Join Ground

5.3% of sends produce 41% of email revenue - that’s Klaviyo’s own benchmark. The brands beating it aren’t writing better emails. They’re supplying better signals. Here’s how AI agents pipe them in from Shopify and your CRM.

Watch what happens when email revenue flatlines. The playbook is identical at every brand: crank the send volume, punch up the subject lines, bump the discount from 10% to 15%.

A quarter later the chart looks the same, and someone in the room declares the channel maxed out.

It isn’t - not even close. The median Klaviyo brand pulls 27% of total revenue through email. The best operators run at 30–40% across email and SMS. And the distance between those numbers has surprisingly little to do with what you put into Klaviyo. It’s determined almost entirely by what Klaviyo knows at the moment a send fires.

The channel isn’t tapped out. Your CRM is under-informed. And an under-informed CRM is a fixable one.

Klaviyo Already Told You Where the Money Is. Almost Nobody Listened.

Ask most growth leads what Klaviyo does and the answer amounts to “we send email there, then we check the numbers.” A dashboard wearing a send button. Glanced at on Mondays, actually managed never.

Now the stat that should rewire how you think about the channel. Klaviyo’s 2026 benchmarks, drawn from 183,000+ customers:

• Flows account for ~41% of all email revenue on a mere 5.3% of send volume

• Per recipient, flows earn ~18x what campaigns earn

• 48% of flow revenue comes from first-time buyers - campaigns manage 16%

Sit with that split for a second, because it ends the debate. Campaigns shout at everyone. Flows close individuals. And before a flow can do anything at all, it has to do one thing first: hear a signal.

Which means the revenue ceiling inside Klaviyo was never about creative output or send volume. It’s a signal supply problem. You have plenty to say. Your CRM has too little to go on.

Strip it down and your CRM is missing exactly three pieces of knowledge: who’s on the site, what they came for, and what they’d buy next.

Signal Gap One: The Welcome Offer Doesn’t Know Who It’s Interrupting

The welcome offer is the most critical entry point in any e-commerce funnel. It’s the digital equivalent of walking into a physical store and being greeted by a sales associate:

“Hey - did you know you can get 15% off your first purchase today?”

Yet the average welcome offer converts under 10% of visitors - meaning 90% of high-intent shoppers never opt in. In e-commerce, once they move on, they’re gone. There’s no second chance to intercept, assist, or close.

🔌 This is exactly the job Greet AI was built for. It reads 370M+ behavioral signals a month to predict purchase intent - like a best-in-class sales associate - and times the welcome offer for the moment a shopper is most likely to convert.

By firing at peak intent, Greet AI captures demand that static, one-size-fits-all offers miss, turning anonymous traffic into first-time customers and measurable revenue. Greet AI plugs into your CRM in 15 minutes, with no lift from your engineering team.

Book a Demo

Signal Gap Two: The Best Customers You’ve Never Met

Somewhere between 70% and 90% of your traffic is anonymous. That stat gets quoted everywhere. What almost never gets explained is the machinery under it - and the machinery is where the money is.

When an unknown visitor moves through your store, Klaviyo does track her - into her own browser’s local storage, where the record sits for up to 14 days, going nowhere. It only reaches Klaviyo after she identifies herself. If she eventually does, the backfill can trigger flows retroactively - assuming she hasn’t already sailed past their time delays. If she never returns to volunteer an email within the window, the whole record evaporates.

So the shopper who pulled up the same ring four times, read the reviews, opened the shipping page and left isn’t just unidentified. She’s inert, at exactly the moment she was persuadable. Nothing fires while she’s anonymous, and the entire rescue plan is hope she comes back and fills out a form.

Her intent was captured. It lives on her laptop. It’s earning you nothing.

🔌 This is the job ReCartify was built for. It resolves identity - converting anonymous sessions into profiles your existing flows can actually address, instead of gambling on a form fill inside a 14-day countdown. At Heaven Mayhem it produced a 30% lift in recovery, and 43% of new CRM signups came from visitors who had been anonymous - people their Klaviyo could never have named on its own.

That isn’t a smarter abandoned-cart email. It’s the exact same email, finally reaching the 43% who were invisible to it.

Book a Demo

Signal Gap Three: Your Customers Are Telling You What They'll Buy Next - And When. You're Not Listening.

AI can model each customer’s likely depletion date from their specific purchase history and order quantity - and trigger a replenishment flow in the exact window they’re running low, not on a fixed 30-day calendar that ignores how much they actually bought. The cost of getting this timing wrong isn’t a missed email. It’s a customer who reordered from a competitor two weeks earlier.

🔌 This is the job ReBeat was built for. Per-customer prediction of the next purchase and its timing - an individual’s predicted next-order date, not a segment average she happens to fall into. Adapt Naturals used it to push subscription conversion up 22% and nearly double subscription revenue, by showing up when the behavior said ready instead of when the calendar said send.

Book a Demo

Enriching Your Klaviyo With Customer Data Can Go a Long Way

This is where revenue agents change the model. They collect customer data across the full funnel - from the first anonymous visit to the day someone becomes a loyal customer - and continuously act on the moments that move them to the next step.

Ground’s suite of AI revenue agents were built for this purpose. Greet AI, our acquisition agent acts before the first order, helping brands drive 30-50% more first-time revenue from existing traffic. ReCartify, our revenue recovery agent, resolves identity so the models understand who is engaging and abandoning. ReBeat AI, our retention agent, driving 2-4x repeat purchase revenue. Terra, our new product, connects the intelligence and execution across the journey.

These models do not operate independently. They work as one revenue brain.

Book a Demo

Join Ground

If this is the first time you’re hearing about AI revenue agents, September is the right month to catch up - and act fast.

Revenue agents are AI agents with one job: grow revenue across your funnel. They plug into your Shopify, your Klaviyo, your ad accounts, and read what your visitors are already telling you they want. Because they are telling you - with every click, hesitation, and return visit. Most stacks just never capture that intent. The agent does, and then it triggers your own native comms at the moment the signal fires, turning intent you were blind to into purchases.

September matters because revenue agents run on data, and data takes time. Every week an agent is live, it gets better at recognizing the patterns behind why your customers buy.

The September advantage: put acquisition agents to work catching shoppers who are already planning their holiday purchases

According to Attentive’s 2026 shopper survey, 46% of BFCM shoppers start buying before November even starts, and 62% plan their holiday purchases before the month begins.

Here’s the problem that keeps happening: 95% of your site traffic is anonymous and never opts into your welcome offer. You don’t know who’s checking out your website, you don’t know who’s planning their holiday purchases, and you can’t send them your welcome flows.

Which means the real Q4 question isn’t the one in your planning doc right now. It isn’t “what’s our offer?” It’s this: on November 1st, how many of your visitors will your stack actually recognize - and how much will it know about what makes them buy?

The ceiling is set before the season starts

What if you could lift identification from 5% to 30% or more? AI revenue agents - acquisition agents specifically - solve for that. They read how each visitor behaves: what they look at, where they hesitate, which ad brought them in. Then they show the right welcome offer at the exact moment that visitor is deciding. They can even mirror the ad the visitor clicked, so your site picks up the conversation the ad started instead of restarting it.

🔌 This is exactly why we built Greet AI. It’s an edge AI model that processes 370+ million behavioral signals every month, reading the pattern behind the clicks rather than the clicks themselves. The pause on a SKU right after someone finishes the product description. The time spent hovering over reviews. The sequence of pages they move through. None of that means anything to your site. To Greet AI, it’s the difference between a visitor who isn’t ready and one who’s warmed up and ready to buy.

After integrating Greet AI into their stack, Heaven Mayhem saw a 49% incremental new customer lift, and 43% of their new CRM signups came from visitors who would have left anonymous. Book a demo to see Greet AI in action.

Book a Demo of Greet AI

Why September is the best month to have revenue agents learning, testing, and optimizing before the rush

September traffic is clean. Signals aren’t distorted by promo chaos. The patterns your AI learns this month are the durable ones - what your customers actually respond to at full price, which is precisely the intelligence you’ll spend against when everything goes on sale.

And the learning compounds into the peak. A suite of revenue agents installed now enters Cyber Week with two months of your data behind every decision. A system installed in November enters cold, at the exact moment your traffic is most expensive and your team has the least bandwidth to notice what’s breaking.

That’s the flywheel, and it only spins one direction:

More conversions → more revenue → more data on what works → smarter agents → more conversions.

Your ads can’t compound like that. Your discounts definitely can’t. A discount you ran last summer taught you nothing. An agent that ran last summer made you money and left you smarter.

Book a Demo - Ground's Agents

What AI revenue agents actually learn

“Collecting data” undersells it. Here’s what a revenue-grade intelligence layer accumulates between now and November:

What makes a visitor buy the first time. Which entry point, which category, which moment of hesitation - and which intervention resolves it. Adobe’s read on a trillion US retail visits found AI-referred traffic converting with 37% higher revenue per visit than the average session. High-intent traffic is coming. The question is whether your site responds to that intent or serves everyone the same 10%-off offer.

What makes a buyer buy again. Replenishment cycles, cross-sell affinities, the browsing shift that signals someone is ready for a second category. This is the data most stacks throw away in November, treating BFCM buyers as revenue instead of as a cohort. The brands that lose Q4 in January are the ones whose stack learned nothing from their biggest month.

