAs using VisiCalc to compress month-long projects into days. Yet number of accountants continued to rise. Lower costs made analysis cheaper, which lead to more analysis (a classic case of Jevons paradox), while regulation created new categories of work and the role itself continuously evolved under the same title. The Census job category for “billing, posting and calculating machine operator” emerged and disappeared within a decade, which is a reminder that when technology truly eliminates work, entire job categories can vanish, rather than gradually declining.
An even harder problem is that jobs can remain intact while the economic structure supporting them collapses. The internet didn’t change what made a good journalist, but it dismantled the local ad monopolies, particularly classifieds, that funded journalism, forcing a structural reset of the industry. Similarly, no one modeling “smartphone exposure” in 2005 would have flagged taxi drivers, whose $1M NYC taxi medallions were about to become effectively worthless. It’s difficult to construct a predictive framework that would have correctly identified all three outcomes: the persistence of accounting, the disappearance of clerical roles, and the collapse of taxi medallion economics. Against that backdrop, Evans argues that current efforts to score AI exposure, often based on O*NET task descriptions, should be treated with a healthy degree of skepticism.

Quick Hits:

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #28]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-ecfhttps://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-ecfThu, 04 Jun 2026 17:38:17 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

Can AI responses be influenced? The SEO industry is trying (Mia Sato, The Verge): At its I/O conference last month, Google said AI Overviews now reaches more than 2.5B monthly users and AI Mode has surpassed 1B just a year after launch, with the two now merging into a single, seamless search experience that flows from an AI Overview into a follow-up in AI Mode across desktop and mobile. The Verge maps what this shift is doing to the ecosystem built on Google’s legacy ranking system. A new wave of firms is now promising to get brands “cited” by AI through generative engine optimization, but the tactics remain largely experimental. One firm, Fabric, runs thousands of synthetic prompts like “jeans similar to Everlane or Uniqlo” to track how often a model names a client vs. competitors. Its CEO says AI discovery has quickly become a top-one-or-two priority for the retailers he advises, after they watched homepage traffic fall while product pages held.
What no one can reliably do yet is measure it. Even the vendors concede the gap: the cofounder of Growtika, an agency that markets itself on warnings about collapsing search traffic, told The Verge there is “still not a right way to measure” AI visibility. SparkToro’s Rand Fishkin estimates that money and focus on AI search are running 10 to 100 times ahead of actual usage, even as traditional search continues to dominate. The more durable change may instead be where credit lands. The signals most likely to earn an AI citation are third-party mentions on Reddit, in news coverage, and across other high-trust sources, the exact channels performance marketing deprioritized for years. Gartner now expects brands to double earned-media and PR budgets by 2027.
Reverse Engineering The Black Box (Eric Seufert, Mobile Dev Memo): The major ad platforms have increasingly converged on end-to-end automation. Advertisers hand over a budget and an objective, and the system handles targeting, bidding and, increasingly, the creative itself. Meta’s Advantage+ and Google’s Performance Max are the templates, with generative tools pushing the marginal cost of producing an ad toward zero. Alphabet reported 70M creative assets flowing through its text-customization tools in Q4 2025 alone. For advertisers, the result is effectively a black box with just two meaningful levers left: creative and budget. On Stratechery, Eric Seufert paraphrases the platforms’ pitch, echoing Mark Zuckerberg: just provide a credit card and the system will take care of everything else.
Seufert’s counter, outlined in his recent Harvard master’s thesis, is not to fight the black box, but to replicate it. His model, DeCANT, treats platforms like Meta as “teachers” and trains a smaller “student” model on an advertiser’s own performance data to predict a creative’s return before spending against it. The rationale is straightforward: advertisers routinely allocate ~10% of budget to creative testing with little return accountability, spend the platforms collect regardless of outcome. DeCANT was trained across every creative variant for roughly $1,000 of cloud compute, making it cheap enough to retrain daily. Platforms are unlikely to welcome advertisers reverse-engineering their economics this way, but as Seufert notes, there is little they can do to prevent it.
The Hunt for Milk Chocolate (Cedric Chin, Commoncog): In consumer products, many enduring franchises are built on a single product innovation, one that introduces a new taste or form factor to consumers. Hershey’s was exactly that: a new kind of milk chocolate. Milton Hershey sold his caramel company for $1M in 1900 and then spent three years trying, and failing, to reinvent the recipe from scratch, largely ignoring the decades of technique European producers had already developed. When he finally succeeded, with a process that slow-boiled heavily sweetened milk in a vacuum, it produced an unexpected side effect chemists wouldn’t fully understand for decades: the breakdown of milk fat into free fatty acids, giving the chocolate a faintly sour note. One of those compounds, butyric acid, is also found in aged cheese. European chocolatiers were appalled (a Swiss cocoa official reportedly said Hershey had “completely ruined the American palate with his sour, gritty chocolate”), but for American consumers, the flavor was entirely new, and it stuck. Hershey priced every bar at a nickel and pushed distribution beyond traditional candy shops into grocers, newsstands, and bus stops nationwide, while rivals like Walter Baker remained regionally constrained. By 1907, revenue approached $2M, entirely self-funded. Distribution enabled the scale, but the invention was the foundation: a taste Americans had never encountered and, having grown up on it, never abandoned. That divergence still holds today, with European consumers, raised on Cadbury or Lindt, often finding Hershey’s profile off‑putting. More than a century later, product innovation remains at the center of industry growth, with next-generation brands such as Celsius, Poppi, and Olipop building meaningful businesses by introducing taste profiles and form factors incumbents weren’t providing.
Soft Serve & Soft Power (Rachel Brown, Listomania): A wave of Chinese chains has opened across New York over the past year, from Luckin Coffee, now at roughly ten locations, to soft-serve chain Mixue, with most new entrants clustered around NYU and the East Village. The city saw a similar influx of Japanese concepts in 2017, some of which stuck (ramen chains like Ichiran and Tonchin) while others have exited (Zauo, Ikinari Steak). Mixue will be an interesting test case. It is now the largest restaurant chain in the world by store count, with more than 53K locations, ahead of both McDonald's and Starbucks, yet it generated just $6.5B in system sales in 2024, compared to McDonald's $131B on a much smaller footprint. The model is clearly built for scale and cost efficiency, optimized for China’s hyper-competitive domestic market. The question is whether that model translates to New York’s operating environment. These brands are likely entering the U.S. assuming relatively softer competition than at home, where price wars have been punishing. They may find the opposite. Incumbents like Starbucks and Dutch Bros remain formidable, and a new wave of U.S.-grown chains (Better Buzz and Blank Street) have already reset expectations around price, format, and convenience. The first real signal of success will be whether any of these chains expand meaningfully beyond a handful of Manhattan zip codes.
How Swatch and Audemars Piguet Defied Collaboration Fatigue (Cathaleen Chen, Business of Fashion): Brand collaborations have become so ubiquitous that most now feel incremental, selling out to existing fans without reaching meaningfully new audiences. This launch looked different from the start. Five days before release, fans were already lining up and sleeping in folding chairs outside Swatch stores for a chance at the Royal Pop, a $400 plastic pocket watch co-designed with Audemars Piguet (AP). On launch day, crowds were large enough to shut stores and require police presence from New York to Dubai. The backdrop was Swatch’s 2022 MoonSwatch, its $260 bioceramic take on the Omega Speedmaster, which sold more than 1M units in its first year and lifted full-price Speedmaster sales by 50%. Accessible collaboration drove halo demand for the luxury partner, and set expectations for a repeat. When Swatch teased a partnership with one of watchmaking’s “Holy Trinity,” traders assumed a repeat and bid the stock up almost 18% in a week. What buyers got instead was not a wristwatch. For the first time since the Royal Oak’s 1972 debut, AP licensed its design, but only for a pocket watch. RBC raised its price target on Swatch to 220 francs from 160. The stock dipped initially on the surprise format, but quickly recovered and now sits near a three-month high, up roughly 17.5%. Swatch got volume; AP got the lowest-ever entry point into its universe with a $400 product that introduces the brand to consumers priced out of a $20,000 Royal Oak, while avoiding the dilution of putting plastic versions on wrists. Early resale data suggests many Royal Pop buyers are first-time participants, which is precisely the audience AP was paying to reach.

Quick Hits:

Subscribe now

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #27]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-185https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-185Sat, 16 May 2026 15:06:26 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

Meta Q1 2026 Earnings (Eric Seufert, Mobile Dev Memo): Meta’s Q1 2026 ad revenue grew 32.9% to $55B, driven by a 19% increase in impressions (the fastest pace since Q1 2024) and accelerating price growth, with an average price per ad up 12% vs. 6% in Q4. In a typical auction market, that degree of new supply would pressure clearing prices. Instead, recent quarters have shown the opposite. Daily Active People rose just 3.8%, meaning the incremental inventory isn’t coming from user growth, but from Meta’s AI-driven ranking and retrieval systems. By surfacing more relevant content, Meta is lifting engagement across Reels and Feed and increasing the number of ads shown per session. Pricing rising alongside volume suggests each impression is becoming more valuable to advertisers, even as the bidder pool continues to deepen. 8M advertisers now use at least one of Meta’s generative creative tools, and a new MCP endpoint allows agents like Claude Code to plug directly into ad accounts.
This long-tail dynamic is one of Meta’s core moats. During the 2020 Stop Hate for Profit boycott, more than 1,000 advertisers paused spending, including Coca-Cola, Unilever, Verizon, and Ford, yet ad spend revenue continued to grow. The top 100 advertisers accounted for just 6% of total revenue, prompting Zuckerberg to say at the time that Meta wouldn’t change its policies over “a threat to a small percent of our revenue.” AI extends this moat by expanding the bidder pool while increasing the value of each bid. As Seufert puts it: “AI is not the metaverse.” The returns are measurable now, which likely explains Zuckerberg’s visible irritation when the first analyst pressed him on AI ROI. Still, the market remains unconvinced. Meta raised full-year capex guidance to $125-145B (from $115-135B), and the stock traded down 6% on the print, reflecting investor concern over whether gains in matching efficiency will continue to compound faster than the infrastructure required to deliver them. Against a backdrop of rapid change in digital advertising, it’s a useful reminder for founders and management to reflect that long-term durability comes from a truly differentiated product and a brand that supports it.
Retailers Flock to TikTok Shop to Find New Shoppers (Jennifer Williams, WSJ): TikTok Shop generated $4.9B in U.S. sales in Q1, nearly double YoY, with Ralph Lauren, Olaplex, and Ulta Beauty all opening storefronts in the past six months. While the platform still represents just ~1% of U.S. retail and ~3% of e-commerce, Circana expects that share to quadruple within a year and approach 10% by 2028. The primary draw for brands is customer acquisition. ThredUp sold more than 100K of its $2.99 Clean Out Kits in two weeks; 98% of those buyers were new to the brand, effectively giving ThredUp zero net customer acquisition cost (CAC) on customers that typically cost more than $3 to acquire. Dan Frommer’s 2026 Consumer Trends report projects $15B of U.S. GMV in 2025, in only the platform’s second full year. Growth is also broadening demographically: the 45+ cohort is now TikTok Shop’s fastest-growing segment. At the same time, sales are top-heavy, too. Roughly 8,450 creators (the top 1% of 845K tracked) drove 53% of $7.7B in trailing-twelve-month video sales. The week the WSJ piece ran, QVC’s parent filed for Chapter 11 to shed more than $5B in debt, after 2024 sales fell roughly 30% from a 2020 peak above $14B, as consumers have continued shifting shopping behavior from traditional formats to social platforms (Bloomberg).
Amazon’s Durability (Ben Thompson, Stratechery): Ben Thompson frames Amazon’s durability around a single playbook he calls “primitives.” Amazon builds infrastructure where it is the first and best customer, which justifies the capex, then sells that same infrastructure to third parties, deepening the moat. AWS was the original template. This week’s launch of Amazon Supply Chain Services, with P&G and 3M as anchor customers, extends the model to freight, trucking, and last-mile delivery, nearly a decade after Thompson first predicted Amazon would move into logistics as a service. The next primitive in line is Amazon’s satellite constellation, Leo (formerly Project Kuiper). On the most recent earnings call, Andy Jassy described Leo as “reminiscent of AWS,” with 20+ launches planned this year and 30+ in 2027. The endgame isn’t simply selling satellite connectivity to enterprises, the way Apple currently pays Globalstar, a company Amazon agreed to acquire for $11.5B in April, with Apple remaining a customer of the combined Leo network. Leo also gives Amazon a proprietary connectivity layer for its eventual drone delivery fleet, removing dependence on third-party systems like Musk’s. Each new business builds on the last.
This is Nick Sleep’s “Scale Economies Shared” thesis playing out at infrastructure scale. In the Nomad letters, Sleep argued that the most durable businesses are those that pass scale benefits back to customers rather than retaining them as margin. Costco is the canonical consumer example, with renewal rates above 90% and efficiency gains routinely handed back through lower prices. Amazon was Sleep’s other concentrated position for the same reason. Every new ASCS customer makes Amazon’s freight network denser and cheaper to operate, lowering costs for Amazon.com retail and customers like 3M simultaneously. Each additional Bedrock customer drives down the cost per token on Trainium. The flywheel operates over decades, which is why public markets routinely underprice it. The discipline, and the risk, is resisting the temptation to extract once the network is built, the point at which most platform businesses eventually falter.
China Shock 2.0 (Ryan McMorrow, FT): Twenty years ago, the first "China shock" hollowed out Western producers of textiles, toys, and low-end electronics. The FT's new series argues a second shock is now hitting the high end: EVs, solar, batteries, and advanced industrials. The opening case is Mega-Senway, a Shanghai-based sensor maker whose EV-charger sensors fell in price from ¥200 six years ago (where German and Swiss incumbents had been pricing) to ¥10 today. The Europeans exited the category entirely. In China, the mechanism is called neijuan, or involution: relentless domestic competition reinforced by deep engineering talent, massive industrial scale, and state subsidies that the OECD estimates run 3-9x those of advanced-economy peers, largely via below-market state bank loans. Layer on a renminbi the IMF estimates is undervalued by ~15% and you get a system that produces dominant exporters and zero-margin domestic businesses at the same time: BYD's average selling price fell from ¥143,100 to ¥119,223 in three years; China's six largest solar manufacturers lost a combined ¥43B in 2025 while still collecting subsidies; and exports grew nearly 15% YoY in Q1 2026 on top of a roughly $1T trade surplus.
Neijuan maps closely to the “Red Queen” effect Buffett has described in commodity industries, where everyone runs faster and faster just to stay in place. Ask Swiss sensor maker LEM, whose net margins collapsed from 19.4% to 2.7% as its shares fell 84% over four years; its response was to hire 30 more R&D staff in Shanghai. Even Volkswagen moved all new-model engineering to China. Shein and Temu are the consumer-facing expression of the same system. At this point, it has rarely been riskier to sell a product without clear differentiation or a brand strong enough to support it.
The Google Fitbit Air is an AI-infused take on Whoop wearables (Cherlynn Low, Engadget): Google unveiled the Fitbit Air, a $100 screenless tracker shipping May 26 that borrows directly from the Whoop playbook: no display, passive 24/7 data collection, swappable bands, and a Gemini-powered Google Health Coach that does the interpretive work across sleep, HRV, AFib detection, and automatic activity tracking. They look nearly identical and each unit ships with three months of Google Health Premium. The pricing math is where things get interesting. Whoop’s three tiers (One at $199/year, Peak at $239/year, Life at $359/year) bundle hardware into a subscription. Oura Gen 4 runs ~$349 plus a $5.99/month membership, while Apple Watch SE starts at $249. Google now offers a credible alternative at roughly one-third the annual cost of Whoop Peak, with the same foundation model that powers Gemini sitting under the coaching and insights layer. If differentiation in wearables is shifting from sensor accuracy to software interpretation, Google has a structural advantage: it owns the foundation model, controls mobile OS distribution, and has introduced a hardware price point that makes subscription-only pricing models harder to defend. The open question is whether Whoop's brand equity with serious athletes and Oura's leadership in sleep and women's health can hold against a credible $100 substitute, or whether the category ultimately commoditizes around whichever player owns the AI layer.