What your best customers have in common before they become your best customers. By the peak, the system isn’t guessing which of your 100,000 visitors is worth a VIP experience. It’s pattern-matching against months of evidence.

That’s the goldmine. Not a discount engine. A memory your revenue stack didn’t have last year.

🔌 This is exactly what Ground’s suite of revenue agents does. They plug into your Shopify, Klaviyo, ad accounts, SMS platforms, and more. Greet AI converts first-time visitors. ReCartify recognizes the shoppers your stack forgot. ReBeat brings customers back before they drift. They run as clearly labeled flows inside your own account - your data, your relationship, your customer. Setup takes about 15 minutes, and brands typically see 30–50% more first-time revenue without having to manage another tool.

Book a Demo

Final Word

Every brand in your category will enter the same four weeks with roughly the same discounts, the same channels, and more expensive versions of the same traffic. The only durable edge left is how much your stack knows about your customers when the surge hits - and knowledge has a lead time. The offer can be decided in November. The intelligence can’t.

September isn’t prep season. It’s the part of BFCM where the outcome is still negotiable.

Curated Intelligence

• Salesforce: 2026 Holiday Predictions: 20% of holiday ecommerce traffic is expected to come from AI agents, and consumer reliance on AI assistants grew 200% in a single year. Discovery is being rewritten mid-season.

• Attentive: 2026 BFCM Shopper Report: 46% of shoppers start before November; 67% have used AI chatbots for shopping in the past three months. Your “early access” list is now your main list.

• Adobe via Yahoo Finance: AI traffic to US retailers up 393% in Q1: and it’s outspending human-referred traffic per visit. Small share, outsized quality.

• NovaData: AI shoppers arrive ready to buy but rarely close: high-intent AI-referred sessions are converting poorly on sites that can’t respond to that intent. The gap is the opportunity.

Join Ground

Common Ground is Ground’s newsletter highlighting the most important consumer news, tech trends, artificial intelligence, in-depth analysis, along with a few extra thoughts from our team. Subscribe today to get the latest issue delivered to your inbox.

This issue is brought to you by Sheida Mirjahani, Growth Lead at Ground - a suite of AI revenue agents for commerce, driving incremental top-line revenue for 200+ consumer brands.

All week, my LinkedIn feed kept circling the same theme: brands are planning to convert more from their CRM marketing. Which makes sense, BFCM is a quarter away, and you really have find out what’s working and double down on it or fine tune the leaks.

So I wanted to combine what operators are saying out loud on Linkedin with what we’re seeing from our customers portfolio at Ground: how AI is turning lifecycle marketing into a sustainable growth engine.

What we’ve seen move the needle in 2026 comes down to optimizing three things in your lifecycle marketing:

• Use what customers are telling you they want to buy next by capturing their on-site browsing data

• Timing the next touchpoint based on what they’re already signaling you + what’s the common pattern across all your CRM data

• Matching the offer to where someone actually is in their buying cycle

AI agents are making all these three inputs became knowable per customer, and it’s what I want to unpack in this issue. Before diving in, I’d like to share this post which really resonated with me this week:

Optimization #1: AI Agents Are Watching Who’s On Your Site And Why

Between 70% and 90% of your traffic is anonymous - and your CRM can’t send to a person it can’t identify. The shopper who viewed the same ring four times, read the reviews, and checked the shipping page isn’t in your abandonment flow. She isn’t in any flow. The best-built automation in your account has no one to fire at.

That’s not a creative problem or a segmentation problem. It’s an identity problem, and until this year the accepted answer was a popup and a prayer.

🔌 This is actually a problem one of our AI agents solves. At Ground, we’ve built a suite of AI revenue agents that integrate with your existing DTC stack - Shopify, Klaviyo - and generate the revenue it’s currently leaving behind. One of them, ReCartify, is our abandonment recovery agent. It resolves identity - turning anonymous sessions into profiles your existing flows can actually reach. At Heaven Mayhem, the modern jewelry brand worn by Hailey Bieber and other celebrities, it drove a 30% lift in recovery revenue, and 43% of their new CRM signups came from visitors who had been anonymous - people their CRM, on its own, would never have been able to name.

Book a Demo

Optimization #2: AI Agents Are Now Passing The Buying Intent To Your Flows In Real Time

The thing that moves the needle - and now is the time to add it - is an intelligence layer on top of your stack that watches when a customer comes back and what they were looking for. Then getting trained on your CRM data to find the patterns behind when people are actually ready to buy again, subscribe, or replenish. Then acting on its own to send them the email or sms they should get, when the signal shows up.

🔌 This is what ReBeat, our retention agent, is for. Per-customer prediction of what someone buys next and when - a predicted next order for that individual, not an average applied to a segment they happen to sit in. Adapt Naturals used it to lift subscription conversion 22% and almost double subscription revenue, by making the offer the data said each customer was ready for.

Book a Demo of ReBeat AI

Optimization #3: AI Agents Know The Best Time to Trigger Each Flow

The 30/60/90-day win-back is a calendar guess wearing a lifecycle strategy’s clothes. Nobody’s repurchase cycle is a round number - and the gap between the calendar’s guess and the customer’s actual rhythm is where repeat revenue quietly dies.

Tyler Neely, our very own customer operations lead, puts it plainly:

None of that requires a bigger team. It requires a CRM that knows the moment.

🔌 This is what per-customer timing does at portfolio scale. Across the brands running Ground’s agents, letting the data pick the moment produces 2–4x higher repeat purchase revenue - and with Greet AI, our on-site acquisition agent, reading intent in real time and greeting customers when they’re ready with a welcome offer, is why our brands see 30–50% more first-time customer revenue on traffic they already paid for.

➡️ See what your flows would earn with per-customer timing - Book a demo with our team.

Book a Demo

Curated Knowledge

• Klaviyo ships its first agents. Composer and Customer Agent - the platform layer is racing toward the same conclusion: CRM is becoming a system that decides, not just sends. Klaviyo, Jun 2026. Read More

• The 2026 email benchmarks, in full. Flows drive ~41% of email revenue from 5.3% of sends - the strongest public evidence that triggered beats scheduled. Klaviyo. Read More

• How Klaviyo handles anonymous visitor activity. The local-storage mechanic that decides whether your flows ever meet the anonymous majority. Klaviyo Help Center. Read More

🔌 Ground’s AI’s suite of revenue agents sit inside every growth channel your brand operates on -paid ads, email, SMS, Shopify, etc - and make decisions 24/7 about which levers to push and pull to drive more revenue.

Brands using Ground see 30–50% more first-time revenue and 2–4x higher repeat purchase revenue. Setup takes 15 minutes with no time taken from your engineering team.

Book a Demo

Join Ground

Black Friday is fourteen weeks out, and the standard playbook is already being written in a thousand planning docs: raise the ad budget, deepen the discount, send more email.

CPMs spike every November because every brand in your category floods the same auctions in the same four weeks - you’ll pay more per click in Q4 than at any other point in the year. So the brands that win BFCM aren’t the ones that buy the most traffic. They’re the ones that convert the most of the traffic they were already going to buy.

If I were running retention for a DTC brand right now, this is exactly what I’d do in order:

Project #1: Fix the Experience Between the Ad and the Landing Page

Start where the most expensive dollars are. If you’re increasing paid spend for BFCM - and nearly every brand does - the first place your money leaks is the three seconds after the click.

Someone clicks a piece of creative built around a specific need. A specific product. A specific color. A specific promise. And then they land on your site and hit a generic welcome offer that has nothing to do with what brought them there. The ad said one thing; the site says another. The thread breaks, and so does the conversion.

That break gets more expensive every day. During BFCM, when you’re paying peak auction prices for every visitor, it’s the single costliest gap in your funnel.

This is where AI agents are quietly pulling CAC back down. Connected directly to your ad platforms, your CRM, and your storefront, an agent can read which creative brought each visitor in - and reshape the on-site moment to match it. Same offer, same language, same promise. Then it carries that thread into the welcome flow that follows. More people opt in, and more of the ones who do actually convert, because the experience never stopped being the one the ad sold them.

🔌 This is exactly why we built Greet Halo. It connects to your ad platforms, reads the creative behind each click, and mirrors it in the on-site moment and through the welcome flow after it. Brands running it are seeing a 10% lift in first-time customer revenue. See Greet Halo run on your own ad account →

Book a Demo

Project #2: Audit Your Flows

Most brands haven’t looked hard at their flows since they were built. If I had to bet on what an honest audit would find in the median Klaviyo account fourteen weeks before Black Friday, it would be some mix of: segmentation that hasn’t been rethought since the list was a third of its current size, a welcome flow that treats a $200 ad-click visitor the same as an organic browser, and at least two revenue-critical flows - browse abandonment, post-purchase, winback - that simply don’t exist.

🔌 We built a pack of Claude skills that run this audit for you. Add them to your Claude, connect your Klaviyo, and it runs a full audit of your flows, segmentation, and structure - and tells you exactly what’s missing before Q4 traffic finds out for you. It’s free. Download the Retention Claude Pack →

Project #3: Install the Thing That Sells While You Sleep

What moves the needle, and the time to act on is now, is adding an intelligence layer that analyzes your customer data, finds the patterns behind what makes each buyer buy next - or when they’re ready to opt into a subscription or replenish - and then acts on its own when it gets the right signal. Not a dashboard that tells you what happened. An agent that sells more of your inventory while you’re running the rest of Q4.

🔌 This is exactly why we built ReBeat AI. It connects to your stack - Shopify, ads, your email and SMS system, your popup tool - reads purchase history alongside real-time browsing, and pulls the lever on its own when the signal is there. If someone’s due for a replenishment but suddenly browsing something else entirely, it catches the shift, reads the intent behind it, and surfaces the right product at the moment the desire is actually there. Replenishment, cross-sell, one-time-buyer-to-subscription - driven by what the customer is telling you right now, not by a calendar.

Book a Demo

Customer Story: Ground helped Chief Nutrition triple (3x) their DTC revenue!

Chief Nutrition is a health-focused brand making clean, whole-food snacks - beef jerky, protein bars, the kind of products people reorder when the timing is right. In under a year running Ground’s revenue agents, they tripled their DTC revenue. The receipts:

• 12x in repeat purchases, while growing subscriptions

• 15x ROI

Justin Babet, Co-founder & COO: “Ground AI is a great product! This is how all AI products should be - where it actually GROWS your business for you. It was a super easy onboarding and in 15 minutes Ground was up and running doing what it does best... lifting sales.”

Notice what’s in that quote and what isn’t. No replatform. No new headcount. No six-month implementation. Fifteen minutes of onboarding, and the acquisition side and the repeat side started compounding on each other - new customers in, repeat purchases up 12x behind them.

🔌 Ground AI drives up to 50% more email and SMS revenue with zero effort - scanning your site for revenue-generating opportunities from first-time visitor all the way to repeat buyer, and acting on them on its own. Get a BFCM-readiness read on your own store → Book your audit

This issue comes from Sheida Mirjahani, Growth Lead at Ground.

Join Ground

If I were running an ecommerce brand today, I would have asked myself:

• Which first purchase creates my highest-value customers?

• Which product opens the door to the most cross-sells?

• Do customers who came in on a big discount actually stick around?

• Which one-time buyers are most likely to become subscribers?

• Which products bring people back again and again?

Simple questions that every founder has asked them but almost nobody can answer them.

And there’s a new way to reach to these answers: Deploy custom revenue agents (yes, they exist, and yes, leading consumer brands are already generating extra revenue with them! - more on that below) that both generate you revenue and capture you the customer data you need.

What a revenue agent actually is

Most people hear AI agents for b2c and immediately think chatbots. A widget in the corner of your website answering "where's my order?"

I get it. That's what the market has been trained to expect.

But there’s another category of agents called revenue agents that are quietly getting more and more popular. Read how VIOLETTE_FR or RoC Skincare generated at least extra 10% DTC revenue with Ground’s revenue agents.

But what are revenue agents exactly?

A revenue agent is an AI with one job: make you more money. It sits inside the channels you already run - your site, your email, your SMS - and makes small decisions all day long. Who should see a welcome offer, and when. Who’s about to leave. Who’s ready to buy again. Which message to send, and which to hold.

Think about what a great growth team does all week. They look at what’s converting and adjust. They change email triggers based on how customers behave. They time offers so margins don’t bleed. Now imagine that rhythm running every hour of every day, without waiting for Monday’s standup to decide what changes. Those are revenue agents jobs.

They also bring 3 core value to your business, beside generating additional revenue:

AI Revenue Agents - Value #1: higher LTV

We at Ground have built a suite of AI agents that are running inside leading consumer brands and are generating them revenue.

Take ReBeat, our retention agent. It watches every customer after the first purchase - what they bought, how they browse when they come back, and what customers like them usually want next. Then it sends the right message at the right moment, instead of a “we miss you” email on day 45 because day 45 is what someone set in 2023.

The difference sounds small. It isn’t. Calendar-based winbacks talk to everyone the same way. Prediction-based winbacks talk to each customer when that customer is ready. Brands using Ground see 2–4x more repeat purchase revenue -not because the emails got prettier, but because the timing finally matched the person.

More second purchases, faster. More subscribers from one-time buyers. That’s LTV going up.

Book a Demo

AI Revenue Agents - Value #2: Lower CAC

Here’s the thing about acquisition costs: you don’t have to lower your ad prices to lower your CAC. You just have to waste fewer of the visitors you already paid for.

Most brands convert 1–2% of their traffic. The other 98% leaves - and most leave without ever telling you who they are. You paid for every one of them.

Our other revenue agent, Greet AI, works that problem. It reads how each visitor behaves - what they look at, where they hesitate, which ad brought them in - and shows the right welcome offer at the exact moment that visitor is deciding. It even mirrors the ad they clicked, so the site picks up the conversation the ad started instead of restarting it.

Same traffic. Same ad spend. More customers out the other end. When more of your paid clicks turn into buyers, the math is simple: your cost per customer drops without touching your ad account.

If you want to see what that looks like on your own traffic, book a demo - we’ll show you the visitors your current setup is missing.

AI Revenue Agents - Value #3: You Get More Customer Data

Every decision an agent makes creates a data point you didn’t have. Every visitor Greet identifies is a name, an email, and a story - which ad they came from, what they browsed, what finally convinced them. Every prediction ReBeat makes - right or wrong - teaches you something about how your customers actually behave.

Go back to those five questions at the top. Which first purchase creates your best customers? After a few months of agents running, you can just look. The answer is sitting in your own CRM, attached to real people, because the agents were collecting it while they were converting.

That’s the flywheel, and it only spins one direction:

More conversions → more revenue → more data on what works → smarter agents → more conversions.

Your ads can’t compound like that. Your discounts definitely can’t. A discount you ran last summer taught you nothing. An agent that ran last summer made you money and left you smarter.

Book a Demo of Ground Agents

🔌 Where Ground fits

Ground’s agents plug into the Shopify, Klaviyo, ad accounts, sms platforms, etc. Greet AI converts first-time visitors. ReCartify recognizes the shoppers your stack forgot. ReBeat brings customers back before they drift. They run as clearly labeled flows inside your own account - your data, your relationship, your customer. Setup takes about 15 minutes, and brands typically see 30–50% more first-time revenue without having to manage another tool.

Book a demo - see the flywheel running on brands like yours

💡 Prompt Smarter

Want a head start on those five questions with the data you already have? Export your last 12 months of orders and paste this into Claude:

“Group my customers by their first product purchased. For each group, show me: average number of repeat orders, average total spend, and the % who bought a second time within 90 days. Rank the groups. Which first purchase creates my most valuable customers?”

Finding Common Ground, The Ground Team

If this changed how you think about what your tools should be doing for you, share it with a founder who’s still buying traffic to make up for the customers they can’t see.

Join Ground

When CRM revenue plateaus, every brand reaches for the same three levers. Send more. Rewrite the subject line. Raise the discount from 10% to 15%.

Then the next quarter looks like the last one, and the conclusion is that email is tapped out.

It isn’t. The median brand on Klaviyo drives 27% of total revenue from email. Best-in-class brands sit at 30–40% across email and SMS. The gap between those two numbers has almost nothing to do with what you’re putting into Klaviyo, and almost everything to do with what Klaviyo can see when the send goes out.

Your CRM isn’t underperforming. It’s under-informed.

Stop Managing Klaviyo Like a Report. Start Managing It Like a P&L Line.

Ask a growth lead what Klaviyo is for and you’ll usually get some version of “it’s where we send email and check how it did.” A dashboard with a send button. Reviewed on Monday, optimized never.

I was reading Klaviyo’s own 2026 benchmarks and one number caught my attention: Flows generate ~41% of total email revenue from just 5.3% of sends

And when you compare that to how campaign broadcasts perform, you see that Flows are responsible for sustainable revenue and campaigns are responsible for awareness, making excitement and time-to-time revenue spikes. Flows do their best jobs when they are told when, why, and who to reach out to.

Which means revenue generation inside Klaviyo isn’t a content problem or a volume problem. It’s a signal supply problem. You are not short of creative. You are short of things your CRM knows.

There are exactly three things it doesn’t know: who showed up, why they’re here, and what they’ll buy next.

🔌 This is why we built Ground. Ground is a suite of three AI revenue agents that capture the on-site signals your CRM never sees - who showed up and why - and read purchase behavior across your entire customer base to learn what similar customers buy, why, and when. Together, they turn Klaviyo from a send tool into a sustainable revenue channel.

Book a Demo

Blind Spot One: You Don’t Know Who Showed Up

Between 70% and 90% of your site traffic is anonymous. Everyone knows that stat. Almost nobody knows the mechanic underneath it, and the mechanic is the interesting part.

When an unidentified visitor browses your store, Klaviyo does record what she does - into her browser’s local storage, where it sits for up to 14 days, waiting. It’s only sent to Klaviyo after she identifies herself. If she does, the backfill can retroactively trigger flows, provided she hasn’t already blown past their time delays. If she never comes back and hands over an email inside that window, the record is discarded.

So the shopper who viewed the same ring four times, read the reviews, checked the shipping page and left is not merely unknown to you. In the moment that matters, she’s inert. Nothing fires while she’s anonymous, and the entire recovery depends on her volunteering an identity later, unprompted.

The evidence of her intent exists. It’s on her laptop. It is doing nothing for your business.

🔌 This is what ReCartify, our abandonment agent, is for. It resolves identity by analyzing millions of first-party data in a month and turning anonymous sessions into profiles your existing flows can actually reach. At Heaven Mayhem it drove a 30% lift in recovery.