Quick Hits:

From Nikkei Asia: Toto, the Japanese toilet maker, now earns more from advanced ceramics used in semiconductor manufacturing equipment than from toilets and tubs. That division is projected to generate 55% of operating profit this fiscal year, with profits up 32% to ¥27B ($169M). Investors have taken notice as the stock is up 79% YTD.
From Food Dive: A looming shortage of whey protein concentrate, the core ingredient in many protein powders, is pushing prices sharply higher. Whey protein concentrate with 80% protein content is now topping $11 on the spot market, while protein isolate (90%+ protein by weight) remains “firmly in the $12s,” according to the USDA (~3-4x normal pricing).
From the BBC: The British Beer and Pub Association reports that ~160 pubs closed in Q1 2026, nearly two per day, compared with ~335 closures for all of FY 2025. Wales was the only region to see an increase in pubs, as the industry continues to grapple with a decades-long structural decline.
From the WSJ: Premium-format moviegoing appears to have hit a new ceiling. Regal sold out opening-night 70mm IMAX seats for Dune: Part 3 at $50 a pop within minutes, even as average ticket prices sit at $18 nationally ($30 in major cities). Exhibitors have stopped pretending movies are a cheap night out and are increasingly running a Broadway-style price-discrimination model on the top 1% of seats to offset still-soft post-pandemic attendance.
From the WSJ: NeeDoh, a $10 squishy gel cube from 50-year-old Massachusetts toymaker Schylling, has gone full Labubu, selling through a year's worth of inventory in nine weeks and pushing Amazon resale prices to $28, a 155% spike. Knock-offs and stockouts have hit Target, Walmart, and Five Below, while a poorly-timed Lunar New Year shutdown at Schylling’s Chinese factories means the company (owned by BDC Gladstone Investment Corp / GAIN) won't fully catch up to demand until summer.

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #26]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-5e4https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-5e4Tue, 05 May 2026 12:29:07 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

Subscribe now

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

Subscribe now

The Roundup:

Mondelez overhauls its $3.5 billion digital commerce strategy in era of AI search (Kimeko McCoy, Digiday): Mondelez (Oreo, Ritz, Sour Patch Kids) has rebuilt its $3.5B digital commerce operation around AI search after retail partners projected 30% of site traffic would come from agents by 2028. Internal data showed that Oreo, one of the world’s most recognized cookie brands, appeared in just 10% of AI chatbot cookie recommendations. Part of the issue was self-inflicted: Mondelez had blocked AI bot crawlers to protect its IP. After unblocking them and rebuilding its backend with clean sitemaps, machine-readable content, and structured product data, Oreo’s citation rate jumped to ~70% on cookie queries and ~30% on broader snack queries.
VP of global digital commerce Andrew Lederman told Digiday that Mondelez now treats AI search like a performance channel, actively tracking visibility, citation, and sentiment. He expects 20-30% of Mondelez product purchases to flow through agents within the next year. This case highlights how even a brand with decades of equity and dominant shelf presence can be effectively invisible in agent-mediated discovery, and the fix looks more like technical SEO than traditional brand building. What remains unclear is how well “agentic” traffic will convert to actual sales, and whether it delivers meaningful incrementality for a brand like Oreo that already enjoys near-universal awareness in the U.S.
Slate Auto Raises $650M (Sean O’Kane, Techcrunch): Slate Auto closed a $650M Series C earlier this month led by TWG Global, bringing the Bezos-backed startup’s total funding to roughly $1.4B ahead of production beginning at year-end. The truck is expected to list in the mid-$20Ks, can convert to an SUV via a $5K kit, and has already amassed more than 160K refundable reservations. The leadership team is notably Amazon-heavy: co-founder Jeff Wilke previously ran Amazon’s consumer business, and new CEO Peter Faricy ran Amazon Marketplace.
In Maintenance: Of Everything, Stewart Brand argues that “the world is still waiting...for an all-electric equivalent of the Model T, Volkswagen Bug, or Lada, something dirt cheap, brilliantly simple, and durable, that invites embellishment and repair by the owner.” The Model T, Beetle, and Lada sold roughly 58M units over the 20th century and shared three defining traits: low cost, a basic design maintained for decades, and owner repairability. Tesla cracked open the EV market, but in the opposite direction. Brand notes that body repairs on a Tesla can cost tens of thousands of dollars and take weeks, and that the megacasting approach Tesla introduced in 2021 made repairs harder, not easier. Slate’s leadership has reportedly cited the Model T and Beetle as its north star, and the design choices are consistent with that ambition: manual crank windows, no infotainment screen, a polypropylene body designed for vinyl wraps instead of paint, and an open-source service philosophy under which Slate says owners can perform DIY warranty repairs, an industry first, according to the company. Whether Slate can scale where Rivian and Lucid have struggled remains an open question, but in terms of intent, it is the closest attempt yet at the archetype Brand describes.
Peloton’s Latest Leader Thinks He Can Coach It Back to Health (Jordyn Holman, NYT): Peter Stern, who actively pursued what he called his “dream job,” is now one year into running a company whose market cap has fallen from ~$50B at its 2021 peak to ~$2B today. Quarterly revenue slid from a high of $1.3B in Mar’21 to $657M in Dec’25, and management lowered FY26 guidance earlier this year. Profitability, however, has moved in the opposite direction: EBITDA improved from -$1.3B in FY22 to +$141M in FY25, Peloton’s first profitable year, following repeated layoffs (the latest cut ~11% of staff) and aggressive cost reductions. The stock sold off sharply in January but has since rebounded, as investors now weigh whether profitable retention can translate into profitable growth.
Stern’s bet is that subscriber stickiness can support Peloton’s evolution from a bike company into a broader wellness platform, and the churn data supports that premise. Peloton’s average monthly Connected Fitness churn of ~1.5% sits is comparable to Spotify Premium (~1%), modestly better than Netflix (~1.8-2.5%), and roughly half the ~3% monthly attrition typical of U.S. gyms. The proposed growth wedge is strength training: ~2M members already engage in strength workouts on the app, GLP-1 usage is driving demand for resistance training to preserve muscle mass, and Peloton’s data shows that members who engage with two or more disciplines churn at ~60% the rate of single-discipline users. Stern has aligned product strategy around this thesis with the Cross Training series, the Tread+ Pro for commercial gyms, and Peloton IQ for AI-driven personalized coaching. What remains unsolved is that subscriber growth has historically followed hardware sales, and there is no obvious catalyst to restart that flywheel. Peloton has shown it can survive a ~50% revenue decline with subs down only ~10% from peak. The open question is whether it can grow member base again from here.
Toys Diverging (Joost van Dreunen, SuperJoost): Joost van Dreunen frames the latest toy industry earnings as a clear split: companies that have built a secondary, high-margin revenue engine alongside physical product are winning, and those still primarily moving boxes through retail are not. Hasbro is up 55%, Games Workshop 44%, and Asmodee 26% since Asmodee’s February 2025 IPO. Mattel is down 32%, and Spin Master 38%. The margin profiles map directly onto that gap. Hasbro’s Wizards division posted $2.2B in 2025 revenue (+45% YoY) at 46% margins, with Magic: The Gathering alone up 59%. Games Workshop earned ~42 cents on the dollar across $1B in revenue, while its licensing line nearly doubled to $66M after Space Marine 2 sold 7M copies. Asmodee, despite record sales, retains just 7.6 cents per dollar because much of its business still distributes physical product through retail. Mattel’s Uno has 300M lifetime downloads but generates only ~$3M/month, roughly 1/30th of Monopoly Go!, and is down 16% over the past eight months.
The reason this model works for Magic and Warhammer is what Bebop Value calls “cohort stacking”: new customers join every year, and old ones rarely leave. The mechanics are durable and cumulative: persistent value of accumulated product (your 1995 Space Marine still plays today), low re-entry friction, identity formation, local network effects, multi-generational handoff, and multi-layer engagement across building, painting, playing, reading, and watching. Uno has reach, but not depth. It doesn’t form identity, doesn’t accumulate over time, and doesn’t pull a 35-year-old back after a decade away. That distinction explains why digital extensions compound for some IP and just subsidizes others. Hasbro and Games Workshop aren’t winning because they built apps. They’re winning because they own IP that people built lives around, with licensing, video games, and digital monetization flowing downstream from that density. Mattel’s $269M in incremental digital spend is buying distribution into IP that may not have the underlying cohort behavior required to support it.

Quick Hits:

From Sherwood: Starbucks raised its FY26 guidance after posting a second consecutive quarter of positive same-store sales growth, following six straight quarters of declines.
From FastCompany: Walmart is rebranding its 10,000-SKU Great Value private label, which is now present in 9 out of 10 American households and outsells Coca-Cola by Nielsen's count, after focus groups admitted they weren't proud to leave it out when guests came over. The refresh follows Target's Good & Gather and Aldi's recent updates, and lands just as $100K+ households became Walmart's fastest-growing customer cohort last quarter.
From CNN: Kraft Heinz is leaning into protein and away from sugar, starting with a defensive move to shore up Kraft Mac & Cheese with “Power Mac.” An effort to stem share losses to better-for-you brands like Goodles.
From NOSH: Yough Pizza raised funding from venture studio Founders Row and secured nationwide distribution at Target, as protein continues to push into adjacent categories.
From FT: Uber plans to invest more than $10B to buy thousands of self-driving cars and partner with robotaxi makers (it is now the second-largest shareholder in Lucid). This is a big shift from its asset-light model of connecting drivers with riders. Investors are skeptical so far, judging by the stock’s reaction.
From Allrecipes: Grillo’s Pickles has never been afraid to lean into unconventional collaborations, and its latest launches are no exception…

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #25]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-cb4https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-cb4Thu, 16 Apr 2026 13:19:15 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

Subscribe now

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

The Medvi Files (Brendan Keeler, Health API Guy):
On April 2, The New York Times profiled Medvi, a GLP-1 telehealth startup reportedly on pace for $1.8B in 2026 revenue with just two full-time employees, as evidence that AI could enable what Sam Altman has predicted: a “one-person billion-dollar company.” Altman told the Times he had won a bet with his CEO friends and wanted to meet the founder. The story went viral, and then began to unravel almost immediately. Medvi has no outside investors, no audit, and no official valuation; the $1.8B figure is simply a revenue run-rate extrapolation. The Times did not mention that the FDA had issued Medvi a warning letter six weeks earlier for misbranding compounded semaglutide and tirzepatide, that a class action suit was filed in California alleging Medvi’s affiliate marketers sent deceptive spam emails from spoofed domains (one exhibit listed the return address as “123 Wellness Blvd, Healthtown, USA”), or that Medvi’s clinical infrastructure partner, OpenLoop Health, is facing a RICO complaint alleging the compounded oral tirzepatide sold through its network has no demonstrated efficacy. The AI framing was likely the wrong lens from the start. Medvi built an affiliate-marketing-powered checkout page on top of someone else’s medical infrastructure, and scaled it with aggressive, and allegedly illegal, customer acquisition tactics.
What sits underneath Medvi is more interesting, and more consequential, than Medvi itself. OpenLoop Health, run by the Lensing family out of Iowa, controls the technology platform, the medical group whose clinicians write prescriptions, and, through a holding-company structure, the compounding pharmacies that fulfill them, including Triad Rx and Link Pharmacy. Novo Nordisk is already suing Link. OpenLoop offers prospective storefronts a turnkey operation: marketing playbooks, suggested landing page layouts, and a prescribing pipeline ready to plug in. Medvi was simply the storefront that scaled fastest. Court filings from a breach-of-contract lawsuit brought on by one of Medvi’s own affiliate partners show the economics directly. One affiliate alone drove “tens of millions of dollars in revenues to Medvi per month” before the relationship collapsed after a competitor acquired the affiliate’s assets. In a sworn declaration, Gallagher admitted Medvi “has the funds to pay the due balance but chooses not to,” prompting a judge to grant an attachment order against him personally. This is the real architecture of the compounded GLP-1 boom: a vertically integrated backend, white-label storefronts competing on ad spend, and affiliates paid per conversion. As regulatory pressure mounts from multiple directions with Novo litigating against the pharmacy layer, the FDA targeting storefronts, and 35 state attorneys general flagging AI-generated weight-loss ads to Meta. The fallout from any serious enforcement action will not stop at Medvi. Every company, public or private, that has built a growth story on access to compounded GLP-1s is exposed.
FERRARI PART 1:
The Prancing Horse and the Business of Desire (David Rosenthal & Ben Gilbert, Acquired):
Acquired’s latest episode traces Ferrari from Enzo Ferrari’s founding of Scuderia Ferrari as an Alfa Romeo racing team in the 1920s to its current position as one of the most valuable luxury companies on earth. The numbers tell a clear story of scarcity as strategy. Over its entire 79-year history, Ferrari has sold just 330K cars, at an average price of ~$500K. Of the ~14K vehicles it produces each year, 80% are reserved for existing owners, leaving fewer than 3K slots for new buyers annually. That’s the volume Toyota moves every 10 hours. And yet Ferrari trades at a market cap of ~$63B - worth more than Ford, Volkswagen, or Mercedes-Benz, on trailing revenue of ~$8.4B and an EBITDA margin of ~39%, the highest in the auto industry.
The most instructive chapter of the episode is Luca di Montezemolo’s turnaround of the company following Enzo’s death in 1988. Under Fiat’s early management, Ferrari committed the cardinal sin of luxury: expanding supply to chase growth. Production climbed from 2,200 cars in 1982 to 4,500 by 1991, at which point cars sat unsold on dealer lots, and a Honda NSX was outperforming the Ferrari 348 on the test track. Montezemolo reversed course. He cut production nearly in half, reinvested heavily in F1 (most notably recruiting Michael Schumacher, who went on to win five consecutive championships from 2000-2004), and applied a luxury playbook borrowed from Europe’s great fashion houses: long wait lists, highly ritualized delivery ceremonies, and the discipline to always produce one fewer car than the market demands. Ferrari would not surpass its 1991 production peak again until 2006, 15 years later. By then, the waitlist, the mythology, and the margins were strong enough to support a $10B IPO. Enzo sold 50% of the company to Fiat in 1969 for a valuation of $6.8M. Today, Ferrari is worth nearly 10,000x that.
FERRARI PART 2:
A Study in Rosso Corsa: How Ferrari became “the Ferrari of Fundamental Value Investments” (Nadav Mandam, PIB Academy):
Nadav Manham of PIB Academy breaks down how Ferrari compounded shareholder value at ~23% annualized over the decade ending November 2025. The thesis is simple. Revenue per vehicle increased at a 3.5% annualized, and unit shipments grew at 5.9% (from ~7,700 to ~14,000). Neither figure would excite a growth equity investor on first glance. But operating leverage did the work. Cost of sales grew at 8.7% against 9.6% revenue growth, while SG&A expanded at 6.6%, and R&D at just 5.0%, producing 16.9% annualized EBIT growth. Add declining tax rates and consistent share buybacks, and Ferrari turned that 5.9% volume growth into 22.9% annualized dividend-per-share growth, re-rating from 18x earnings post-spinoff to routinely north of 40x. The balance sheet made it possible. Negative net working capital, with customer prepayments effectively financing operations, meant retained earnings could fund capex without diluting shareholders. At the same time, a 25-40% dividend payout policy kept management disciplined. Nadav aptly describes the result as a "slow-motion LBO with training wheels." A brilliant strategy, brilliantly executed by Sergio Marchionne.
A Token Is Not a Fixed Unit of Cost (Anjali Shrivastava, Essay 1, Essay 2; MBI Deep Dives, Summary 1, Summary 2): Anthropic’s run-rate revenue hit $30B in April, roughly tripling since December. Meta employees consumed an estimated 60T tokens in a single 30-day period (every book ever published is estimated at ~20T tokens). But as Anjali Shrivastava argues in a pair of essays well worth your reading, revenue growth and profitability are fundamentally different problems in AI. Because of how large language models work, generating the 10,000th token in a conversation requires meaningfully more compute than generating the first. A token is not a fixed unit of cost. In traditional SaaS, the law of large numbers smooths this out: low-usage customers subsidize power users, and gross margins converge in predictable ways. AI breaks that model. Compute costs in AI are “fat-tailed.” A small fraction of users running long, complex, multi-step tasks can rack up compute expenses large enough to wipe out the gross profit generated by everyone else. Some Claude Code users reportedly burned $400+ in compute on a $20 monthly subscription. Anthropic’s response was that opaque usage caps rather than transparent token limits says more about the limitations of token-based pricing than any earnings call ever could.
The clearest public-market evidence of this dynamic sits in China. Zhipu AI (Knowledge Atlas Technology), the closest analog public markets have to an Anthropic-like company, has grown revenue ~12x since 2022 to $724M. Cloud API revenue is tripling annually, and nine of China’s ten largest internet companies are customers. Yet gross margins have moved in the opposite direction: they peaked at 66% in 2023 and have since fallen to 41%. Zhipu has raised prices aggressively (API pricing up 83%, and coding plans are up 30%) and still describes itself as operating in a state of “compute panic,” where demand consistently outstrips supply. The cause is mechanical. True compute costs depend on real-time server congestion, batch sizes, memory pressure, and scheduling efficiency. No static pricing model, per token or per seat, can accurately capture the actual cost of serving a request. Brad Gerstner of Altimeter recently claimed Anthropic’s gross margins are in the 50-60% range and rising. Zhipu’s trajectory suggests that claim warrants scrutiny. Growing revenue in AI, it turns out, is relatively straightforward. Preserving margins while doing so is a much harder problem entirely.
AI Birds (Matt Levine, Money Stuff): Allbirds, the sustainable sneaker brand once valued at $4B, announced that it is pivoting to AI compute infrastructure and rebranding as “NewBird AI.” The stock, down 99% from its peak and valued at ~$22M at Tuesday’s close, surged 650%+ by midday. As Matt Levine noted in Money Stuff, there are two levels of analysis here. One: Allbirds is pivoting its business to AI. That seems unlikely to work. Two: Allbirds is pivoting its stock into an AI meme stock. That already did work.
The financing tells its own story. An unnamed institutional investor is providing a $50M convertible facility, but only $5M is committed upfront. The remaining $45M is entirely at the investor’s discretion. The base conversion price is ~$2.99 per share, vs. a stock trading north of $18 today. There is also an “alternate conversion” feature that allows the investor to convert at 93% of the lowest trading price over the prior 10 days, a structure Levine calls a “death-spiral convertible.” Levine lays out the playbook succinctly: find a near-defunct public company with residual brand recognition, fund a pivot to AI, commit only 10% up front, and let the meme do the rest. If one out of ten works, you triple your money. For context, the Allbirds brand previously sold for $39M. As of today, the AI meme attached to it is worth ~$164M.