Book a Demo

Blind Spot Two: You Greeted Them Before They Were Ready

There’s something wrong about most welcome offer setups: they fire on a timer. Someone lands, a countdown starts, and five to ten seconds later the popup appears - regardless of who that person is or what they’re doing.

And 95% of people close it and this is a timing problem.

People land in browsing mode. They’re orienting - checking prices, scanning products, deciding whether your brand is even relevant to them. They are not ready to hand over an email address to a brand they met four seconds ago.

🔌 This is exactly why we built Greet AI. It’s an edge AI model that processes 370+ million behavioral signals every month to understand when the customer is actually locked-in to buy and greets them with a welcome offer. As an example of using Greet AI, Ghia saw a 42% lift in first-time customer revenue through CRM within 30 days.

Book a Demo for Greet AI

Blind Spot Three: You Don’t Know What They’ll Buy Next

What usually happens today is that brands miss the signals their customers are giving them. They miss that someone came back to the site, ready to buy a product they have never tried, or came back to replenish their product sooner than you expected. This is a high-intent signal for retention flows that your Klaviyo is not seeing.

🔌 This is what ReBeat is for. Per-customer prediction of what someone buys next and when - a predicted next-order date for that individual, rather than an average applied to a segment they happen to sit in. Adapt Naturals used it to lift subscription conversion rates 22%, and almost doubled their subscription revenue, by reaching people at the moment the behavior said they were ready to commit rather than on the day the calendar said to ask.

Across Ground’s brands, ReBeat produces 2–4x higher repeat purchase revenue.

Book a Demo - ReBeat AI

Final words

The constraint on your CRM in 2026 is not the quality of the intelligence sitting on top of it. It’s the quality of the data underneath it. Ground’s three models sit between Shopify and Klaviyo and fix the second problem - which is what makes everything built on top of it work.

The beauty of Ground is that you don’t replatform, rebuild your flows, or retrain the team. The flows you already have start firing on evidence instead of on timers.

Learn about Ground

Curated Intelligence

• Heaven Mayhem is using revenue agents that lift CRM revenue on their own. Watch this video to learn how.

• The 2026 email benchmarks, in full. The flows-versus-campaigns split is the most under-discussed number in retention. Klaviyo. Read More

Common Ground is Ground’s newsletter highlighting the most important consumer news, tech trends, artificial intelligence, in-depth analysis, along with a few extra thoughts from our team. Subscribe today to get the latest issue delivered to your inbox.

Join Ground

You already paid for every visitor who lands on your store. More than 90% of them leave without telling you who they are - and the moment they walk, every future touchpoint you could have had with them walks too. It’s an identification problem, and it’s the cheapest revenue in your business.

The Metric You’re Optimizing Is the Wrong One

Most brands track popup conversion rate. It’s the wrong number. The one that decides your quarter is visitor identification rate - of everyone who lands on your store, how many do you actually know?

The math is unsentimental. Take a brand with 100,000 new visitors a month, a 2% site conversion rate, and a $100 AOV. That’s $200,000 in first-time revenue from anonymous traffic converting on its own. Now move identification from 5% to 15% - a range brands hit through better timing and matched creative, not more spend. Those visitors enter your welcome flows, which convert at a meaningfully higher rate than anonymous traffic, and the shift alone generates 30-50% more first-time customer revenue.

No new ad spend. Same traffic. Same site.

95% of People Close Your Popup. That’s a Timing Problem, Not an Offer Problem.

Here’s the honest read on most welcome offer setups: they fire on a timer. Someone lands, a countdown starts, and five to ten seconds later the popup appears - regardless of who that person is or what they’re doing.

And 95% of people close it.

That isn’t a verdict on your offer. It’s a verdict on your timing. People land in browsing mode. They’re orienting - checking prices, scanning products, deciding whether your brand is even relevant to them. They are not ready to hand over an email address to a brand they met four seconds ago.

The brands winning this aren’t firing on a timer. They’re firing on intent. And to be clear: basic clickstream logic - click a product, scroll halfway, trigger a popup - has been around for over a decade. It’s not novel and it’s not enough.

🔌 This is exactly why we built Greet AI. It’s an edge AI model that processes 370+ million behavioral signals every month, reading the pattern behind the clicks rather than the clicks themselves. The pause on a SKU right after someone finishes the product description. Time spent hovering over reviews. The sequence of pages they move through. None of that means anything to your site. To Greet AI, it’s the difference between a visitor who isn’t ready and one who’s warmed up and ready to buy.

Heaven Mayhem, the cult accessories brand worn by Hailey Bieber and Sydney Sweeney, was already running a strong operation - their welcome popup was live, it just fired on entry, before most shoppers had made up their mind. High-intent visitors dismissed it and kept browsing. With Greet AI reading behavior instead of a countdown, they saw a 49% incremental new customer lift, and 43% of their new CRM signups came from visitors who would have left anonymous. Book a demo to see Greet AI in action.

Book a Demo

The Creative That Earned the Click Should Be the Creative That Greets Them

Someone clicks a Meta ad for your SPF moisturizer, lands on your site, and gets a popup about your entire skincare collection. You just broke the thread. The creative that earned the click is the strongest signal you have about what that person wants, and most brands throw it away at the door.

🔌 That’s what Greet Halo does. It connects to your ad platforms and layers your best-performing Meta, Google, and TikTok creative into the welcome offer, so the shopper sees a continuation rather than a non-sequitur. Brands running Greet AI and Halo together see a 30-70% lift in sign-up conversion, and about 20% more email and phone signups across our portfolio. Book a call to see Greet Halo.

Book a Demo of Greet Halo

Revenue Per Signup Is the Metric You Should Track

A popup producing 1,000 new subscribers a month means nothing if two of them buy. The number that tells you whether your capture strategy is working is how much revenue each new subscriber generates in their first 30 days. Most brands don’t track it. They optimize for subscriber count, then wonder why the revenue doesn’t follow.

Ghia, the non-alcoholic apéritif brand, had a welcome series in place. The flow wasn’t the problem - who was entering it was. High-intent visitors were dismissing the timed popup before they ever opted in. Once Greet AI started engaging shoppers at the moment their behavior signaled genuine readiness, more qualified subscribers entered the flow and the revenue followed: a 42% incremental welcome series lift and 3.5x ROI since going live.

As their Director of Growth Marketing & E-Commerce, Hanna Pandya, put it: “Most ‘growth tools’ create more work. Ground does the opposite. It captures missed revenue automatically. It’s how AI should be.”

You don’t have a traffic problem. You have an identification problem - and you’re already paying for it, every month, in a line item you’ve stopped questioning. Ground installs on Shopify in 15 minutes, works inside your existing Klaviyo stack, and your native popup still fires first. We catch shoppers it missed.

See Ground in Action

Curated intelligence

• How H&M, Chobani, UrbanStems and Teleties are refining their partnership strategies. Read more

• ShipBob Launches First Anthropic-Verified Fulfillment Connector, Anchoring its AI Suite. Read more

• Meta launched an AI-powered “Seller” app for Marketplace. Read more

Subscribe now

AI agents aren't a future concept anymore. They are now running inside hottest brands in commerce, acting as a new channel that generates up to 20% more revenue across the entire funnel autonomously.

Join Ground

AI Agents Are the New CRO: Improving Your Ad Conversion

AI agents are improving ad ROI by connecting directly to your ad platforms, your CRM, and your storefront - reading which creative brought each visitor in, and reshaping the on-site moment to match it: same offer, same language, same promise. Then they carry that thread into the welcome flow that follows. More people opt in, and more of the ones who do actually convert.

The problem before this was that the experience often breaks the moment someone clicks your ad and lands on your site for the first time. They click a piece of creative built around a specific need - a specific product, a specific color - and then hit a generic welcome offer that doesn’t match what brought them to the store in the first place.

CAC is rising every day, but AI agents are bringing it back down, because they know each customer at an individual level, can match their needs and their language, and convert more of them.

🔌 This is exactly why we built our new AI agent, Greet Halo. It connects directly to your ad platforms, reads which creative brought each visitor in, and mirrors that creative - the offer, the language, the promise - in the on-site moment and through the welcome flow that follows. More people opt in, and more of the ones who do actually convert.

Brands running it are seeing a 10% lift in first-time customer revenue. Book a demo to see Greet Halo in action.

Book a Demo

AI Agents Convert More First-Time Buyers into Repeat Buyers

AI agents can now understand when a customer comes back to check out your store, and what they’re looking at - is it a new product they’re interested in, or are they trying to replenish? - and use that to win them back or remind them they can now opt in to a replenishment subscription.

What usually happens today is that brands miss the signals their customers are giving them. They miss that someone came back to the site, or they’re running an old-school stack that can’t predict the optimal moment to reach a customer through their ESP or SMS, because the stack simply isn’t built for it.

🔌 This is exactly why we built our retention agent, ReBeat. Ground’s ReBeat predicts when an existing customer is likely to buy again and what they’ll reach for, rather than waiting on a rules-based timer that fires at day 60 or day 90 because someone picked a round number in a planning doc.

Fixed timers treat every customer as the average customer, so half your winbacks land too late to matter and the other half land before anyone’s ready - both feel like a brand talking at you rather than to you. ReBeat reads purchase history alongside real-time browsing, so if someone’s due for a replenishment but is suddenly browsing something else entirely, it catches the shift, interprets the intent behind it, and surfaces the right product at the moment the desire is actually there. That’s what drives replenishment, cross-sell, and the one-time-purchase-to-subscription conversion - not a calendar, but a read on what the customer is telling you right now.

Request a Demo

🗣️ From the Community

Nothing we write about what AI agents are now able to do lands as hard as a customer saying it themselves.

One of our customers Dan Castle, former Managing Director at Martha Stewart and now founder of CastleFlexx, just shared publicly what Ground has meant for their growth - and we’re still a little floored by it.

Dan Castle runs CastleFlexx - the premium mobility and back-pain device - with a small team and a handful of AI agents (Humbly: ours). By his own words, he’s scaling the brand faster than Vuori and Under Armour did at the same stage, and he calls Ground the first growth lever in the business. See what Ground can do for your brand

Join Ground

Finding Common Ground

The Ground community comes together sharing different stories, but common ground.

Learn More

Our very own Growth Strategist at Ground, Jacqueline, brings you the hottest news in commerce and what that means for growing brands today.

The Churn Behind Churn

We talk to hundreds of DTC brands, and this season one thing keeps coming up: the person who knew why everything worked has left.

You know the meeting. A new ecommerce director starts. Within three weeks there’s a vendor review, and the question in the room is “what does this one do for us?” Nobody can answer it - not because the vendor stopped working, but because the two people who chose it are gone. The wins are in a Q3 deck somebody archived. The reasoning behind them was never written down at all.

So the review does what reviews do without history. Flows get paused “just to look at them.” A test gets re-run that already failed last spring. The dashboard gets rebuilt, and last year’s best-performing segment quietly stops sending because it wasn’t in the new build. Four months later, revenue is down and nobody can point to the decision that did it.

Your customers didn’t churn. Your knowledge of them did.

And the calendar is unforgiving about it right now. A brand that changes ecommerce leads in September has one holiday season to get it right, with a new operator learning your customer base in real time - during the eight weeks that decide the year.

AI Concentrates the Risk

The teams doing this work are getting smaller. AI-driven restructuring has flattened ecommerce orgs, so fewer people now own more of what happens, and knowing how to use AI has moved into the job description for director and VP roles.

Beauty leaders are treating it as a daily habit rather than a project. Tarte runs 15-minute weekly check-ins on how the team is actually using AI. Beekman 1802 has called it non-negotiable - while being clear the point isn’t to shrink the team.

Here’s the part that should make you uncomfortable. When one strong operator can do the work of three, more of what’s working lives inside that one person: their prompts, their spreadsheets, their sense of which customer is worth chasing. Most of it sits in a personal ChatGPT account and a few Slack threads. None of it is yours.

AI hasn’t made great marketers less valuable. It’s made losing one significantly more expensive - we’ve watched brands spend a full quarter rebuilding what one person carried out the door.

Build a Brand That Remembers

The answer is not another folder of final decks. Brands need to preserve the reasoning behind the work.

Capture the signal that triggered a decision, the hypothesis that followed, the experiments that failed, and the evidence that earned a vendor its place in the stack. Move approved AI workflows out of personal accounts and into shared systems. Give every category and partner a value story a new stakeholder can understand without a six-week archaeology project.

This isn’t documentation for documentation’s sake. It’s a compounding learning system.

AI can retain captured signals, hypotheses, results, workflows, and decision logic. It cannot automatically retain taste, relationships, trust, or judgment. Those still belong to people.

The goal is not to make operators interchangeable. It is to make the brand’s learning cumulative. Ground’s revenue agents do these all automatically. Book a chat with us.

Request Demo

The Continuity Advantage

Operator churn is inevitable. Institutional amnesia is not.

The brands that win in the age of AI will not be the ones that never lose talented people. They will be the ones whose intelligence keeps compounding through every new hire, agency review, and reorg.

When your best operator leaves, will your brand lose a person, or everything that person taught the business?

🔌 This is where Ground’s AI Revenue Growth Engine matters. Greet AI learns how new visitors buy. ReCartify resolves who they are. ReBeat AI learns what brings them back. Terra connects that intelligence into one brain inside your existing Klaviyo & Shopify stack.

When the ecommerce leader or agency changes, the new operator inherits a running revenue system, not a blank dashboard. Ground does not replace your best operator. It prevents your revenue system from starting over when that operator leaves.

Book a Demo

Curated Intelligence

• BeautyMatter / Who Moved Where in Beauty, Q2 2026 - Beauty’s leadership reshuffles show how quickly teams and mandates can change.

• Glossy / How beauty executives are implementing AI - Tarte and Beekman 1802 are making AI fluency part of the operating cadence, not a side project.

• eCommerce Placement / The AI-native restructuring of ecommerce teams - Leaner team structures are concentrating more responsibility in fewer, higher-stakes operators.

• Fast Company / The hidden cost of employee turnover in the age of AI - Outputs are not enough; companies need to preserve the reasoning behind decisions.

💡 Prompt to Use

“Assume our E-commerce leader and agency are both replaced tomorrow. Using our approved campaign briefs, performance reports, decision logs, vendor evaluations, and shared AI workflows, identify: which decisions will likely be reopened, the evidence behind each one, undocumented logic, failed experiments we may repeat, and what the new team must understand in its first 30 days.”

If that prompt exposes more gaps than answers, book a revenue continuity audit. We will map where your revenue logic is trapped in people, reports, or disconnected systems before the next handoff tests it.

Audit Your Revenue System

Follow the Thinking Behind Common Ground

I share the ideas behind Common Ground on LinkedIn - what I’m hearing from DTC founders and operators, the AI and commerce shifts worth paying attention to, and how I turn those signals into each issue.

Follow along for the research, writing process, and practical takes that don’t always make it into the newsletter.

Follow Jacqueline on Linkedin

Subscribe now

Join Ground

The 48-Hour Window

Traffic that arrives from AI converts about 50% higher than organic search. That’s the good news, and it’s the last piece of good news in this issue.

Here’s the rest. Those shoppers behave nothing like your organic audience. They didn’t wander in from a Google rabbit hole with a vague idea and three weeks to think about it. They asked a specific question, got your product page as the answer, and are roughly 48 hours from a purchase decision. Then they’re gone - to a competitor, to a marketplace, to nothing at all.

Almost every DTC brand is trying to catch that buyer with a retention system designed in 2019 for leisurely browsers. A seven-day abandoned-cart drip that opens with “still thinking about it?” on day three is a system optimized to arrive after the decision has already been made.

The mismatch isn’t a copy problem. It’s a timing problem, and timing is architecture.

What the premium operators figured out first

Mytheresa’s parent company, LuxExperience, is deploying AI across Mytheresa, Net-a-Porter, Mr Porter and Yoox. CEO Michael Kliger has been direct about where the value actually sits: the highest-return application isn’t automating customer service. It’s identifying future VIP customers early enough to assign human concierge resources before their spending patterns prove they deserve it.

Read that again, because it inverts the standard AI pitch. Kliger isn’t using AI to cut cost. He’s using it to know where to spend more.

That’s the split forming across premium commerce right now. Brands using AI to automate support are competing on cost structure - a race that ends with everyone’s margins in the same place. Brands using AI to decide which inbound shopper gets the personal stylist, or which AI-referred visitor needs intervention in hour six rather than day seven, are competing on lifetime value. Those are two different businesses. Within a year they’ll be two different revenue categories.

Two builds before Q4

One: get quoted, not ranked. Ranking in Google still matters, but showing up inside AI citations is what actually moves traffic now. That means specifications, comparisons, and use-case language written directly into your page copy - not buried in a PDF or a size chart no model can parse.

Two: build retention calibrated to the window. Not a longer sequence. A faster one, triggered by intent signals rather than a timer.

🔌 This is exactly what we built. Ground’s revenue agents work inside your existing Klaviyo and Shopify stack - greeting every visitor, converting the ready buyer, and bringing back the lapsed one, while the relationship and the first-party data stay on your side of the table.

Greet AI owns first-touch conversion. ReCartify resolves identity. ReBeat owns retention. Each agent is pointed at a single revenue KPI.

Brands running Ground see 30–50% more first-time revenue and 2–4x higher repeat purchase revenue. Book a Demo

Book a Call

The through-line

AI is making your traffic better and faster, and “faster” is the part nobody built for. The brands still measuring success in seven-day windows will watch high-intent buyers arrive and leave, and they’ll read it as a traffic quality problem.

The sale is going to happen either way. The only question that matters is who keeps the customer.

Curated Intelligence

• Ecommerce Fastlane / The AI referral retention window - AI-referred shoppers convert 50% higher but churn inside 48 hours when they hit retention systems built for casual organic browsers. Read More

• Glossy / Mytheresa’s AI for VIP identification - Luxury ecommerce’s highest-value AI application isn’t automating service, it’s predicting which new shopper justifies expensive human attention before they’ve earned it. Read More

• Glossy / Oner Active’s salaried creators - $52M in creator-driven sales this year, from treating creators as exclusive salaried employees rather than transactional affiliates. Proof that solving for lifetime value beats solving for campaign CTR. Read More

🗣️ From the Community

Nothing we write about tightening your retention programs lands as hard as a customer saying it themselves.

One of our customers Dan Castle, former Managing Director at Martha Stewart and now founder of CastleFlexx, just shared publicly what Ground has meant for their growth - and we’re still a little floored by it.