Quick Hits:

From Unilever: Unilever signed an agreement to acquire Grüns, the gummy supplement brand founded in 2023, for a reported $1.2B. Grüns reportedly exited 2025 with $270M in revenue, implying a ~4.5x revenue multiple. The deal fits Unilever's broader pivot as they divest the food business (Hellmann's, Knorr) via a ~$45B merger with McCormick to free up capital for higher-growth wellness and beauty categories, where it already owns Liquid I.V., Olly, and Nutrafol. A good interview with Chad Janis on TBPN below:
From PR Newswire: Mark Anthony Group has acquired The Finnish Long Drink for a reported $325M, likely at a 2.5-3.0x revenue multiple. The deal adds to a wave of consolidation as organic growth gets harder and harder to come by. Recent transactions include Sazerac’s acquisition of Goodwell Brands (Dirty Shirley and Country Luau) and Buzzballz, AB InBev’s purchase of BeatBox, and Molson Coors’ acquisition of Atomic Brands, producer of Monaco Cocktails.
From Making Cents: Suja Life, the vertically integrated cold-pressed juice brand owned by Paine Schwartz Partners, has officially filed to go public. The company reported ~$330M of sales, ~$40M of Adj. EBITDA (a 12.4% margin), and -a $23M of GAAP net loss.
From The Hollywood Reporter: According to MoffettNathanson YouTube’s revenue has surpassed Disney’s, making it the world’s largest media company.
From Bloomberg: A pound of lean ground beef costs $8.20 in the U.S. as of March, up from $7.48 in March 2025 and $5.68 in 2016 - before factoring in any potential inflationary impact from the Strait of Hormuz closure. Downstream effects for major restaurant chains are likely, with higher prices, lower margins, or some combination of both.
From Snaxshot: Tiger Global invested in Popup Bagels at a reported $300M valuation. The chain currently operates ~30 locations, with an aggressive pipeline targeting ~300 units nationwide.
From Sherwood News: Even historically resilient beer brands like Modelo and Corona are feeling the pressure of lower beer consumption. Last week, Constellation Brands, the beer and wine giant, reported a 10% decline in net sales for the year ended February 2026, driven by a 3% drop in beer sales and the divestiture of lower-end wine labels.

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #24]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-b21https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-b21Tue, 31 Mar 2026 18:09:21 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

Subscribe now

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

The Buy Button: Bain’s 10th annual Insurgent Brands report shows that this year’s cohort of 113 brands captured roughly 36% of total FMCG market growth in tracked channels in 2025, despite accounting for less than 2% of overall market share. This growth is not price-driven: insurgent unit volumes rose approximately 55% year-over-year in a market where overall volumes were flat. Since 2017, nearly 400 brands meeting Bain’s insurgent criteria have generated ~$60B in incremental U.S. retail sales - 50% more than the top three CPG companies combined. Bain expects insurgents could capture as much as 50% of category growth over the next five years. Notably, TSG‑owned DUDE Wipes was named an insurgent brand for the third consecutive year. In short, the fastest-growing segment of consumer products sits largely outside the portfolios of the major CPG companies best positioned to distribute it. At the same time, strategic capital is accelerating into the category. Nearly $9B has been deployed into insurgent-style brands over the past 18 months, including PepsiCo/Poppi, Unilever/Dr. Squatch, e.l.f./Rhode, Hershey/LesserEvil, and Church & Dwight/Touchland. Unilever alone has committed ~€1.5B annually to acquisitions, with a clear focus on U.S. assets. While large CPG R&D engines can reliably produce line extensions, they have rarely built the kind of repeat-purchase loyalty that Olipop or Poppi earned from scratch. CPG deal value rose more than 50% in 2025 despite fewer transactions, indicating that buyers are paying up for quality.
Bain defines insurgent brands as those that generate more than $35M in annual revenue in NielsenIQ-tracked channels, have grown at least 10x faster than their category average over the past five years, have maintained positive growth over the past two years, and have remained independent.
The great peptide debate (TBPN, John Coogan & Jordi Hays): TBPN recently hosted a debate between Martin Shkreli and Superpower CEO Max Marchione on Silicon Valley’s growing fascination with peptides. Peptides are short chains of amino acids that the body naturally produces, but the compounds driving today’s craze go well beyond weight loss. Molecules such as BPC-157 and thymosin alpha-1 are being promoted for everything from gut healing to injury recovery to cognitive performance - largely sourced from Chinese labs and sold online as "research chemicals."
Shkreli’s argument is straightforward: most peptides, particularly compounds like BPC-157 and thymosin alpha-1, break down in the body within seconds, lack a clearly established binding mechanism, and have never been through rigorous human trials. If no one has tested them properly, his view is that it’s unlikely every pharma company simply missed something. Marchione’s counter is structural rather than scientific. Many of these compounds are endogenous (your body already produces them), can’t be patented, and therefore lack a commercial sponsor willing to fund the $1B+ FDA approval pathway. He pointed to ketamine as precedent, a compound dismissed for decades before emerging as a breakthrough treatment for depression. From that perspective, the more relevant question may not be whether individual peptides “work,” but what happens when consumer demand outruns the regulatory framework designed to evaluate it. That tension is already visible. RFK Jr. has pledged to expand access to compounded peptides, though the FDA hasn’t acted. Meanwhile, Eli Lilly’s retatrutide (a next-gen GLP-1 that has produced up to one-third body-weight loss in Phase III trials) still has seven Phase III readouts pending and has not submitted an approval filing. A Polymarket contract pricing FDA approval this year sits at ~23%, albeit with thin liquidity. The fact that GLP-1 drugs are themselves modified peptides makes a blanket “peptides don’t work” position hard to hold. Ultimately, the debate highlights a broader fault line: incumbents benefit from a narrow, capital-intensive approval pathway, while a gray market exists precisely because that pathway excludes unpatentable compounds. If RFK Jr. follows through, the compounding pharmacy market may get a lot bigger.
For LLMs, reviews are the new SEO (sort of) (Modern Retail, Julia Landow): Brands are scrambling to generate more customer reviews, not because the usual conversion logic has changed, but because LLMs increasingly pull from review content when making product recommendations. Stackline estimates that U.S. ChatGPT users make more than 84M shopping-related queries per week, and DTC brands such as Fireclay Tile and dog-food subscription Pawco report that reviews on public forums like Reddit (which many LLMs appear to overweight), Yelp, and YouTube (which Gemini reportedly emphasizes) influence which brands surface in AI-driven search results. If LLMs are synthesizing answers from crawlable web content, then brands with the most detailed, widely distributed reviews are most likely to appear. In effect, this is early-stage SEO logic applied to a new interface. That said, the 84-million-queries figure warrants scrutiny. Query volume is not purchase volume, and there is limited public data showing that LLM-driven product discovery is converting into actual orders. As discussed in previous editions, Amazon, ever the dominant product search engine, has blocked OpenAI’s crawlers from accessing its site and has effectively flipped the script by advertising on ChatGPT, turning it into yet another customer-acquisition channel rather than a true substitute for marketplace search.
There is also Goodhart's Law problem lurking beneath the surface: when a measure becomes a target, it ceases to be a good measure. Reviews were valuable to LLMs because they reflected genuine customer sentiment; the more brands incentivize and manufacture them, the weaker they become as a signal. This mirrors the cycle that degraded both Amazon reviews and Google rankings, but arguably with two important differences. First, opacity. On Amazon, consumers can still eyeball the reviews, look for verification badges, and spot obvious manipulation. LLMs remove that transparency entirely, delivering a synthesized recommendation with no visibility into, or ability to audit, the underlying inputs. Second, surface area. Gaming Amazon meant gaming a single platform. LLMs draw simultaneously from Reddit, Yelp, Google, brand sites, and niche forums, creating incentives to seed reviews everywhere and broadening the scope of the problem.
Dick’s Sporting Goods Is Winning the $40 Billion Youth Sports Arms Race (WSJ, Ben Cohen): American families now spend more than $40B annually on youth sports, according to the Aspen Institute, and Dick’s Sporting Goods has quietly become the default destination. Over the past decade, revenue has roughly doubled to a record $14.1B. With 700+ stores nationwide, Dick’s functions as a one-stop shop for parents, while its loyalty program, now with ~8M premium-tier members spending $500+/year, drives ~50% of total sales. The purchase data generated through this ecosystem allows the company to personalize outreach down to regional sports calendars, such as the start of local baseball seasons. Baseball, in particular, has been a windfall. Average family spending on the sport rose nearly 70% between 2019 and 2024, fueled in part by MLB’s decision in 2019 to loosen cleat color restrictions. That change sparked a youth equipment fashion cycle in which $450 composite bats now routinely sell out on arrival. Dick’s is not just capturing demand, it is shaping it. Its 2016 acquisition of GameChanger, a stat-tracking and livestreaming app, now generates ~$150M in annual revenue and gives Dick’s a level of ongoing engagement with parents that no other sporting-goods retailer can match. On the physical side, Dick’s is scaling its “House of Sport” megastores (150,000 sq. ft., roughly three times the size of a typical store), complete with batting cages and climbing walls, from 35 today to as many as 100 within two years. More time in-store translates into higher spend, and that kind of experiential retail remains something Amazon has yet to replicate.

Quick Hits:

From the NYT: An entertaining look at the fast-growing dog-grooming industry. The U.S. pet-grooming market is expected to generate $19.5B this year and grow to ~$45B by 2036, driven largely by younger, urban pet owners who increasingly extend their own wellness beliefs to their animals.
From The Geek Way: An interesting chart from Andrew McAfee’s latest post serves as a reminder that while the rate of occupational change is on the rise, it remains well below the levels seen in the 1940s and 1950s.
From The Publish Press: Kimberly Ho, CEO of skincare brand Evereden, shared that the company granted equity to three Gen Alpha creators—Madison Rae (14), Embreigh Courtlyn (15), and Kaili Asa (17) as part of its Generation E campaign. It will be worth watching whether this begins to set a precedent for how creator partnerships are structured.
From ShipMatrix: Amazon has overtaken USPS to become the largest delivery provider by volume, another reminder that, at its core, Amazon is an infrastructure company.
From Sherwood News: Continuing the theme of large CPG companies re-entering the M&A market, Danone acquired Huel for $1.2B to strengthen its “presence in functional nutrition.”
From FoodDive: Chobani is investing ~$550M to fund a La Colombe plant expansion. La Colombe’s ready-to-drink business grew 32% year-over-year, with the broader category expected to grow by ~5% this year to ~$8B and reach $11B by 2031.
From Anthropic: A survey of ~81K Claude users across 159 countries found that the leading desire for AI is quality-of-life improvement. While 19% cited professional productivity, many users ultimately want to reclaim time for family, health, or personal growth. Hope and anxiety coexist: top concerns include unreliability (27%), job displacement (22%), and loss of autonomy (22%), with respondents in lower-income countries notably more optimistic than those in the U.S. and Western Europe.