Dan Castle runs CastleFlexx - the premium mobility and back-pain device - with a small team and a handful of AI agents (Humbly: ours). By his own words, he’s scaling the brand faster than Vuori and Under Armour did at the same stage, and he calls Ground the first growth lever in the business. See what Ground can do for your brand

Chat with Us

Subscribe now

While others rush to replace humans with AI, the smartest companies are focused on making humans unstoppable.

Join Ground

At Y Combinator's Startup School in July, Jensen Huang the CEO of Nvidia said the AI job-loss narrative is "exactly backwards" - "AI automates tasks away, but it doesn't necessarily eliminate jobs. The job of a person has a purpose, and that purpose has many tasks." He backed it with radiology: Hinton predicted in 2016 radiologists would be obsolete in 5 years; the field grew ~12% from 2010–2022 and is projected to grow another 25–40%. He's also said "the more AI we use, somehow the more people we have to hire" and, at Stanford, "most people will lose their job to somebody who uses AI"

Most people are missing what Jensen means: AI is not about replacing humans. It's about making them superhuman.

The Augmentation vs. Replacement Battle

While the tech world obsesses over AI that can do human jobs, the person with the most complete view of AI’s actual capability is making the opposite argument - and backing it with his own headcount.

“The more AI we use, somehow the more people we have to hire,” Huang has said, pointing to the fact that agentic systems are genuinely new skills requiring new roles. His framing: “This is a classic example of productivity increasing growth. Increasing growth drives more employment.”

This isn’t philosophical positioning. It’s a strategic read on where the leverage actually lives. The real AI revolution isn’t about fewer humans - it’s about more capable ones.

The DTC Parallel

This mirrors exactly what the smartest DTC brands are discovering about AI implementation. The companies winning with AI aren't the ones laying off their teams. They're the ones making their existing teams dramatically more effective.

Your email marketing specialist doesn't need to be replaced by AI, they need AI that can analyze thousands of customer behavioral data points in real-time. Your customer service team doesn't need to be eliminated, they need AI that can instantly surface every previous interaction and predict customer intent. At Ground, we see this philosophy in action every day - we're not trying to replace your marketing team or growth operators, we're giving them capabilities that make every decision smarter and every campaign more targeted.

🔌 Ready to turn supercharge your revenue team? Speak to Ground AI's growth experts today.

Let's Chat

The Immediate Results Reality

The brands that understand augmentation over replacement are seeing immediate, measurable results. Teams that were already performing well suddenly deliver results that seem impossible. Not because they've eliminated human decision-making, but because they've amplified human intuition with AI precision.

When your lifecycle marketing manager has AI that can predict exactly when each customer is ready for their next purchase, they don't just send better emails - they become a revenue-generating machine. When your retention specialist has AI that identifies at-risk customers and automatically crafts personalized win-back campaigns, they don't just reduce churn - they become a profit center.

The Strategic Reality

Nvidia is betting its hiring plan that augmented humans beat AI replacement. Smart DTC brands are making the same bet on a smaller scale - and seeing immediate returns.

While competitors debate whether AI will eliminate jobs, the winners are busy making their people unstoppable.

The question isn’t whether your team needs AI. It’s whether you’ll use AI to replace them - or to make them the coworker everyone else loses out to.

🔌 Brands using Ground’s AI revenue agents typically see 20 to 50% more revenue out of the same traffic they're already getting. A few August onboarding spots left - book a time with our team here.

Book a Call

🗣️ From the Community

Nothing we write about augmentation lands as hard as a customer saying it themselves.

One of our customers Dan Castle, former Managing Director at Martha Stewart and now founder of CastleFlexx, just shared publicly what Ground has meant for their growth - and we’re still a little floored by it.

Dan Castle runs CastleFlexx - the premium mobility and back-pain device - with a small team and a handful of AI agents. (Humbly: ours.) By his own words, he’s scaling the brand faster than Vuori and Under Armour did at the same stage, and he calls Ground the first growth lever in the business. See what Ground can do for your brand

Let's Chat

Finding Common Ground

The Ground community comes together sharing different stories, but common ground.

Learn More

Our very own Growth Strategist at Ground, Jacqueline, brings you the hottest news in commerce and what that means for growing brands today.

What Actually Happens After The First Customer Arrives

On July 14, Prenetics announced that IM8, the wellness brand it co-founded with David Beckham, had secured up to $1 billion in growth financing from General Catalyst’s Customer Value Fund.

This was not a billion-dollar check dropped into IM8’s bank account. General Catalyst can finance up to 70% of the brand’s selected monthly marketing spend, then collect a capped return from the revenue generated by each funded customer cohort.

But here is what most people are missing. General Catalyst is not betting $1 billion on David Beckham’s reach. It is betting on what IM8’s customers do after they click.

This is not a traffic problem. It is a yield problem.

Expensive Can Still Work

IM8’s disclosed customer acquisition cost rose from $104 in Q1 2025 to $305 in Q1 2026. That is nearly 3x as much to acquire a customer in one year.

Yet its mature cohorts reportedly paid back in roughly 3-6 months. Across those cohorts, every $1 of CAC produced $1.44 in blended gross profit. Those economics gave General Catalyst enough confidence to finance customer acquisition cohort by cohort.

IM8 did not prove that CAC is falling. It proved that higher CAC can still work when the customer produces enough value after acquisition.

The pressure is not going away. Meta reported that its average price per ad increased 12% year over year in Q2 2026. Brands can keep fighting for a cheaper click, but the auction will always belong to the platform.

What happens after the click belongs to the brand.

Own the Yield

Most acquisition dashboards end too early. They track CPM, CTR, CPC, ROAS and CAC, then treat the first purchase like the finish line.

It is not. The first purchase is where the customer economics begin.

The full equation includes how many paid visitors convert, how much first-order revenue they produce, how quickly CAC is recovered and how often those customers return. Shopify identifies a 3:1 LTV-to-CAC ratio as a commonly used benchmark for sustainable growth.

You may not control what Meta charges for the visitor. You can control how much revenue that visitor produces once they reach your Shopify store.

That leaves two levers. Convert more of the traffic you already paid for. Then create more value from the customers you worked so hard to acquire. Ground’s revenue agents does this all automatically with it’s suite on revenue agents.

See How Ground Converts Your Traffic

From Click to Compound

This is where revenue agents change the model. Instead of waiting for a team to build another campaign, they continuously act on the moments that determine whether an expensive visitor becomes a valuable customer.

Ground is an Autonomous Revenue System across Shopify and Klaviyo. Greet AI acts before the first order, helping brands drive 30-50% more first-time revenue from existing traffic. ReCartify resolves identity so the models understand who is engaging and abandoning. ReBeat AI acts after acquisition, driving 2-4x repeat purchase revenue. Terra connects the intelligence and execution across the journey.

These models do not operate independently. They work as one revenue brain.

The result is not another audience to rent. It is incremental revenue from the audience you already paid to reach.

🔌 If paid traffic is getting more expensive, Ground can show you where first-time and repeat revenue are escaping across your Shopify journey.

Book a Demo

Join Brands that have a Revenue System that makes Every Customer Compound

The most important asset in IM8’s financing is not the billion-dollar commitment. It is the customer cohort that can convert, repay its acquisition cost and keep purchasing after the fund’s return has been capped.

That is the new growth advantage. Not access to more traffic, but an autonomous revenue system that makes every hard-earned customer worth more.

If every click is getting more expensive, are you still trying to buy cheaper traffic, or are you building a revenue system that makes every customer compound?

If you’re curious about learning more, let’s chat.

Book a Demo

Curated Intelligence

• Prenetics and General Catalyst investor presentation - The CAC, payback and cohort economics behind the $1 billion commitment

• Prenetics’ official financing announcement - How the financing structure funds marketing without issuing equity

• Meta’s Q2 2026 results - Average ad prices increased 12% year over year

• Shopify’s 2026 customer lifetime value guide - How retention, purchase frequency and customer lifespan determine how much CAC a brand can support

• Common Ground’s revenue-agent explainer - Why the revenue after discovery belongs inside a system the brand owns

💡 Prompt Smarter

Give your AI an anonymized cohort table with acquisition month, spend, new customers, first-order gross profit and 30-, 60- and 90-day repeat revenue. Ask it to calculate CAC, gross-profit payback and LTV-to-CAC by cohort, then identify the largest revenue leak. Verify every result against your finance source data before acting.

Follow the Thinking Behind Common Ground

I share the ideas behind Common Ground on LinkedIn - what I’m hearing from DTC founders and operators, the AI and commerce shifts worth paying attention to, and how I turn those signals into each issue.

Follow along for the research, writing process, and practical takes that don’t always make it into the newsletter.

Follow Jacqueline on Linkedin

Learn More

There are enough founders who know the pattern by now.

We’ve sat in a good number of calls with founders that go something like this: Someone shows us a revenue screenshot.
It looks great.
Then we get one layer deeper, and suddenly cash flow is tight, margins are thin, and every month feels like survival instead of scale.

We’ve learned the hard way that a good revenue number can hide a bad business. Profit is what keeps a brand alive.

The instinct when profit is thin is almost always to fix acquisition: lower the CAC, find a cheaper channel, cut the ad budget. It’s the number everyone’s staring at. But it’s usually not where the actual leak is.

There’s a LinkedIn post that compared 50 dead DTC brands (Allbirds, Casper, Bonobos, Outdoor Voices) to a graveyard which was quite interesting. Completely different products, completely different categories, same cause of death: they built the whole business around buying customers cheaply through Facebook and Instagram, and never fixed what happened after checkout.

That worked right up until ad costs caught up with them (CAC climbed 222% between 2013 and 2022), and the traffic that had been propping up the whole model dried up with it.

The math underneath is brutal. The average DTC brand loses about $29 on a customer’s first order once you account for ad spend and returns. Only 27% of first-time buyers ever place a second order. The other 73% leave before you’ve earned back what you spent to get them.

What moves this number? In short: data. In long: an AI that learns from every transaction and gets smarter over time - so it knows the moment an existing customer is ready to buy again or ready to explore something new.

Ground is what that looks like in practice. Three AI models working together as a brain across your full customer journey: bringing in high-intent buyers, converting them on site, and compounding retention over time. It plugs into your existing stack in 15 minutes. No data science team. No technical lift. Just intelligence that starts working from day one.

Book a Demo

Why “profitable” got harder to see, not just harder to achieve

The brands doing this well aren’t necessarily smarter, they’ve just built the visibility to act early instead of reacting late. Real-time sell-through data lets teams shift investment into what's working while there's still time left in the season to react, instead of finding out three months later in a spreadsheet. That's the muscle DTC brands like U Beauty, Orebella, Nodpod, The great and hundreds more have already built using Ground’s Rebeat AI.

Request Demo for Rebeat AI

Where AI actually moves the needle

So where does AI help, concretely? Not by making ads cheaper. By making the after visible and personal:

• Predicting who’s about to leave, before they leave. Instead of a reactive win-back email three weeks after someone’s already checked out mentally, AI can catch the churn signal early enough to actually act on it with dynamic blocks and messaging.

• Replacing blanket discounts with the right offer for the right person. A sitewide “20% off, last chance” trains your best customers to wait for the next one and quietly erodes your positioning. A model that knows who actually needs the nudge (and who was going to buy anyway) protects both margin and brand.

• Giving you real-time contribution margin visibility, not just revenue, per SKU and per cohort, so decisions happen while there’s still time to act on them instead of after the quarter closes.

• Forecasting demand instead of guessing at it, so cash doesn’t sit trapped in overstock waiting for a markdown to move it.

• Personalizing the retention layer itself - loyalty perks, subscription cadence, post-purchase content - the same way strong omnichannel brands retain 89% of customers compared to 33% for brands with a weak one.

These are all why we built Ground. Ground accounts for 10% of DTC revenue for the brands we work with. That number grows as the AI models learn. The brands that moved early are now the ones with the compounding advantage.

Book a Demo

The honest caveat

None of this works if you’re just adding another AI tool because it’s trendy. If it doesn’t improve retention or margin, it’s overhead with better branding - and at that point the brand might as well skip the tool entirely than bolt one on and call it a strategy.

This is exactly the gap Ground was built to close - Where most DTC brands are still flying blind between campaigns, Ground's AI agents work continuously in the background capturing high-intent visitors who slipped through the first time, re-identifying returning shoppers whose cookie data has expired, and surfacing the right product to the right customer at the right moment. The result isn't just more automation. It's better visibility into who's on your site, what they want, and when to reach them - turning traffic you're already paying for into revenue you were previously leaving on the table.

If you’re curious about learning more, let’s chat.

Book a Demo

Curated Intelligence

• Why Analytics and Agility Are Critical to Success in Today’s Retail Market Read More

• D2C’s Profitability Crisis Has 1 Root Cause: Focus on Acquisition, Instead of Loyalty Read More

• The Fashion Marketer’s Guide to AI Read More

💡 Prompt Smarter

Did you know: you can ask an AI model to calculate your true contribution margin per SKU right now, if you feed it your CAC, discount rate, shipping cost, and return rate by product line. Most brands have never seen that number broken out this granularly, and it usually tells a very different story than top-line revenue does.

Finding Common Ground

The Ground community comes together sharing different stories, but common ground.

This issue comes from Sheida Mirjahani, Growth Lead at Ground, who writes about AI in commerce, ecommerce business math, and how brands are actually growing online.

Learn More

I saw the IM8 news go by on LinkedIn a couple of weeks ago and I keep thinking about it.

IM8 is the supplement brand David Beckham co-founded raised $1 billion, and gave up no ownership at all. It’s a rev-share deal: General Catalyst pays for IM8’s customer acquisition upfront, then takes a capped cut of the revenue those customers generate until it’s paid back. Once the cap is hit, IM8 keeps everything.

What This Deal Actually Proves

I don’t think the takeaway here is “celebrity brands can raise money.” It’s that can you prove your LTV: CAC making sense?

General Catalyst has published the criteria to qualify as a brand: you should be able to get your acquisition money back within 24 months.

Read that again. It’s all about can you convert visitors to customers? Are they coming back? How fast and why? Do you have a repeatable growth engine?

A few brands have quietly figured this out.

Getting More First-Time Customers From Traffic You Already Earned

Picture this: ChatGPT recommends your moisturizer as the best option for a woman in her 40s with sensitive skin. She lands on your site, sees your welcome offer within five seconds, closes it, browses for a few minutes, and leaves.

Based on our research, 95% of visitors never opt in - so you can’t capture them, and they never enter your email flows. All that hard-earned traffic walks out the door.

That’s what Ground’s Greet AI was built to solve. It sits on your Shopify store, observes every individual visitor, identifies the right moment, and re-engages them exactly when they’re ready to buy.

You don’t need new flows. You don’t need new creative. You just need more chances to talk to the customer you already paid for.

Book a Demo

Getting Those Customers to Buy Again

Now the other half.

Someone bought a face cream from you 40 days ago. They finished it in 32 - faster than most people do.

On day 35 they came back to your site and looked at the same product. Then they left.

Your win-back email was scheduled for day 40. By the time it landed, they’d already bought something else.

ReBeat AI is our answer to that. It’s our AI agent that learns how quickly people actually run out, across your whole customer base. It notices when someone comes back to your site, knows which product they were looking at (maybe it’s a serum this time?) and sends the email or text on that person’s timeline instead of a set schedule.

It activates repeat purchase on it’s own in three ways. Replenishment, Cross-sell, and one-time buyer to subscriber conversion.

Book a Demo

Brands Using Ground Convert 30% More ChatGPT Visitors Into First-Time Customers

(And 3-6x more repeat purchases!)

Most platforms you’ve added to your stack do one thing: they help you do something better. A pop-up tool captures more emails. An ESP sends better sequences. An analytics platform shows you what’s happening.

Ground is different. It’s not a tool you manage - it’s a suite of AI revenue agents that run autonomously, learn from your customers, and generate revenue your existing stack never would have found.

Book a Call

Curated Intelligence

• The IM8 deal, explained. How the rev-share structure works and why IM8 gave up no equity. TechCrunch, Jul 2026. Read More

• The numbers behind the raise. Prenetics’ updated guidance and the $1.44-per-dollar disclosure. GlobeNewswire, Jul 2026. Read More

• Why funding ads separately is becoming a thing. General Catalyst’s own case for splitting growth capital from equity. Read More

• What IM8 does with the money. Hydration in Q4, a premium gummies line in Q1. NutraIngredients, Jul 2026. Read More

• Why your CPMs are up. The structural reason competitive categories are seeing 30–40% jumps. Threadpoint. Read More

Learn More

In our calls with brands recently, I’ve found that there’s been a common overarching theme: AI and AEO mean the same thing. This is not true, but I of all people fully understand how all this AI jargon can be confusing so let me break it down.

AEO and AI (and thus AI agents) are solving two very different problems:

AEO helps your brand appear in the answer.

Agents can act on their own to hit a business KPI: increasing customer acquisition, increasing replenishment, or converting more OTP to subscriptions.

If AEO Is All You’re Doing for Your Brand, You’re Way Behind

For a lot of marketing teams, “AI strategy” now means AEO.

Can ChatGPT find the brand? Does the product appear in recommendations? How often does an answer engine cite the company instead of a competitor?

Those questions matter. If your product information is unstructured, your policies are buried, or nobody credible mentions your brand, AI may never put you in the consideration set.

But the AEO playbook is still recognizably search marketing: publish comparison pages, earn third-party citations, add structured data, and monitor share of voice.

It is a new discovery channel, but AI can do much more than that.

Agents Can Bring You New or Repeat Customers on Their Own

An answer engine recommends a running shoe. An agent can compare the options, check the sizing policy, create the cart, and complete the purchase within rules set by the shopper.

That is the difference between generative AI and agentic AI. One produces an answer. The other works toward an outcome.

AEO asks, “Can AI read your content and use it in an answer?”

AI agents and Agentic commerce asks, “Can AI do business with you?”

Whats the action? Ground’s AI revenue agents are your business partner and are optimized for one thing only:

Can they increase revenue?

Brands like Heaven Mayhem, Ghia, Adapt Naturals, and many others are already seeing great increases in their DTC revenue with Ground’s revenue agents.

Book a Demo

So They Discovered You, Did You Turn Them into Customers?

Picture this: ChatGPT recommends your moisturizer as the best option for a woman in her 40s with sensitive skin. She lands on your site, sees your welcome offer within five seconds, closes it, browses for a few minutes, and leaves.

Based on our research, 95% of visitors never opt in - so you can’t capture them, and they never enter your email flows. All that hard-earned traffic walks out the door.

That’s what Ground’s Greet AI was built to solve. It sits on your Shopify store, observes every individual visitor, identifies the right moment, and re-engages them exactly when they’re ready to buy.