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #23]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-549https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-549Mon, 16 Mar 2026 20:45:18 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

Subscribe now

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

Consumer Trends - Food & Beverage 2026 (Dan Frommer, The New Consumer): The New Consumer and Coefficient Capital released their Spring 2026 F&B trends report, reinforcing what we’ve been tracking on protein. In their survey of 3,000 U.S. consumers, protein ranked as the #2 nutrient Americans want to consume more of this year (behind vegetables), while Google Trends shows searches for “protein” reaching another all-time high. Instacart basket data tells the same story from the purchase side: five of the platform’s ten fastest-growing grocery categories are protein-oriented (Greek yogurt, protein bars, protein drinks, cottage cheese, and ground beef). Per the survey, 65% of Gen Z and Millennials say they are actively trying to eat more protein, and most adopting this behavior within the past year. GLP-1 users are accelerating the trend further, with 73% actively seeking more protein. As that user base grows (current penetration estimates are ~10%), protein demand is likely to continue rising. We do not appear to be at peak protein yet.
The beverage data is more mixed. While 71% of Gen Z and Millennials say they “love” trying new beverages and 55% say their drink choices are part of their personality, their stated favorites remain firmly incumbent brands. When asked to name their top three drinks, Coke/Coca-Cola and coffee at the top ranked highest among younger consumers, ahead of any functional or better-for-you options, with Coca-Cola actually ranking higher among Gen Z and Millenialls than among older cohorts. Despite high experimentation, incumbents continue to dominate core preferences. As far as this data goes, Buffett’s age-old thesis on Coca-Cola’s stickiness appears intact, with no next-generation soda brands breaking into Gen Z and Millennials’ top-10 rankings. That said, the data also points to where new brands can win: 70% of young consumers say they choose beverages based on how they make them feel (energy, calm, focus). Instacart basket data backs this up, with protein drinks, probiotic and prebiotic beverages, and matcha powder among the fastest-growing beverage categories. The best read of these seemingly conflicting signals may be that younger consumers are not replacing their core go-to drinks, but are instead adding new ones across additional occasions.
Younger generations move from barstools to barbells (Joe Wadford, Bank of America): A new report from the Bank of America Institute puts numbers behind a trend the consumer sector has been sensing for some time: young Americans are drinking less. Alcohol spending as a share of household budgets is near a 40-year low, per BLS data, and SAMHSA reports 3.9 million fewer 21-34-year-olds binge drinking compared to a decade ago, with participation dropping from ~40% to 30%. BofA card data, however, suggests this shift is more about moderation than abstinence. Spending at liquor, wine, and beer stores fell 5% YoY in January, while bar spending rose ~4%. Consumers are cutting back on at-home consumption while preserving alcohol’s role in social occasions. All together, the data points to a rebalancing of how and where alcohol is consumed, rather than an outright rejection. We should also note that roughly half of Gen Z remains under 21, limiting the availability of comprehensive alcohol consumption data for this cohort.
So where are those dollars going instead? BofA card data shows Gen Z and Millennials’ spending growth on fitness-related categories (gyms, golf, and country clubs) meaningfully outpacing their spending growth at bars. This aligns with The New Consumer survey data we covered above: 38% of Gen Z and Millennials say they are in “optimization mode,” actively making choices to improve energy, appearance, and long-term health. BofA Global Research suggests GLP-1 adoption could further accelerate moderation trends among older cohorts as well. Consumer dollars are not disappearing, but they are clearly shifting toward health, wellness, and self-optimization.
Agentic Commerce is a Mirage (Eric Seufert, Mobile Dev Demo): OpenAI is scaling back Instant Checkout, the feature that allowed users to buy products directly within ChatGPT, and is instead routing purchases through third-party apps. Eric Seufert argues this pullback was predictable for three structural reasons:
(1) The 4% transaction fee is untenable for merchants already operating on thin net margins of roughly 3-8%
(2) Merchants lack control over volume, since OpenAI’s algorithm determines what products are surfaced, unlike advertising where spend can be dialed up or down to manage demand
(3) Advertising is structurally superior to affiliate-style commerce because bidding allows merchants to set their own price for demand, rather than accepting a flat percentage imposed by the platform.
The timing is also notable. Less than a week before the Instant Checkout pullback, Amazon announced a $50B investment in OpenAI (with plenty of caveats around the structure of the financing). Given Amazon’s massive advertising business ($68.6B in 2025, +22% YoY), it has little incentive to support independent shopping agents that could cannibalize its core platform. Seufert’s hypothesis is that any go-forward partnership between OpenAI and Amazon will more likely resemble Amazon’s prior deals with Meta and Snap, where Amazon placed its own first-party ads within those platform’s surfaces without directly sharing any customer data. Amazon is also reportedly exploring ways to expand its demand-side-platform to place ads across third-party chatbot apps. This is a complete inversion of the agentic commerce thesis: rather than chatbots disintermediating Amazon, Amazon turns chatbots into another ad surface. As Seufert puts it, you cannot beat Amazon by building a better search layer on top of commerce. You would have to build a better Amazon.
AI-generated images are breaking ecommerce refund systems (Jamie George, Ravelin): A recent blog post from fraud-prevention company Ravelin (content marketing, but directionally useful) highlights a growing vulnerability in ecommerce refund processes: consumers are using AI tools like ChatGPT and Gemini to fabricate photos of “damaged” products in order to claim refunds on items received in perfect condition. It’s as simple as receiving an order, uploading a photo, prompting an AI to add cracks or stains, and submitting the image as “evidence.” In a Ravelin survey of 6,200 shoppers, 65% said AI has made it easier to falsely claim refunds. The tactic exploits a foundational assumption embedded in many ecommerce systems: that photos don’t lie. For low-value items, perishables, and products that are expensive to return-ship, merchants often waive returns entirely and issue refunds based on photographic evidence alone. Fraudsters know this. Ravelin argues that addressing this at scale requires a shift in the unit of analysis from the individual refund claim to the customer. Rather than attempting to authenticate every photo, merchants need to segment customers by risk profile and apply friction dynamically based on factors like transaction history, account age, and lifetime value. This issue is worth watching closely because the problem scales with AI capability, and any company whose refund process relies heavily on customer-submitted photos is exposed.
The "Slop Bowl" Slowdown Is a Quality Problem, Not a Budget Problem: Recent coverage in The New York Times has framed the slowdown at chains like Cava, Chipotle, and Sweetgreen as consumers tighten their wallets. And sure, restaurant prices are up 22% since 2022, with Chipotle and Sweetgreen raising prices by ~25% and ~20%, respectively. But Dan Frommer made a sharper observation in The New Consumer: the problem at Sweetgreen isn’t that $14 is too much for a salad, it’s that $14 buys what still looks and tastes like a $9 salad. Anecdotally, consumers appear perfectly willing to spend more when quality justifies it. In New York, concepts like ThisBowl have no trouble drawing crowds for $17+ salmon bowls. That’s not a pullback, it’s a trade-up! Cava’s own numbers complicate the belt-tightening story too: its strongest results are coming from its lowest-income neighborhoods. If the slowdown was really about shrinking budgets, that outcome would be counterintuitive. Instead, the data suggests a clearer divide: chains that maintain or improve product quality are holding traffic, while those that let the product drift are losing it. Sweetgreen’s own data makes the point. After overhauling its miso salmon prep (longer marination, better presentation), salmon sales velocity jumped 20%. The solution isn’t cheaper bowls, but better ones.

Quick Hits:

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #22]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-252https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-252Fri, 27 Feb 2026 18:03:34 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

Subscribe now

The Roundup:

An Investor’s Guide to Gen Alpha (Josyana Joshua and Dina Katgara, Bloomberg): Gen Alpha (born 2010-2024) is roughly 2 billion people strong globally, already commands more than $100 billion in annual U.S. direct spending power, and is forming financial habits earlier than any prior generation thanks to native access to digital commerce. Bloomberg highlights five trends shaping where that money goes:
1. AI native. 64% of teens use AI chatbots, per Pew. For Gen Alpha, AI isn’t a tool they adopted later in life; it’s a interface for school, content creation, and search.
2. Gaming as a social network. 85% of U.S. teens play video games. Roblox, Fortnite, and Discord have joined, if not supplemented, Instagram and Facebook as the primary social layer. Gen Alpha communicates while playing, not while scrolling. That said, gaming broadly is facing pressure: legalization of sports betting and prediction markets, plus a macro battle for attention against short-form content, are all weighing on engagement.
3. Wellness from birth. 59% of Gen Alpha cite mental health as a top concern. Therapy platform Headway logged 650,000 appointments from Gen Alpha users last year.
4. Resale is mainstream. The global secondhand market is projected to reach $360 billion by 2030, growing 10% annually, per BCG. This cohort treats vintage as a feature, not a compromise.
5. Physical retail is back. 66% of Gen Alpha shoppers prefer buying in-store, and just 3% would rather shop exclusively online, per Numerator. AMC is leaning in with themed merchandise drops, while retailers like Ralph Lauren are expanding in-store cafés to extend dwell time.
Shopify and the SaaS Panic (Ben Thompson, Stratechery): Shopify posted Q4 revenue of $3.7 billion (up 31% YoY) and $124 billion in gross merchandise volume, both beating estimates. The stock fell 13% after adjusted EPS of $0.48 missing consensus expectations of $0.51. This feels like a SaaSpocalypse panic in miniature: the market is treating every software company as an AI casualty, regardless of whether AI is a headwind or a tailwind. As Ben Thompson eloquently argued on Stratechery last week, Shopify’s logic should run in the opposite direction. Shopify sits atop trillions of data points from billions of transactions across millions of merchants, and it’s deploying that data through Shop Campaigns, a new Product Network that cross-sells inventory across Shopify stores, and the Universal Commerce Protocol co-developed with Google to power agentic commerce.
Anyone who has used Shopify knows that it does far more than build storefronts. It runs checkout, payments, taxes, shipping, fraud prevention, and fulfillment. That full-stack coverage is exactly what makes Shopify hard to displace and easy to layer AI on top of. When an AI agent surfaces a product for a consumer, someone still has to process the payment, ship the box, and handle the return.
Orders from AI search to Shopify stores are up 15x since January 2025 (off a small base). Thompson’s broader point is that AI-driven commerce will favor the long tail of merchants, which is Shopify’s core constituency. Shopify now represents over 14% of U.S. e-commerce. To anyone running a business on Shopify, this sell-off smells like pattern-matching gone wrong: Shopify got lumped in with software companies AI threatens, when it’s likely one of the platforms AI makes more valuable.
Waymo’s $126B Fundraise (Om Mallick, On My Om): Waymo closed a $16 billion round ($13 billion of it from parent Alphabet) at a $126 billion valuation, roughly 360x its ~$350 million annualized revenue run rate. For context, Uber trades at 3-4x revenue. Om Malik digs into the physics of the business, and the numbers are stubborn. Waymo’s ~2,500 Jaguar I-PACEs average about 25 trips per day and already operate ~16 hours daily. To reach its target of one million weekly rides by the end of 2026, the fleet would need to more than double to 5,500-6,000 vehicles, at roughly $175,000 per car, which is over $600 million in vehicle costs alone. New city launches offer limited near-term relief. Austin, nine months in, accounts for just 8% of rides with roughly 200 vehicles. The valuation debate echoes the famous 2014 exchange between Aswath Damodaran and Bill Gurley over Uber. Damodaran sized Uber against the existing taxi and limo market and arrived at a modest valuation. Gurley argued that Uber would expand the market by making rides cheaper, faster, and more accessible than car ownership in many contexts, and he was right. Valuing Waymo requires a similar leap: that autonomous vehicles don’t merely substitute for rideshare trips, but create entirely new demand by being cheaper (no driver), safer, and available 24/7. The obvious downside is that, unlike Uber’s asset-light model, Waymo’s growth is constrained by atoms, not bits. Every incremental ride requires a physical car, a mapped city, regulatory approval, and charging infrastructure. Both sides of that argument deserve serious consideration.
Why Are Chinese EVs So Cheap? (Gregor Williams, Rhodium Group): The conventional explanation for Chinese EV pricing points to subsidies and cheap labor. Rhodium Group's cost analysis tells a different story. Comparing BYD's Seal to Tesla's Model 3 (both produced in China), BYD enjoys a roughly $4,700 per-vehicle cost advantage, about 15% of the Model 3's sticker price. Subsidies explain just 5% of that gap (~$292 per vehicle). Labor productivity actually favors Tesla, which generates six to seven times more revenue per employee than BYD.
The real advantage is vertical integration. BYD manufactures far more components in-house, avoiding supplier markups worth an estimated $2,369 per vehicle relative to Tesla. The second largest factor is lower overhead per vehicle ($1,766 advantage), largely due to concentrating R&D in China.
So where does this leave U.S. automakers? In January, Ford CEO Jim Farley pitched the Trump administration on allowing Chinese EV makers to manufacture in the U.S. through joint ventures where American firms retain controlling stakes. Bloomberg described this as a mirror image of China’s own requirements decades ago, when GM, VW, and Toyota needed local partners to access the Chinese market. The proposal was rebuffed, but the logic is hard to ignore. Ford already licenses CATL's battery technology (CATL is a Chinese battery manufacturer) for a $2-3 billion plant in Michigan. GM is importing CATL cells for the new Chevy Bolt. Ford is reportedly in talks with Geely (a Chinese auto group) on manufacturing partnerships. Canada isn't waiting around: in January it cut Chinese EV tariffs from 100% to 6.1%, contingent on joint ventures with Canadian firms within three years. The cost gap Rhodium identifies is structural, not temporary. Western automakers must decide whether to replicate these advantages through partnerships, or simply cede the ground.

Quick Hits:

From Uber’s Q4 Earnings: Uber shared early data showing that AV launches accelerated overall usage, particularly among first-time riders, evidence that autonomy is expanding the category. AVs complete 30% more rides per day and arrive 25% faster than human drivers. One open question: given highly variable demand, will it make sense for AV manufacturers to build to peak demand or build to a baseline, with surges being met by traditional drivers?
From Sherwood: Despite peanut butter ranking as America’s favorite condiment in a recent YouGov survey, ketchup remains the most ubiquitous, present in ~84% of U.S. households. The category is seeing renewed M&A activity, with Bachan’s acquired by The Marzetti Company and Tapatio by Highlander Partners
From ScreenRant: McDonald’s Valentine’s Day caviar-and-McNuggets kit crashed the company’s website and sold out instantly. Kits have resold on secondary marketplaces for as much as $600, for a bundle that includes a 1-ounce tin of caviar, crème fraîche, a mother-of-pearl spoon, and six nuggets that would set you back ~$100 retail.
From RestaurantDive: Sticking with McDonald’s, the company is rolling out craft sodas and energy drinks to the McCafé lineup after a “highly successful” 500-store test
From Sherwood: Wendy’s had a brutal year capped by an even worse Q4, with SSSG of -11.3%. The previous low point was Q2 2020 at -4.4%, during COVID. Compared with McDonald’s recent performance, the contrast underscores how critical sustained innovation is to long-term SSSG

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #21]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-6achttps://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-6acTue, 17 Feb 2026 20:05:54 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

Subscribe now

The Roundup:

A Reflection on SEO & AI Search in 2025 (Lily Ray): Lily Ray, VP of SEO Strategy at Amsive, published a comprehensive reflection on SEO and AI search in 2025 that cuts through the noise around “Generative Engine Optimization” (GEO). She argues that despite an industry-wide identity crisis sparked by ChatGPT’s rise to roughly 800 million weekly active users, GEO is not a replacement for SEO but an extension of it. Ray’s observes that much of what is marketed as revolutionary GEO advice is simply repackaged SEO best practices: structured data, clear content hierarchy, and concise answers to user questions. In practice, if you rank well in Google today, you are already well positioned for AI-powered search.
That said, large language models (LLMs) do introduce meaningful mechanical differences that marketers need to understand. An important mechanical change is query fan-out. When a user asks ChatGPT something like “What’s the best CRM for a 50-person sales team?”, the model does not search that exact phrase the way Google traditionally would. Instead, it decomposes the prompt into multiple sub-queries and runs them in parallel, such as: “best CRM software mid-market 2025,” “CRM comparison Salesforce HubSpot,” “CRM pricing 50 users,” and “CRM ease-of-use reviews G2.” The model then synthesizes those results into a single response.
New tools like Google’s Gemini Grounding API and iPullRank’s Qforia allow marketers to see the actual sub-queries LLMs generate, creating a valuable new data source for content strategy. The practical takeaway is not to chase individual fan-out queries (they are highly personalized, long-tail, and shift with every prompt), but to aggregate them and identify recurring themes. If fan-out analysis consistently shows that models break “best CRM” queries into dimensions like pricing, integrations, ease of onboarding, and peer reviews, that signals which facets your content needs to cover in depth. In other words, the unit of optimization is shifting away from individual keywords toward comprehensive topical coverage. This is not a new idea in SEO, but fan-out data gives brands a much sharper lens into which subtopics actually matter to systems generating answers.
Multimodal content has also become more important. Models such as GPT-5 and Gemini 3 natively process video, audio, and images, meaning a YouTube transcript, podcast episode, or virtual asset can now directly inform an AI-generated answer. At the same time, off-site signals have been meaningfully reprioritized. According to Semrush data, Reddit, Quora, LinkedIn, YouTube, and review platforms like G2 are among the most frequently cited domains in AI search results. This shifts emphasis away from traditional link-building toward earning organic brand mentions and co-citations across trusted third-party platforms.
It’s worth remembering that the largest AI search surface is still Google. By mid-2025, Google’s AI Overviews had reached ~2 billion monthly users and appeared in an estimated 16-25% of all U.S. queries. During that same period, Google’s search revenue grew 17% YoY in Q4 2025 to $63 billion, with CEO Sundar Pichai noting on the earnings call that the company sees “no evidence of cannibalization” from AI features. Brightedge data shows Google still commands over 90% of global search market share, while AI platforms like ChatGPT account for less than 1% of global referral traffic.
Taken together, this suggests the overall search pie is growing, not shrinking, even as its composition changes. AI Overviews are reducing click-through rates for certain informational queries, while longer, more complex, and more personalized searches are expanding. For brands, the strategic imperative is not to abandon SEO for GEO, but to double down on clear, authoritative, well-structured content that performs across both traditional and AI-powered search surfaces simultaneously.
The Creator Fund: Slow Ventures’ Creator Fund, led by Megan Lightcap, invests at the holding company level of creators who are building real businesses, not just audiences. The thesis rests on a set of interlocking ideas. First, we are firmly in the era of the individual. Institutional trust is eroding, while people increasingly follow people. Creators who own their reputation and are directly accountable to a community offer higher signal and higher fidelity than traditional brands. Second, niche beats mass. Slow is explicitly “short the mass market,” arguing that follower counts are increasingly decoupled from conversion and revenue. Algorithms are now precise enough that undifferentiated content rarely earns sustained distribution. What matters is not what someone consumes, but why they follow. Lightcap’s sourcing framework focuses on five categories where creator-audience bonds run deepest: skill-based hobbies (needlepoint, cycling), widening knowledge gaps (hormone health, personal finance), taboo topics (sex, compensation), the paradox of choice (skincare, romance novels), and cultural milieu (faith, geography, ethnicity).
Slow’s thinking on exits is very novel. The fund invests at the creator holdco level but expects liquidity to come from monetization events at the individual brand level beneath it (think SKIMS going public, or a creator-led food brand selling to a CPG company). The logic is that creators take multiple shots on goal, and the businesses they build are rooted in community insight rather than pure distribution. At the same time, a newer path is emerging: the entire creator holding company exiting as a single entity. Martha Stewart pioneered this model with her 1999 IPO. More recently, Rich Sparkle, a Nasdaq-listed firm, acquired Khaby Lame’s full commercial operation for $975 million. MrBeast’s Beast Industries has also hinted at a future IPO spanning media, Feastables, a theme park, and more. While these whole-entity exits are likely limited to creators operating at enormous scale, they materially expand the range of credible liquidity outcomes for the asset class.
For most creators, however, Slow believes the more durable strategy is to build individual brands that can survive independent of the creator’s personal distribution. That independence increases enterprise value, reduces key-person risk, and creates greater optionality at exit.
The Live Shopping Update (Dan Frommer, The New Consumer): Live shopping generated $22 billion in North America and Europe last year, more than doubling year-over-year, according to Whatnot's State of Live Selling Report. Whatnot, the largest live shopping platform in the West, accounted for more than $8 billion of that total. The model pairs live video with real-time auctions and instant purchasing across categories like sports cards, beauty, and electronics. Growth has been especially pronounced in beauty (up ~800% YoY) and electronics (up ~400%).
Engagement metrics suggest this is more than a passing novelty. Shoppers spend an average of 95 minutes per day in the app. Sellers who stream daily average $69,000 per month in sales, and roughly one in eight sellers now works on the platform full-time. These figures point to a level of habit formation and economic viability that earlier Western attempts at live commerce struggled to achieve. Previous efforts by Meta and Amazon largely fizzled, in part because live video was bolted onto an existing shopping experience rather than designed around it. Whatnot appears to have succeeded by building from the ground up around niche communities and highly passionate sellers. In China, live commerce is approaching a $1 trillion market and accounts for ~60% of all e-commerce. By comparison, the U.S. remains in the early innings.
The Booming Business of Luxury Grocery Stores: (Lane Florsheim & Chavie Lieber, WSJ): A new wave of luxury grocers is booming in Manhattan. Meadow Lane, which opened in November, was profitable by week two and now rings in 1,000+ daily tickets, selling everything from $15 chicken nuggets to $750 caviar to lines of Gen Zers waiting behind a bouncer. Rigor Hill Market has doubled sales since 2023 on items like $26 cauliflower bowls and $23 lasagna. Erewhon, the LA original, now has 11 locations and is expanding east.

Grocery is a notoriously thin-margin business (even Amazon has struggled to make it work), but these stores are succeeding by selling prepared food and an “experience” priced closer to restaurants, while maintaining lower build-out and operating costs than a traditional restaurant. The target customer is rarely doing a full weekly shop; they are more often picking up a rotisserie chicken and a $15 juice on the way home from work. It’s also interesting to understand who is spending. As noted in prior posts, the top 30% of earners account for roughly 75% of all U.S. consumer spending, with the top 10% representing about half. These stores are designed for that cohort, as well as aspirational consumers adjacent to it. This concentration of spending power helps explain why luxury grocery can thrive even as broader affordability concerns mount. It also helps explain why the format works as a discovery and distribution channel for premium CPG brands: the customer walking into Meadow Lane is the early adopter willing to pay $15 for a glass bottle of oat milk, post about it, and help pull the product into wider distribution.

Subscribe now

Quick Hits:

Share The Roundup

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #20]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-3d9https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-3d9Fri, 30 Jan 2026 18:58:50 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

OpenAI embraces the inevitable: ads come to ChatGPT (Mobile Dev Memo, Eric Seufert): OpenAI announced it will begin testing ads in ChatGPT’s free and $8/month Go tiers, a notable reversal for a company whose CEO once described advertising as a “last resort.” CFO Sarah Friar disclosed that compute and revenue have been growing in parallel at roughly 3x per year (from 0.2 GW to 1.9 GW, and from $2B ARR to $20B+ from 2023-2025), leaving the company without the operating leverage typical of traditional tech economics, even if R&D spending likely overstates how closely those curves track. Advertising is the only model capable of reshaping that trajectory, expanding the addressable market by turning free users into ad inventory and deepening monetization by allowing ARPU to scale with advertiser demand. With 900 million WAU and $1.4 trillion committed to infrastructure, subscriptions alone cannot close the gap. The timing carries risk. Google does not run ads inside Gemini, yet Gemini underpins AI Overviews and AI Mode, both of which already carry advertising. Google can subsidize its chatbot with its dominant search business, while OpenAI must now compromise parts of its user experience to fund operations. The initial ad format, contextual placements at the bottom of relevant responses, is also limiting. It constrains inventory to contextually aligned moments and raises questions about whether answers serve users or advertisers, and it fails to address that the most compute-intensive, and therefore most expensive, sessions are the ones where ads are least appropriate. Imagine ChatGPT is tutoring you on a complex concept: how would it feel for an ad to pop up in the middle of that exchange? As Seufert notes, OpenAI needed to begin this shift years ago; the real destination, personalized ads decoupled from chat context (the “Instagram-style” model Altman has described), remains far off.
Universal Commerce Protocol (Stratechery, Ben Thompson): Google and OpenAI have each introduced “open” commerce protocols to enable agentic shopping. OpenAI’s Agentic Commerce Protocol allows purchases directly within ChatGPT, with merchants paying a 4% fee to OpenAI on top of Shopify’s standard transaction costs. Google, by contrast, is charging nothing. Its Universal Commerce Protocol, developed with Shopify, Walmart, Target, and others (though notably not Amazon) carries no fee at all. That pricing gap underscores the same tension visible in the companies’ respective advertising strategies: OpenAI must monetize every possible surface, whether through ads or transaction fees, while Google does not need Gemini itself to make any money directly. Gemini’s role is simply to keep users inside Google’s ecosystem, where monetization is already robust. By offering its commerce protocol for free and taking nothing at checkout, Google can subsidize merchant adoption, while OpenAI introduces friction by charging a fee.
Do Commodities Get Cheaper Over Time? (Construction Physics, Brian Potter): Brian Potter’s analysis of 124 commodities finds that agricultural products have historically been clear winners in deflation. Of the 25 U.S. crops he examined, 24 are cheaper in real terms today than at the start of their respective time series, with 17 experiencing declines greater than 50%. Potter hypothesizes that the more a production process resembles manufacturing (automated, continuous, and machine-driven) the more its costs follow the familiar downward curves of factory-made goods. Industrial diamonds illustrate this dynamic: once production shifted almost entirely to lab-grown processes, real prices collapsed by 99.9% between 1900 and 2021. However, this pattern has notably reversed in recent decades. Since 2000, only four crops (cotton, peanuts, tobacco, and sweet potatoes) have declined in real price; the other 21 have increased. Meat prices show similar trends with beef up 41% since 1995 and pork up 15%. Potter’s analysis of 20-year rolling windows reveals a distribution that was once left-skewed (more price declines) and has now become right-skewed (more price increases). The implication is that the long-running agricultural productivity improvements that drove decades of falling real input costs may have reached diminishing returns. It is increasingly reasonable to assume we are operating in a structurally different commodity environment than the one that prevailed from roughly 1860 to 2000.
America’s Own Goal: Who Pays the Tariffs?: The Kiel Institute analyzed 25.6 million shipments worth nearly $4 trillion in trade and found that foreign exporters absorb only about 4% of U.S. tariff costs; the remaining 96% is passed directly to American importers and, ultimately, consumers. U.S. customs revenue jumped by roughly $200 billion in 2025, representing a transfer of wealth from American households to the U.S. Treasury, not from foreign producers. Event-study analyses reinforce this pattern. After a 50% tariff was imposed on Brazil in August 2025 and 25-50% on India, export prices barely moved. Instead, trade volumes collapsed. Indian exporters maintained their dollar prices and simply shipped less to the U.S.
Exporters can behave this way because, as long as global demand is strong, they can redirect supply to Europe or Asia rather than slash prices. On the U.S. side, supply chains are sticky and importers cannot rapidly switch suppliers, so foreign exporters face less competitive pressure to absorb tariff costs than you might expect. The institute concludes that tariffs function less like a tool to extract concessions from foreign producers and more like a selective consumption tax on Americans. They also create deadweight losses through reduced product variety and supply-chain disruption. The political claim that foreign countries “pay” the tariffs is, according to the data, only about 4% true.
Beyond Meat’s New Protein Drink Feels Like a Cry for Help (New Fare Partners, Elly Truesdell): Beyond Meat’s latest product reveal is a carbonated pea-protein drink called Beyond Immerse. This is the same company that once persuaded Whole Foods to merchandise plant-based patties in the meat case and saw its stock peak at $239 per share in July 2019. Today it trades under $1 and is selling 36-packs of slim cans through a “Beyond Test Kitchen” landing page. The pivot speaks volumes: the U.S. plant-based meat category has contracted from $1.43B in 2022 to $1.17B in 2024, and Beyond’s Q3 2025 revenue fell 13.3% year-over-year to $70.2M amid shrinking distribution and softer velocities. The strategic logic isn’t entirely off base. Protein is a massive category and beverages offer a far larger TAM than a meat-alternative segment that never fully found product-market fit. Still, it’s jarring to see the Beyond wordmark (cow logo intact) wrapped around a slim can that looks like a Celsius knockoff. The brand was built on bleeding beet juice and disrupting the butcher case; whether that equity carries into the beverage aisle, or simply confuses the few consumers still open to the brand, is unclear. Beyond also never had a slam-dunk product in its core category, cycling through multiple burger reformulations without landing on something enduring. Against that backdrop, Beyond Immerse reads less like a coherent expansion strategy and more like a company searching for a reason to exist.

Quick Hits:

From Google: Another marketing-AI case study to add to the growing library: Colgate-Palmolive partnered with Google on a gen-AI pilot that produced video ads 4-6x faster, at lower cost, with equal or better consideration lift. (Note: this is a Google case study)
From Business Insider Africa: Rich Sparkle Holdings, which is historically a capital-markets administrative services firm, has acquired Khaby Lame’s company, Step Distinctive Limited, in an all-stock deal valued at $900M. This move marks a dramatic strategic pivot toward influencer-led digital commerce
From Sherwood: America’s love of spicy sauces continues, but the question now is whether we have hit “peak-heat”
From The WSJ: Amazon is joining the big-box league with its largest store ever: a 230,000-square-foot location in the Chicago suburb of Orland Park
From CNBC: Amazon has angered a wave of online retailers by cross-listing their products on Amazon’s marketplace without consent. It’s an ironic twist given Amazon is currently suing Perplexity for essentially the same behavior!

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #19]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-704https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-704Sat, 17 Jan 2026 16:24:55 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

Quick Hits:

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #18]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-7f0https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-7f0Mon, 05 Jan 2026 20:28:43 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

The 2026 Consumer Trends Report (Dan Frommer, The New Consumer): Dan Frommer’s annual Consumer Trends report, published in partnership with Coefficient Capital, has become essential reading for anyone trying to understand where American spending is headed. Three themes stood out this year:
Creator Economy: Creator-led brands continue to enjoy a structural advantage in awareness and distribution. Across beauty, apparel, health & wellness, and food & beverage, creator brands command two to three times more Instagram followers than even the top 5% of traditional brands in each category. TikTok Shop is the clearest proof point, projected to hit $15B in U.S. GMV in just its second full year. But the power law is stark: the top 1% of creators (roughly 8,450 of 845,000 tracked) drove 53% of TikTok Shop’s $7.7B in video sales over the past twelve months. The flywheel rewards reach, and reach compounds.
GLP-1s: The fear that GLP-1s would crater food spending hasn’t materialized. One year after starting medication, users actually index 37 points higher on CPG food and beverage spending versus the average household. When you eat less, taste matters more, and you can afford to trade up. 93% of users report weight loss, 91% say they feel like a different person, and 90% express interest in brands specifically formulated for them. With 23% of U.S. households now including at least one GLP-1 user (up from 19% a year prior) and oral formulations on the horizon, this is a durable tailwind for premium products.
Health & Wellness: A generational fault line is emerging in healthcare. 53% of Gen Z and Millennials believe they could manage their health better than their doctor given the right data and tools; only 25% of Gen X and older agree. Behaviors reflect this mindset: 40% of younger consumers who have taken blood tests have uploaded results to ChatGPT for analysis, and 44% would rather pay out of pocket than navigate insurance. For this cohort, self-optimization is a status symbol (58% agree that being fit signals status), and they’re willing to act on it outside traditional systems.
The full report covers much more, including the state of the “two Americas” wealth divide, prediction markets going mainstream, and how life satisfaction varies across demographics. It’s worth your time.
The Great Car Reset: The EV transition increasingly looks like an industry shakeout masquerading as a technology shift. This month, Ford announced a $19.5B charge as it abandons larger electric vehicles and rebrands its Tennessee EV Center as the “Tennessee Truck Plant” to build gas-powered pickups instead. The EU simultaneously softened its 2035 combustion engine ban from 100% to a 90% emissions reduction, allowing hybrids and vehicles with internal combustion engines (gas-powered cars and trucks) to continue selling with e-fuels. Western automakers and regulators are retreating in unison. Ford's pivot to plug-in hybrids and extended-range EVs may make sense in the near term, but it comes with trade-offs. Maintaining a full combustion supply chain limits potential cost savings as batteries keep improving. The question is: how will these hybrids compete with full EVs in 10 or 15 years?
China, meanwhile, is running the smartphone playbook. Bill Gurley, speaking on Tim Ferriss Show after touring Xiaomi’s SU7 factory, described brutal provincial competition winnowing hundreds of manufacturers down to a few champions, creating significant domestic overcapacity. The SU7 itself is telling: Marques Brownlee (MKBHD), YouTube’s most influential tech reviewer with 20M+ subscribers, called the $42K sedan “a $75,000 car” after two weeks of testing. Ford CEO Jim Farley admitted he drove one for six months and “doesn’t want to give it up.” The Xiaomi factory Gurley toured runs at one-third the employees per car compared to legacy automakers. Ford missed Japan. Ford missed Korea. Farley says they can’t miss China. Based on these announcements, they may already have. Are legacy carmakers that can't make profitable, desirable EVs destined to become the Nokias and Blackberrys of this transition?
The Relentless Rise of YouTube (Christopher Grimes, Financial Times): According to the Financial Times, YouTube now leads all TV and streaming consumption in the U.S., ahead of Netflix, Disney, and Amazon Prime. Since 2024, Americans have watched YouTube primarily on televisions, not phones. The platform has paid creators over $100B since 2021, and channels earning $100K+ from connected TVs grew 45% year-over-year. This month, YouTube secured exclusive streaming rights to the Oscars starting 2029, ending a 50-year relationship with ABC. As MoffettNathanson analyst Michael Nathanson put it: “For the past year or so everyone in Hollywood has woken up and figured out how big YouTube is.”
The incumbents are responding. Netflix’s recent moves make more sense through this lens. The $83B Warner Bros. acquisition wasn’t just about library content or blocking Paramount, it was about owning assets YouTube can’t replicate: decades of premium IP, from HBO’s prestige catalog to DC Comics, that provide churn prevention in ways algorithm-driven recommendations cannot. The same logic applies to Netflix’s podcast deal with Spotify, which brings 16 shows from The Ringer exclusively to Netflix and off YouTube. Expect 2026 to bring more 'anti-YouTube alliance' moves as the platform continues swallowing the media industry.
The Impact of Visual Generative AI on Advertising Effectiveness (Hyesoo Lee, Vilma Todri, Panagiotis Adamopoulos, Anindya Ghose): A rigorous new study from NYU and Emory researchers compared human-created ads to AI-generated ones in both lab settings and a real Google Ads campaign with over 100K impressions. The results were striking: ads generated entirely from scratch by visual AI tools like Midjourney and DALL-E achieved a 19% higher click-through rate than ads created by human experts. By contrast, AI-modified ads (where AI simply edited existing human designs) showed no improvement at all. The gap widened further when AI was given complete creative freedom to design both the ad and product packaging, adding another 15% lift. This pattern is the inverse of what researchers have found with text-based AI, where large language models (LLMs) perform better editing copy than generating it from scratch. The mechanism appears rooted in what reinforcement learning pioneer Rich Sutton calls “The Bitter Lesson”: when humans impose structural constraints (forcing AI to work within an existing layout), they limit the model’s ability to leverage patterns learned from billions of training images. Remove those constraints, and the system outperforms the experts. The implications for the $470B digital advertising market are significant. Meta’s stated vision of advertisers simply providing a business objective and a credit card while AI handles everything else now has empirical backing. Yet the study also found that disclosing AI involvement cratered CTR by 31.5%, a tension that will only intensify as regulators push for AI transparency requirements. The takeaway seems to be that visual AI should be deployed as an ideation engine at the front end of the creative process, not as an editor at the back end. Human expertise remains valuable for refinement and brand consistency, but the blank-canvas work in advertising may increasingly belong to machines.

Quick Hits:

From Sherwood News: Aldi is winning over both high- and low-income shoppers, and increasingly, so is Walmart
From Chartr: Memory lane has never been busier with consumer interest in retro experiences and products at an all-time high
From CNN: TikTok signed binding agreements to sell its U.S. operations to Oracle, Silver Lake, and Abu Dhabi's MGX, with closing expected January 22. The $14B valuation implied by the deal is strikingly low: TikTok U.S. generated ~$10B in revenue in 2024, meaning the deal values the business at roughly 1.4x sales, far below analyst estimates of $40-60B. ByteDance will retain 19.9% and license its recommendation algorithm to the new entity.
From The Wall Street Journal: Sleep-aid beverages are booming: 22% of U.S. adults now drink them, up from 16% in 2023. Som Sleep doubled sales in 2025, MoonBrew grew 150%, and AG1 launched AGZ at $79-99/month. Wearables are making sleep data visible and fueling demand, even as evidence on key ingredients remains mixed.
From The Wall Street Journal: Nex Playground, a $249 motion-tracking console from a little-known Silicon Valley startup, outsold Xbox during Black Friday week, capturing 14% of U.S. hardware unit sales vs. PS5's 47% and Switch 2's 24%. The company sold 5,000 units in 2023, 150,000 in 2024, and is on track for 600,000 units and $150M+ revenue in 2025. The device is essentially a modern Kinect aimed at families with young kids, a segment Sony and Microsoft have abandoned.

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #17]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-773https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-773Tue, 16 Dec 2025 00:29:06 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

The Complete History & Strategy of Coca-Cola: (David Rosenthal & Ben Gilbert, Acquired) The latest Acquired episode explores what may be the most elegant business model in consumer history. Coca-Cola does two things: it manufactures syrup from a secret formula and spends money on marketing. Everything else, from production to bottling to distribution, is handled by others. This structure dates back to an 1899 deal where Asa Candler gave away bottling rights for a symbolic $1 (never collected), inadvertently creating an asset-light system that could scale globally without Coca-Cola investing in production or distribution. Today, 2.2 billion servings are consumed daily, and the Coca-Cola company still operates with remarkably few employees relative to its reach. Bottlers take on capital risk and operational complexity, while Coca-Cola retains control over the brand and the formula. The episode also uncovers several origin details worth knowing: the drink was invented by a morphine-addicted Confederate veteran, originally contained cocaine and four times today’s caffeine, and introduced the first manufacturer coupon redeemable at retail. It’s an episode worth listening to.
Holiday Spending Hits Records: Black Friday online spending surged to a record $11.8 billion this year, up 9% from 2024, pushing holiday e-commerce toward a projected all-time high of $253.4 billion. This comes despite surveys suggesting consumers planned to cut back. PwC found Gen Z expected to slash holiday budgets by 23% (a reversal from last year’s projected 37% increase), and overall planned spending was down 5%. Yet the registers tell a different story. The pattern is familiar: consumers report feeling squeezed, say they will tighten up, then spend anyway. When everything feels expensive and uncertain, people still want the dopamine hit of a deal, which is made even more accessible as Meta and TikTok collapse the friction to purchase. Whatever the vibes say, revealed preferences remain intact.
The Vibecession, Explained (Or Not) (Scott Alexander, Astral Codex Ten): Scott Alexander digs into a stubborn paradox: real median household income is up 33% since the Boomer era, Millennials and Zoomers earn more at the same age than previous generations, youth unemployment was near historic lows until recently, and the bottom quintile has seen the strongest relative wage gains of the past decade. Yet consumer sentiment sits at multi-decade lows, and the prevailing narrative insists young people have been permanently locked out of opportunity. Alexander methodically stress-tests the usual suspects. Housing? Mortgages doubled post-pandemic, but rents as a share of income are only up ~10%. Debt? Student loan balances have actually declined since 2010 as federal caps eroded with inflation. Inequality? It has plateaued over the past decade, not worsened. None fully explain the gap. The more persuasive culprits are structural. Technology has collapsed application friction, so any given opportunity now draws thousands of candidates and requires exponentially more credentialing to achieve the same outcome. The creative class has concentrated in a handful of metros where effective cost of living runs 50-90% higher than the national average, and these are precisely the people who write about the economy. Meanwhile, media negativity has increased steadily for decades. Alexander doesn’t land on a single answer, but the framework he builds is valuable for understanding why sentiment and spend so often diverge.
Weight Loss Drugs are Bringing A New Commercial Reality (Henry Dimbleby, Financial Times: Early data suggests 2025 will mark the first year on record where the volume of food consumed in America declines. The driver, according to Henry Dimbleby, is GLP-1 drugs like Mounjaro and Ozempic, which are being adopted faster than any pharmaceutical in history. In the UK alone, 2.5 million people have started using them in just two years; statins took 17 years to reach similar penetration. Users cut calorie intake by 15-35%, disproportionately from the high-margin, hyper-palatable snacks that fuel FMCG profits. For an industry built on high volume and low margins, even a sustained 5% drop creates stranded assets in production lines perfectly engineered for what used to be predictable demand. Dimbleby argues the required pivot is from “volume” to “value per bite” - competing on nutrient density rather than addictive taste alone. The recent Unilever ice cream spinoff may be an early signal of who’s reading the room.
AI Eats The World (Benedict Evans): Ben Evans’ latest annual presentation frames generative AI as the next platform shift in the progression from mainframes to PCs to the web to smartphones - but with a critical difference: we don’t know the ceiling. Previous platforms had physical limits that made next year’s capabilities predictable. With LLMs, we are flying blind on capability growth, which helps explain the FOMO-driven investment surge. The hard questions remain open. If models are near-commodities without network effects, where does durable advantage come from? Evans suggests two paths: go down the stack and win on capital scale (aircraft, AWS, chips), or go up the stack and win on product and network effects (software). His “infinite interns” framing surfaces an underappreciated tension: does automation mean doing the same work with fewer people, or more work with the same people? And if recommendation systems evolve from correlation to genuine intent understanding, what gets unbundled? Evans doesn’t claim to have answers, but the questions he’s asking are the right ones.
Estée Lauder’s Fragrance Bet (Claire Yubin Oh, Sherwood): Estée Lauder’s fragrance division is now its fastest-growing segment, up 14.4% year-over-year, and the company is leaning in. Jo Malone recently launched an AI Scent Advisor built on Google Cloud, designed to replicate the in-store consultation experience through natural language prompts. The tool is designed to help online shoppers confidently spend $100+ on a fragrance they have never smelled, a behavior that’s becoming surprisingly common. Gen Z and younger consumers increasingly “blind buy” fragrances, driven by TikTok, review sites like Fragrantica, and trends like “scent layering” and “smellmaxxing.” Piper Sandler data shows fragrance spending by teenage boys jumped 44% last year. After three years of sales declines, Estée Lauder has explicitly positioned fragrance as the core of its turnaround. The company opened a Fragrance Atelier in Paris last month and holds a strong portfolio of prestige scent brands, including Tom Ford, Le Labo, Kilian, and Frédéric Malle. For a category historically dependent on physical retail and sensory experience, the AI advisor is a bold bet that digital discovery can scale without sacrificing premium positioning.

Quick Hits:

From the Financial Times: John Burn-Murdoch explains Baumol’s Cost Disease: the phenomenon where wages rise in industries with little productivity growth, like healthcare and education, because they must compete for workers with industries where productivity has improved.
From QSR: QSR released its annual “Top 50 Fast-Food Chains” list. No surprise: Chick-fil-A, Raising Cane’s, and In-N-Out Burger remain in rarefied air, proving that an intense focus on a signature product pays dividends.
From the New York Times: A Pew Research report shows most American teenagers use YouTube and TikTok daily, and roughly one in five say they are on one of the two platforms “almost constantly.”
From Sherwood: On December 8, shares of The Magnum Ice Cream Company, which spun off from Unilever and is home to Magnum, Ben & Jerry’s, Cornetto, and more, opened at €12.20 in Amsterdam, valuing it at around €7.9 billion ($9.2 billion), slightly below analysts’ expectations.
From Fast Company: Gen Alpha holiday wish lists are shifting away from toys toward digital economies, with 43% requesting in-game currency, ahead of both consoles (39%) and physical video games (37%).

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #16]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-6a5https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-6a5Wed, 19 Nov 2025 13:01:34 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

Ozempic users are going crazy for deli meat, fries, and handbags (Dan Frommer, The New Consumer): Circana data shows that 23% of U.S. households now include at least one GLP-1 user, and by 2030 these households are expected to account for 35% of all food purchases in America. The most significant shift is in motivation: 78% of users now cite weight loss as their primary reason for taking the medication, up from 37% in 2021, while only 43% cite diabetes, down from 75%. Nearly half of users are now “weight-only,” double the proportion from two years ago. Deli meat has emerged as a big winner as users chase protein to preserve muscle mass, while wine, premixed cocktails, and carb-heavy items are losing ground. Interestingly, GLP-1 users over-index on fries at lunch and cookies at dinner. When caloric intake is chemically suppressed, small indulgences become meaningful luxuries, creating opportunities for food companies to focus on premium offerings in small packages that justify higher per-unit pricing. Retail behavior is also changing. GLP-1 users are spending more at club stores like Costco and Sam’s Club, and on pure-play e-commerce, while reducing purchases at dollar stores and traditional grocers. Beyond food, and identity shift appears to be influencing lifestyle choices: 86% of users say they feel like a different person since starting medication, which correlates to increased spending on handbags, casual athletic wear, and jeans. Despite these category and channel shifts, overall spending remains strong. GLP-1 households were already high-value consumers before starting the medication, spending 9% more on food and 20% more at restaurants, and those levels barely change during active use. These consumers are not reducing total spend, they are reallocating it.
The Digital Advertising Industry is Changing (MBI): Digital advertising market growth accelerated to 18.4% in Q3 2025, with Meta and Amazon continuing multi-year share gains at Google’s expense. For the first time, Google’s market share fell below 50%, losing 760 basis points (bps) since Q3 2021, while Meta gained 300 bps and Amazon picked up 400 bps. The numbers mask the underlying dynamics: roughly half of Google’s decline stems from its low-margin network business, and YouTube is intentionally trading ad-supported users for premium subscribers who generate higher gross profits. Internal documents from Google’s antitrust trial suggest this retreat is strategic rather than defensive.
The bigger question is what happens when search becomes AI-mediated. Today, performance marketing works because measurement is direct: a user searches for running shoes, clicks an ad, and makes a purchase. AI-driven commerce breaks that loop. When ChatGPT recommends a product, the conversion path disappears and advertisers can’t track whether campaigns influenced the recommendation or optimize against conversions. This measurement gap could push brands back towards traditional brand advertising, focusing on awareness and recall, in hopes of influencing what the AI recommends. If more brands compete for top-of-mind positioning, auction prices could rise on Meta and Amazon - platforms built on performance marketing infrastructure, not brand campaigns.
AI E-Commerce (Ben Thompson, Stratechery): The digital economy increasingly operates in two modes: push and pull. Push is when platforms like Meta, TikTok, and YouTube show you products you didn’t know existed. Pull begins with a query on Google or Amazon, where consumers are searching for something specific. In an interview with Ben Thompson, Michael Morton of MoffettNathanson argues that AI threatens pull, not push. When you ask ChatGPT for running shoes, it scans expert reviews, checks product specs, and delivers an answer in seconds. That eliminates the extended search behavior that Google and Amazon monetize. The implications are most significant for Amazon, despite its strong position in AI commerce. MoffettNathanson data shows that when ChatGPT, Gemini, and other AI models evaluate e-commerce queries, Amazon consistently ranks first on the three criteria that matter most: price, trustworthiness, and speed. But there’s a catch. Amazon’s $60 billion advertising business (roughly 90% retail-focused) depends on consumer inefficiency. The company has disclosed that shoppers typically click through 14 products before making a purchase. Each click represents auction inventory where brands compete for placement. AI search compresses that journey to one or two clicks. The funnel collapses, auction density drops, and the profit engine breaks even as Amazon “wins” the recommendation. This dynamic creates an opening for competitors. Walmart, with no massive ad business to cannibalize, benefits by partnering with ChatGPT giving it access to 60 million U.S. users over 25 - more than use the Walmart app and more than half of Amazon’s app user base. The move forces Amazon into a prisoner’s dilemma: sit out and lose share, or join and risk undermining its most profitable business line. Still, as noted in the last issue, this thesis hinges on whether ChatGPT can truly divert purchase queries away from the Amazon app, which is by no means guaranteed.
Generative ad creative and the creative control barrier (Mobile Dev Memo, Eric Seufert): Mondelez has invested $40M in an AI-powered tool designed to automate ad creative, developed in partnership with Publicis and Accenture. The goal is to cut production costs by 30-50% compared to existing agency spending. Currently, the tool is producing 8-second social clips for brands like Chips Ahoy and Milka chocolate, but the ambition is bigger: TV spots by next holiday season and a Super Bowl commercial by 2027. The economics make sense when you consider the alternatives. Building an in-house creative team isn’t always structurally feasible for a company with Mondelez’s scale and brand portfolio. So the choice isn’t “pay agencies $300K per animation” versus “do it in-house for $50K.” It’s “pay agencies $300K” versus “invest $40M upfront in AI, then produce at a fraction of the cost.” Another constraint is channel diversity. Mondelez runs campaigns everywhere from Instagram to the Super Bowl so it can’t rely on Meta’s creative tools for Facebook ads and Google’s tools for YouTube, as each platform would produce slightly different aesthetics. Maintaining brand consistency requires a single system that works across all channels. Platform tools like Meta’s Advantage+ Creative are fine for smaller advertisers who can tolerate the occasional misfire (one men’s apparel brand recently got an AI-generated ad featuring an elderly woman). But large-scale CPG brands protecting iconic products can’t risk visual inconsistency or brand drift. This dynamic suggests that the advertising industry will bifurcate: platforms will serve the long tail of SMBs comfortable with automation, while large brands build proprietary systems to maintain control.
Why Beauty and Personal Care Brands Win (and Fail): Lessons from True Beauty Ventures’ Rich Gersten: True Beauty Ventures ran a post-mortem across its 21-brand portfolio and uncovered a counterintuitive insight: brands stuck at $2-3M in revenue carry the same risk profile as pre-revenue companies. By contrast, brands that break through $5M tend to scale rapidly, while that that stall at $2-3M rarely recover. If the risk looks identical, the strategic move is to invest earlier with better terms and more influence before mistakes become entrenched. This insight has reshaped how True Beauty deploys capital. Initial checks now run $2-3M to weight investments equally when risk is highest, followed by aggressive concentration in winners. K18 represented 10% of their first fund and returned 80% of called capital in just 3.5 years. Two other investments, Vacation and Crown Affair, now account for 30% of total capital deployed across the portfolio. Early-stage beauty, diversification isn’t the goal; conviction is. Success depends on developing conviction fast enough to back up the truck.
Later in the conversation, Gerson addressed why the M&A market faces a supply-demand imbalance: too many brands seeking liquidity and too few active buyers. Makeup is particularly stuck. Even profitable, scalable brands like Makeup by Mario and MERIT struggle to find exits. Gerson traces the issue to the 2012-2016 M&A bubble, when strategic buyers went heavy on makeup acquisitions that didn’t deliver. P&G exited beauty entirely. When strategics pass, growth equity funds get nervous about their own exit paths and follow suit. L’Oreal remains the most aggressive acquirer because its portfolio spans all categories, geographies, and distribution channels - including professional - so any brand fits somewhere. True Beauty has raised its investment bar for makeup to “very high,” despite it being the largest category. Category selection can matter more than brand quality when exit markets freeze over.
How the best products remove guilt (Rex Woodbury, Digital Native): Gruns launched in 2023 and reached a $100M revenue run rate earlier this year. By August, the company was tracking toward $300M. Founder Chad Janis noticed the success of AG1 and similar greens powders but recognized a flaw: the form factor was wrong. People disliked mixing powders into water, which was messy and inconvenient. His solution was simple but powerful: switch to gummies. The product isn’t fundamentally different from Flintstones vitamins, yet it revitalized a tired category by tapping into consumer psychology better than incumbents. Brian Sugar, an early Gruns investor, framed the insight this way: the company took a form factor people loved but felt guilty about - gummy bears - and flipped it on its head. Eating gummies was no longer a vice; it became responsible behavior. The formula is clear: take something people know they should do but dislike, and turn it into something they enjoy. Gruns now sits alongside Poppi (acquired by Pepsi for $1.95B) and Rhode (acquired by e.l.f. Beauty for $1B) as proof that big companies still emerge from stale or saturated categories when founders bring fresh insights into human behavior.

Quick Hits:

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #15]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-271https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-271Fri, 31 Oct 2025 13:03:37 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

Walmart on ChatGPT, Walmart (and Amazon) motivations (Ben Thompson, Stratechery): Walmart’s decision to join OpenAI’s Instant Checkout program is a strategic move for the #2 e-commerce player. By listing its catalog on ChatGPT, Walmart trades some control for access to customers who might otherwise default to Amazon. Given the different demographics of Walmart shoppers and ChatGPT users, these are likely incremental customers, not just cannibalized from existing channels. Walmart’s third-party marketplace, similar to Amazon’s and hosts thousands of independent sellers alongside Walmart’s own products, stands to benefit as ChatGPT recommendations could redirect sales from platforms like Shopify or Amazon. The more interesting question is whether Amazon is going to join. ChatGPT enables discovery-based commerce for products you didn’t know existed, while Amazon is much more intent-driven, focused on products consumers already want. Amazon has already shifted product searches away from Google, the reigning aggregator king, by training customers to start their shopping journey in the Amazon app. For ChatGPT to disrupt Amazon, it would need to fundamentally change how consumers discover products. If OpenAI amasses enough influence, it could auction product links to the highest bidder, eroding retailer margins as Google did in search.
Taco Bell tries to woo younget consumers with Live Más Café’s flashy beverages (Amelia Lucas, CNBC): Taco Bell is expanding its Live Más Café concept to 30 locations by year-end, building beverage stations inside existing restaurants, rather than standalone stores like McDonald’s now-defunct CosMc’s. Customers order at kiosks and watch “Bellristas” prepare specialty drinks, driving a 16% year-over-year increase in beverage orders. The menu spans blended coffees, lemonade-based refreshers, and energy drinks, all designed to help Taco Bell reach its goal of $5 billion in beverage sales by 2030. Over 60% of Taco Bell orders this year have included a drink, and early results are strong: the Irvine location sells over 900 drinks daily, Chula Vista 750, with specialty drinks making up a significant share. At roughly 300 incremental specialty drinks per day per location and a $5 average ticket, that translates to approximately $500,000 in incremental annual sales per location. The shop-in-shop approach leverages existing real estate and labor, solving the unit economics challenge that plagued CosMc’s, and giving Taco Bell a beverage halo without the capital intensity of new builds.
How Memory Creates Brand Moats (Bordalo, Pedro, Giovanni Burro, Nicola Gennaioli, Gad Nacamulli, and Andrei Shleifer. “Ads as Cues.” NBER Working Paper Series, No. 34387, October 2025.): Why do Coke and Pepsi spend billions on advertising for products everyone already knows? New research from Harvard suggests ads aren’t about information, but memory: they cue consumers to recall past pleasurable experiences with a brand, triggering purchases. Testing this across 20 highly advertised, undifferentiated product categories using Nielsen data, the researchers found the effect is strongest for frequent buyers. An average Coke drinker needs 166 seconds of weekly ads to buy an extra can, while heavy drinkers (95th percentile) need just 44 seconds. Ads build long-term memory associations that maintain habits, not just immediate sales. Competing brands both advertise heavily, boosting total category demand. Memory creates switching costs, giving undifferentiated brands pricing power. Brand moats in CPG are built by owning mental real estate through repeated consumption and consistent advertising.
There’s a shocking disparity between how high-income and low-income earners feel about the economy (Alex Harring, CNBC): JPMorgan’s latest Cost of Living Survey reveals a stark divide in economic confidence by income bracket. High-income respondents rated their outlook at 6.2 out of 10, with more than half choosing ratings between 7 and 10. Low-income consumers scored just 4.4 out of 10, with less than a quarter rating between 7 and 10. The University of Michigan’s consumer survey shows the top third of earners reporting sentiment roughly 25% higher than the bottom third over the past two years. The divergence matters because, as we covered in Issue #13, the top 10% of earners now account for 50% of all consumer spending. High-income confidence translates directly to sustained consumption in premium categories while lower-income cohorts pull back on discretionary purchases. However, consumer sentiment remains a poor predictor of actual spending, as we explored in Issue #11. Real personal consumption has grown steadily for 45 years despite sentiment swings. Wealth gains among high-income households drive spending power, even when optimism lags. The bifurcation is real in balance sheets and consumption patterns, but history suggests spending grows regardless of what any income cohort reports feeling about the economy.
How Depop Captured the Heart of Gen Z (Max Berlinger, NYT): Depop processes 400,000 daily uploads through a peer-to-peer marketplace that blends Instagram’s aesthetic with eBay’s transaction model. Unlike ThredUp’s centralized consignment, covered in Issue #14, where sellers ship items and the company handles everything, Depop lets users curate their own shops, photograph their own inventory, and build followings. The result is what CEO Peter Semple calls preserving the “energy” of a good thrift store digitally. Users stalk sellers’ likes to find new items, sellers write blog-like product descriptions, and the whole experience feels personal rather than industrial. One seller earns 20% of his income through his Depop shop. Another user scored a Bottega Veneta skirt for $20. The chaos is the point. Depop’s Q2 2025 U.S. sales jumped 54% year-over-year by leaning into community over efficiency. Etsy paid $1.6 billion for Depop in 2021, betting that younger consumers want social commerce, not sterile transactions. The trade-off: packages sometimes arrive with cat food inside, sizes get mixed up, and sellers can be flaky. But Gen Z accepts this because buying from an individual with a point of view beats algorithmic perfection. Depop’s new “Outfits” feature doubles down on discovery and aesthetics. The platform’s value depends on a steady supply of desirable secondhand inventory, not just volume.
The remarkable rise of the Honeycrisp and Cosmic Crisp apples(David Crowther, Claire Yubin Oh, Sherwood News): With fall upon us, we thought we’d highlight the story of how the Honeycrisp apple went from near-rejection to industry titan. In 1982, University of Minnesota horticulturist David Bedford rescued the MN 1711 variety, convinced its lackluster performance was due to poor growing conditions. The cross-pollination that created Honeycrisp happened in 1961, but the variety languished for two decades before Bedford gave it another chance. Patented in 1988 and released in 1991, Honeycrisp didn’t reach grocery store shelves until 1997. The gamble paid off: by 2019, Honeycrisp generated $16.5 million in royalties for the university, making it their third-most-profitable invention after an anti-HIV drug and a gene-editing cancer treatment. Today, farmers call Honeycrisps “divas” and “moneycrisps” in equal measure. The variety bruises easily, requires hand-clipping due to thin skin, and only 55-60% of the crop makes it to market. Yet consumers will pay a premium: Honeycrisp retails at $1.88 per pound, compared to $1.49 for Gala and $1.26 for Red Delicious. Production of the three crunchiest varieties (Cosmic Crisp, Pink Lady, Honeycrisp) has surged 3,391%, 63%, and 17% respectively over five years. Claire Yubin calls out that a 2013 academic paper essentially predicted this: when tasting apples, consumers prioritize sweetness and crispness above all else. The apple industry has been reshaped by texture preference and one breeder’s hunch that a rejected tree deserved a second look.

Quick Hits:

Trader Joe’s $2.99 Halloween mini canvas tote bags drew Black Friday-sized crowds in early October, with the pastel version from April reselling on eBay for over $100 and selling out within hours at many locations. (Samantha Bergeson, Vanity Fair)
U.S. online grocery hits a record $12.5B in September, up 31% year-over-year. (Abbas Haleem, Digital Commerce 360)
YouTube continues to dominate:
Gucci is struggling:
Counter-Strike is a multiplayer shooting game where players can buy and trade cosmetic items (called “skins”) like decorative weapon finishes, knives, and gloves. These items don’t affect gameplay, but carry real-world prices ranging from a few dollars to tens of thousands. Valve, the game’s developer, adjusted exchange rates for these digital items on Wednesday, instantly wiping out $1.75 billion in market value. (Cecilia D’Anastasio, Bloomberg)

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #14]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-e6bhttps://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-e6bThu, 16 Oct 2025 13:39:28 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

The Roundup:

Startup’s Are Eating Big Food’s Lunch (WSJ, David Wainer): In 2024, insurgent brands captured 39% of incremental category growth, despite representing less than 2% of the food, beverage, and household products market. This is more than double their share from the previous year before, signaling a significant shift in consumer behavior. Legacy players are being squeezed in two directions: budget-conscious shoppers are turning to private labels for better prices, while affluent consumers are gravitating towards challenger brands that offer higher quality. Although total food sales returned to growth in 2024, large-cap food makers continue to see declining volumes. The meat snacks category illustrates this shift. Valued at $7.7 billion, the category grew 14% over the past year, with nearly all of that expansion going to upstarts like Chomps. Chomps scaled from $70 million in 2021 to roughly $660 million, vastly outpacing traditional rivals Slim Jim and Jack Link’s. The brand succeeded in repositioning jerky from a nostalgic, macho snack to a sugar-free, high-protein option for health-conscious buyers, particularly women.
Distribution advantages once held by incumbents have narrowed with the rise of e-commerce and social media. While Amazon still rewards scale through lower fulfillment costs and more efficient ad spend, it no longer requires a national sales force or slotting fees to reach consumers. Small brands can now access the same digital shelf as Campbell’s, test products in real time based on reviews and conversion data, and iterate quickly without the six-month lag of traditional retail cycles. Platforms like TikTok and Instagram allow brands with modest marketing budgets to reach exactly the consumers who want sugar-free jerky or prebiotic soda, whereas broadcast TV required millions of dollars for meaningful reach. Legacy players built advantages when distribution access was scarce and marketing required massive budgets for broad reach. These shifts have eroded the structural moats that once protected big food companies, creating space for challenger brands to win on product quality and positioning before economies of scale become decisive. The impact is evident in stock performance: over the past three years, shares of Campbell’s, Nestlé, PepsiCo, and General Mills have declined, while the S&P 500 has climbed 80%.
Have Chicken Strips Become The Fast Food Panic Button of 2025 (Sherwood News, Tom Jones): Beef prices have reached historic highs as decade-old underinvestment in America’s cattle herd finally catches up with demand (Sherwood News). The U.S. imported 30% more beef in the first half of 2025 compared to the same period in 2024, but even that hasn’t been enough to contain soaring costs. Meanwhile, chicken consumption continues its six-decade climb, with Americans expected to consume 103 pounds per person this year, which is nearly triple the amount in 1965.
The widening price gap is creating structural shifts across the food system. Four companies control 81% of U.S. beef production, and JBS recently posted nearly $2 billion in profit last year, capitalizing on tight supply. Tyson is responding to the chicken boom by launching portable protein cups with 30 grams per serving, betting that the 80% of consumers who now incorporate protein into at least one meal daily will pay for convenience. Fast food operators are reading the same data. McDonald’s made its first permanent menu addition in four years with McCrispy Strips, which helped boost second-quarter comps. Taco Bell followed two months later with strip-loaded tacos and burritos and continues to outpace the category. Wendy’s recently launched Tendys with six sauces after watching same-store sales slide 3.2% in the U.S. through the first half of the year. Chicken strips have become the panic button, which is an easy-to-execute item that chains deploy when traffic softens and margins compress. It’s no surprise, given that chicken costs less than beef, moves faster, and gives operators room to compete on price without undermining unit economics.
Corporate Anxiety Is Fueling a Multimillion-Dollar Industry of Gen Z Translators (WSJ, Katie Deighton): Companies are spending millions to decode a generation that won’t stop telling them exactly what it thinks. Gen Z’s global spending power is expected to reach $12.6 trillion by 2030, up from $9.8 trillion in 2024, and corporate anxiety around how to engage this cohort is creating an entire industry. Edelman’s Gen Z Lab has generated $76 million since 2022 by positioning 400 Gen Z employees as paid consultants. NinetyEight, founded by 1998-born graduates, consults for PepsiCo and Meta. UTA acquired JUV Consulting and hosts ZCON, a conference designed to educate executives on youth trends. The stated challenge is that cultural inputs now number in the hundreds of thousands, TikTok trends die in days, and monoculture is dead. The real challenge is that companies struggle to operationalize the insights. It’s ironic that the generation easiest to hear from is the one brands are spending the most to understand. Youth marketing agencies have existed since 1945 but what’s changed is the velocity mismatch between trend cycles and corporate decision-making. Haleon hired Edelman’s Gen Z Lab to shape Advil’s TikTok strategy, and Barilla adjusted its protein pasta messaging after consulting the division. Both changes likely took months to implement, while the insights that informed them became less relevant by the day. The real question isn’t whether Gen Z is different enough to warrant specialized translation. It’s whether companies built for six-month campaign cycles can compete in markets where relevance has a one-month half-life, regardless of how much they pay twentysomethings to explain TikTok.
How Din Tai Fung Became America’s Top Earning Restaurant Chain (Bloomberg, Brent Crane): Din Tai Fung generates $27.4 million per U.S. location annually, nearly triple what Nobu brings in, making it the highest-earning restaurant chain in America despite operating just 17 locations nationwide. CEO Albert Yang, 32, says he doesn’t care about profit, insisting that culinary quality, service, and efficiency drive decisions, with profit following naturally if you do things right. The business began in 1970s Taiwan when Yang’s grandfather, Bing-yi, converted half of his cooking oil shop into a dumpling restaurant after supermarkets disrupted his delivery business. For decades, the restaurant served Taiwanese diners consistently well, until a 1993 New York Times listing as one of the world’s best restaurants triggered a surge in demand and international expansion. Yang’s parents opened the first U.S. location in a California strip mall in 2000.
When Yang and his brother took over in 2021, they began obsessively monitoring Yelp and Google reviews, tying corporate bonuses to ratings. The chain’s performance stems from operational fanaticism, with Yang calling high guest ratings his “north star.” To address common complaints like cold food or inattentive service, Din Tai Fung increased cook times to ensure hotter temperatures and built rigorous training programs. Dumpling production is standardized to 16 grams of stuffing, 5 grams of dough, and 18 folds, with training taking over two months. The company scaled from 15 dumpling producers per location to 60, eliminating the bottleneck that had constrained growth. The family refuses most franchise offers and has no plans to go public, prioritizing quality control over expansion speed. Marketing is minimal beyond Facebook and Instagram, but the chocolate mochi dessert dumpling went viral on TikTok organically, with four out of five new guests citing the platform. While most restaurant chains chase unit count, Din Tai Fung built America’s highest-grossing chain by opening fewer locations and making each one exceptional.
“Magic: The Gathering” is just the tip of a $1 billion digital iceberg (Sherwood News, Charlie Hall): Hasbro turned 100 this year, but is betting its future on a bold pivot towards video games. While Consumer Products revenue is down 16% year-to-date, the Wizards & Digital segment grew 28%, prompting Hasbro to raise its full-year guidance in July, based entirely on the strength of Magic: The Gathering. Operating at margins near 50%, Magic crossed $1 billion in annual revenue in 2022, driven by Universes Beyond, which integrates licensed characters from other franchises. Releases featuring Final Fantasy and Lord of the Rings each generated over $200 million in sales, with upcoming releases set to include Avatar, Marvel, The Hobbit, and Star Trek. As the chart below shows, Hasbro released over 12,000 Magic cards in 2024, up from ~3,000 in 2018. This surge in content fuels revenue, it’s also creating fatigue among longtime players who didn’t expect to see Spider-Man fighting planeswalkers.
That success is funding a bigger gamble. Hasbro has invested roughly $1 billion building four internal game studios, aiming to transform itself from a toy manufacturer into a AAA game publisher. The strategy signals a belief that developing games in-house will yield better returns than licensing IP to established publishers. If any of the studios underperform, the shortfall hits a balance sheet increasingly dependent on Magic to carry the company. Licensed digital games like Monopoly Go have proven successful, but Hasbro’s move suggests it’s banking on deeper control and long-term upside. The transformation from toy company to gaming company is already underway. The real question is whether the gaming company Hasbro is building can survive without Magic’s 50% margins indefinitely funding it.
How ThredUp resells 17 million garments every year (Business Insider, David Valerio): Business Insider recently released a behind-the-scenes look at how ThredUp processes 17 million garments annually through its distribution centers, using AI for tagging, pricing, measurements, and photography. A 100,000-square-foot automated hanger system pulls items five times faster than manual operations. Since pivoting from peer-to-peer resale in 2011, ThredUp now manages the entire selling process, accepting everything from Gap to Gucci. Items that pass inspection are QR coded, photographed in 360 degrees, and listed with algorithm-determined pricing that learns from previous sales. Original owners receive payouts ranging from a few dollars to several hundred, depending on brand and listing price. In 2024, the company introduced AI shopping tools including a virtual stylist and image search, which increased purchase likelihood by 85%.
The business model depends on solving for inventory ThredUp can’t sell. The company doesn’t dispose of unsold items directly, instead it passes them to aftermarket partners who manage the global secondhand supply chain. Typical thrift operators sell only ~25% of donated clothing, leaving the remaining 75% to be baled and exported. The U.S. and Canada ship roughly 780,000 metric tons of used clothing annually, and the average American garment now lives in a closet for just 2.5 years before entering the resale system or getting discarded. This creates a massive downstream ecosystem that ThredUp doesn’t control but relies on to keep inventory moving. The company only accepts women’s and children’s clothing, as men tend to buy less frequently and wear items longer, making resale economics more difficult. Looking ahead, ThredUp launched Resale-as-a-Service in May 2025, positioning itself as the AWS of recommerce by handling the operational complexity that most brands can’t build internally. The challenge will be maintaining sell-through rates high enough to justify the infrastructure while managing what happens to the 75% that doesn’t sell.

Quick Hits:

US EV sales are expected to hit 410K this quarter, accounting for 10% of new vehicle purchases. The record high is likely driven by consumers accelerating purchases ahead of the federal tax credit expiration.
Once Upon a Farm, the baby food company co-founded by Cassandra Curtis and Ari Raz in 2015, has filed to go public with trailing twelve-month net sales of $201 million as of June 2025. CAVU led the most recent funding round, a $52 million Series D in 2022.
Taco Bell ranked as the fastest drive-thru by total time for the fifth consecutive year
YouTube, Netflix, and Nike topped Beano Brain’s 2025 survey of the coolest brands among Gen Z

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]><![CDATA[The Roundup by TSG Consumer - Issue #13]]>https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-487https://tsgconsumer.substack.com/p/the-roundup-by-tsg-consumer-issue-487Fri, 03 Oct 2025 18:40:52 GMTTSG Consumer is a leading private equity firm, and trusted partner in building brands people love since 1986.

What you can expect from this newsletter

A bi-weekly round-up of all things consumer. This is a space for people from across the consumer space to stay up to-date on the latest trends, connect with like-minded peers, and continuously learn from each other

Each issue will feature a curated selection of recent consumer headlines - stay tuned for insights that keep you up to speed in the ever-evolving consumer space!

"The greatest part of what we know is the least part of what we know not."
- Michael de Montaigne

The Roundup:

What’s Behind Resilient Consumer Spending (MBI): U.S. consumer spending remains strong, driven by the top 10% of earners who account for half of all spending and still have capacity to spend, while the bottom 80% have depleted pandemic savings and returned to pre-pandemic debt levels amid inflation. This concentration of spending power helps explain why GDP appears strong even as consumer sentiment deteriorates - the aggregate data reflects the prosperity of a thriving minority, not the strain felt by the broader majority. That strain is increasingly visible in overlooked economic indicators. Cardboard box demand has collapsed to 2016 levels, prompting International Paper to close 9% of U.S. containerboard capacity as both e-commerce and home moves slow. Meanwhile, Hamburger Helper sales jumped 14.5% as families stretch ground beef, that now costs a record $6.63 per pound. Purchases of rice, canned meat, and beans are also rising, which is a classic sign of recession-era shopping. Among lower-income households, shipping and moving activity has slowed, and shelf-stable proteins are replacing fresh meat - behavioral shifts that often precede broader economic downturns, even when headline numbers still look strong.
AI Monetization on YouTube (Ben Thompson, Stratechery): At its recent “Made on YouTube” event, YouTube quietly introduced a potentially game-changing commerce feature: AI-powered product tagging that surfaces purchase links at the precise moment products are mentioned in videos - turning the platform’s 500 hours of content uploaded every minute into a massive, shoppable catalog. While creators welcomed the time savings from automated tagging, the broader implication is its monetization potential: every object featured in a YouTube video - from power tools in home improvement tutorials to kitchen gadgets in cooking videos - can now serve as a potential transaction point, either through affiliate revenue or as biddable advertising real estate. This isn’t just an incremental improvement, it’s a strategic leap. YouTube is harnessing its unmatched scale and Google’s AI infrastructure to create an entirely new commerce layer. With over $100 billion paid to creators in the past four years, many of whom also earn through brand deals, and the platform will now multiply that economic engine by making every relevant object monetizable. While Meta has been talking about this vision of turning all content into potential commerce, Google is doing it. YouTube already dominates in living room streaming, now surpassing Netflix, and it nurtures the creator economy, positioning itself to capture commerce at a granular level. What was once a marketing channel could evolve into the world’s largest product discovery and transaction platform, where purchase intent emerges organically from content people are already watching.
Sephora’s Affiliate Program (Clara Ludmir, Forbes): Staying with the theme of creators and affiliate links, Sephora announced “My Sephora Storefront,” an in-house affiliate platform launching in October. This initiative allows influencers to build shoppable storefronts directly on Sephora.com, positioning the brand to compete with third-party platforms like LTK and ShopMy. Creators receive the standard affiliate toolkit - customizable storefronts, shoppable links, and performance analytics - but shoppers can browse and purchase without leaving the Sephora ecosystem. The timing reflects the market opportunity: affiliate marketing is expected to drive over $210 billion in U.S. e-commerce sales this year, rising to nearly $280 billion by 2028. By building the platform internally, Sephora captures commission fees that would otherwise go to intermediaries and gains direct data into what converts from creator content. To further sweeten the deal for creators, Sephora is offering perks like exclusive product launches and other benefits. The move raises questions for the broader creator commerce ecosystem. Large retailers are incentivized to build their own affiliate platforms, but creators now face the challenge of managing an increasing number of proprietary platforms. This forces a trade-off between convenience and deeper brand relationships with better data insights. Third-party platforms like LTK built their businesses on the assumption that retailers wouldn’t invest in this infrastructure. That assumption is now being tested, as retailers increasingly recognize that controlling creator relationships means controlling a critical customer acquisition channel.
The “Sephora Kid” Industrial Complex (Casey Lewis, After School): The Sephora kid phenomenon has reached remarkable scale. In the first half of last year, households with tweens and teens accounted for a third of all “prestige” beauty sales, contributing to $9 billion and in U.S sales for Sephora and $11.3 billion for Ulta. The creator-driven approach was validated when Sincerely Yours sold out at Sephora in under an hour (see last issue’s coverage of the chaotic launch.) Meanwhile, Ulta now hosts $42-per-guest birthday parties “with tweens and teens in mind”, complete with mini-makeovers and gift bags of Gen Alpha-approved brands. Store employees report ten-year-olds testing every acid, leaving with ‘tomato red’ faces, and wielding parent-funded American Express cards, while limited-edition lip balm drops are creating sneaker-release levels of frenzy. The opportunity continues attracting new players. 13-year-old Coco Granderson just launched Yes Day with Ron Robinson (Rhode formulator), betting that skin-safe ingredients and appealing packaging can succeed without influencer backing or Sephora distribution. According to NielsenIQ, nearly half of recent mass skincare growth is driven by this cohort, and Gen Alpha’s spending power projected to reach $5.5 trillion by 2029. What’s striking isn’t just the scale, but the sophistication of the customers. These kids aren’t just playing with makeup, but consumers who parrot influencer terminology and demand specific ingredients, forcing brands to serve them like any other discerning customer segment.
The Toy Company Outpacing NVIDIA (Jaclyn Peiser, The Washington Post): Build-a-Bear’s stock has surged more than 2,000% over the past five years, outpacing tech giants like Nvidia and Palantir by selling stuffed animals in declining malls while navigating 30% China tariffs. A $100 investment in January 2021 would be worth $1,600 today, $200 more than Nvidia. The company just posted its fifth consecutive year of record profits, with Q2 revenue up 11% to $124.2 million, and plans to open 60 new stores in 2025, a bold move as competitors shutter locations. The kicker: adults now drive 40% of revenue, paying $50+ for limited-edition Pokemon bears and NFL-themed plushies. What makes Build-a-Bear instructive is how well they’ve mastered the in-store experience. From the heart ceremony to the naming ritual and birth certificate, CEO Sharon Price John calls these “marking moments in time,” and they can’t be replicated by Amazon or copied by competitors. The company has also cracked the code on invisible price increases. By constantly rotating limited editions tied to trending IP - 75 brand partnerships from Disney to Formula One - Build-a-Bear resets consumer price anchoring with each drop, without triggering sticker shock. In an era of endless digital choice, physical retail thrives when it sells something that can’t be delivered in a box, whether that’s memories, community, or an experience.

Quick Hits:

VF Corporation is selling Dickies to brand aggregator Bluestar Alliance for $600 million in cash, part of CEO Bracken Darrell’s ongoing strategy to simplify the portfolio and focus on outdoor and lifestyle brands like The North Face, Vans, and Timberland. It’s yet another corporate divestiture we’ve seen so far this year.
MrBeast’s Beast Industries generated $450 million in revenue last year (including $200 million from Feastables chocolate) and now commands a $5.2 billion valuation with 450 employees, yet posted a $60 million loss.
Ben Stiller has launched Stiller’s Soda with beverage entrepreneur Alexander Doman (founder of Avec Drinks.) The brand debuts with three flavors - root beer, lemon lime, and Shirley Temple - each with 30 calories and 7g of sugar, using a blend of real sugar and stevia. Positioned between traditional sodas and functional beverages like Poppi and Olipop, Stiller’s targets consumers seeking a healthier soda option without added prebiotics or functional ingredients.
Despite Lululemon’s well-documented struggles with slowing growth, the brand stumbled into an unlikely hit with their $48 Dual Pouch Wristlet, which has garnered 4,500+ five-star reviews and inspired corporate knockoffs.
ChatGPT is now 20% of Walmart’s referral traffic, though referral clicks still account for less than 5% of total site visits.

Disclaimers:

The information contained herein is intended for informational purposes only and is not investment advice. Under no circumstances should any material herein be used or considered as an offer to sell or a solicitation of any offer to buy an interest in any securities or investment fund. Any specific company listed or discussed is not intended to be representative of any or all companies purchased, sold or recommended for TSG’s advisory clients. One should not assume that all investments in the companies identified are or will be profitable. The information provided is believed to be from reliable sources but no liability is accepted for any inaccuracies and TSG is not responsible for the accuracy or continued accuracy of third-party or hyperlinked content. TSG, its affiliates, and its employees make no representation that they have independently verified this information, and its accuracy and completeness cannot be guaranteed.

]]>