Book a Demo for Greet AI

AEO Brings Visitors. Revenue Agents Turn Them Into First-Time Customers.

A complete AI-commerce strategy has three layers:

1. Visibility: Can AI discover and accurately describe the brand?

2. Transactability: Can an external agent evaluate and purchase the product?

3. Revenue execution: Can the brand respond to customer behavior and create incremental revenue?

Most AEO conversations stop at layer one.

🔌 Ground is built for layer three. Our 3 revenue agents act as one brain. They integrate with Shopify and your CRM, and autonomously grow both customer acquisition and repeat purchases through cross-sell, replenishment, and one-time purchase to subscription conversion.

Greet AI turns more visitors into first-time customers. ReCartify recovers the sales your ESP can’t even see. ReBeat AI combines purchase history with live on-site behavior to predict who buys next, what they buy, and when, then surfaces that exact SKU by email or SMS.

Book a Demo

Brands Using Ground Convert 30% More ChatGPT Visitors Into First-Time Customers

Ground’s Visibility-to-Revenue Teardown maps what happens after an AI recommendation sends someone to your store. We will show you where demand becomes anonymous, where intent stalls, and where an Autonomous Revenue System can act.

Book a Call to See Ground in Action

Curated Intelligence

• The Complete Guide to Ecommerce AI SEO - The current AEO playbook for product visibility, citations, structured data, and revenue attribution.

• Agentic Commerce Strategy - A plain-English distinction between AI that advises and AI that decides, executes, and transacts.

• Shopify Spring ’26 Edition - The commerce infrastructure being built for agent-readable catalogs and end-to-end transactions.

• The Agentic Commerce Gold Rush - New adoption data and the trust gap emerging as merchants connect agents to commerce.

• AEO and GEO Platforms for Ecommerce - A useful map of visibility platforms, action layers, and revenue attribution. The publisher also operates an AEO platform.

  Thanks for reading Common Ground! Subscribe for free to receive new posts about AI x Commerce.

This issue comes from our very own customer growth strategist at Ground, Jacqueline Ngan, who’s on a mission to cut through the noise in AI x ecommerce - and bring the clearest read straight to you.

Subscribe now

AI is becoming the interface to commerce. Here’s what’s shaping the next generation of AI-powered growth. Let’s talk Strategy.

Join Ground

There’s no shortage of AI headlines right now. But inside boardrooms, marketing teams, and product meetings, a quieter shift is happening.

Founders, CEOs, CMOs and Heads of E-commerce of being pressured with the task of using AI…having an AI strategy. The biggest question? “How?”

At Ground, we don’t just build AI tools. We act as a partner in your growth. That means giving you the insight, infrastructure, and education to stay ahead of the curve.

Here are the 6 signals we’re tracking most closely right now, and how they’re already reshaping consumer behavior:

1. Search: AI referrals are a discovery channel with their own rules

   ChatGPT is building richer product discovery directly into the conversation. Shopify’s Agentic Storefronts let merchants make their products discoverable across ChatGPT, Microsoft Copilot, Google AI Mode, and Gemini from Shopify Admin. Your product page is becoming both a landing page and an answer delivered somewhere else. If an AI model cannot confidently understand your product, it cannot confidently recommend it. That means structured product data matters as much as persuasive copy. Price, inventory, variants, ingredients, reviews, shipping, and return policies need to be current, consistent, and machine-readable.

2. AI-native advertising is here

   OpenAI now has a live ads manager, while Google has introduced Conversational Discovery ads, Highlighted Answers, and AI-powered Shopping ads. The AI-native ad has to survive a conversation.

   Creative will still matter, but brands must also prove relevance, answer follow-up questions, and make the next action easy.

   🔌 ReBeat AI Arc, one of Ground’s AI agents, matches each returning shopper with the ad creative most likely to bring them back and generate incremental revenue.
   

   Book a Demo

3. Values are the algorithm

   AI shopping assistants can compare the details that brand familiarity once helped consumers skip. Your positioning cannot live only in a campaign. Price, quality, ingredients, impact claims, reviews, and policies have to substantiate it. AI will not reward what a brand says it values. It will reward what the available evidence proves.

   

🔌 This is what Ground’s AI Revenue Agents are built for. Greet AI, ReCartify, and ReBeat AI work through Terra as one connected intelligence layer, learning how visitors engage, abandon, and return, then acting across acquisition and retention to create incremental revenue. The storefront may be changing, but owning first-party customer context is becoming more valuable, not less.

Learn More

4. Voice is now a behavior, not a novelty

   Alexa+ for shopping are 60% more likely to complete a purchase. It can browse options, compare products, reorder a shopper’s usual brand, monitor prices, and buy when the conditions are right. Your product knowledge has to answer the way people actually speak: “Is this safe for sensitive skin?” “Will it arrive before Friday?” “Is there a less expensive option that does the same thing?”

   The brand with the clearest answer will win before a page is ever opened.

5. Automation is now interface-free

   This month, Visa announced it had completed live agentic commerce transactions with AI purchasing from participating merchants on behalf of cardholders.

6. AI isn’t the enemy, it’s the expectation

   71% of consumers expect generative AI to influence at least half of their spending decisions over the next year. Your brand does not need another disconnected AI pilot. It needs product data AI models can read, value they can verify, and customer context your own systems can act on.

AI is about perspective. And those who see clearly will build better.

Curated Intelligence

• The channel already has traffic. Adobe tracks the rapid growth of AI referrals across retail and other industries. Read More

• Ads are becoming conversations. Google explains the new AI-native ad formats appearing across Search and shopping. Read More

• The buyer now has a proxy. Accenture explores how consumers will instruct AI to choose, negotiate, reorder, and switch brands. Read More

• Voice has reached checkout. Amazon shows how Alexa+ can compare, reorder, monitor prices, and purchase. Read More

• AI is already transacting. Visa details its first live agentic commerce purchases in Europe. Read More

Build for the AI Interface

Built for Innovators. Curated for What’s Next.

Ground is the suite of AI Revenue Agents for consumer brands. Its Autonomous Revenue System connects Greet AI, ReCartify, and ReBeat AI through Terra, turning first-party customer behavior into action across acquisition and retention.

Book a Demo to see Ground in Action

Finding Common Ground

If you found this valuable, subscribe and share it with the operator on your team asking where AI actually fits.

Until next time,
Jacqueline and the Ground AI team

AI is getting incredibly overwhelming. Every week brings another platform, model and use case that is supposedly going to change how a brand operates.

So I decided to do a little digging (okay, a lot) into how DTC teams are actually using AI day to day, beyond writing copy and recycling the same prompts. Here are the workflows creating real leverage now.

Learn More

DIY Agents vs Done-For-You Agents

We’re seeing 2 patterns in how brands are adapting agents: either to make them themselves, or using à la carte ones.

DIY (OpenClaw / Hermes) agents: This is when brands build agents internally. In 2026 more brands started to adapt OpenClaw or Hermes, giving them access to their business context such as customer chats, product catalogue, Shopify and Klaviyo data and are building agents to take care of reporting, or managing their day to day.

Done-for-you agents: These are agents built by companies like Ground that get installed in your tech stack, are optimized to hit a business KPI and don’t stop until they do. They get smarter because as more brands use them, they know what works and what doesn’t to hit their goal.

For example Ground’s AI revenue agents are optimized for one goal only:

Can they increase revenue?

Brands like Heaven Mayhem, Ghia, Adapt Naturals, and many others are already seeing great results in their DTC revenue with Ground’s revenue agents.

Book a Demo

Here are 8 other ways we saw brands using AI for DTC growth this month:

1. Make AI argue against the weekly report

Obvi, a wellness brand famous for their flavored collagen powders, combined Shopify, Meta, email, subscription and support data into a weekly adversarial report. It found that 18% revenue growth came from fewer customers spending more, while stronger email open rates were hiding declining click-through.

AI should not just summarize the numbers. It should tell your team where the story is misleading.

2. Build a brand brain

Cumulus Coffee put its mission, customer reviews, social comments, financial targets and best creative briefs into one persistent Claude project. The nthe project acted as a specialist then was serving as creative strategist, copywriter and channel partner for its three-person marketing team.

3. Ask why customers are cancelling

Love Me Gluten Free used Stay AI’s conversational analytics and found that it had 11.5% churn. A single February day generated 50% of that month’s cancellations.

This is not churn reporting. It is churn diagnosis.

4. Use edge AI to observe new visitors and show welcome offers when intent is high

Heaven Mayhem installed one of Ground’s AI agents, Greet AI, into their Shopify and CRM and they saw a lift on both their new welcome offer sign ups and their first-time customer revenue.

Greet AI is an edge AI (the same technology Tesla uses to observe and act intelligently) that sits on Shopify, observes each person coming to the website and interprets what they are doing, when they’re pausing, what their next action is, and show the welcome offer when the visitor is locked in to buy.

Book a Demo for Greet AI

5. Give shoppers a product expert on demand

Redmond built a production commerce assistant in ten weeks with a two-person team. It now handles thousands of monthly conversations with guardrails around recommendations and customer-facing answers.

The opportunity is not removing the human element. It is giving every shopper immediate access to the brand’s best product knowledge.

6. Predict the next purchase instead of pushing another subscription

Built Bar used rePete AI to identify when individual customers were likely ready to purchase again. They report that non-subscriber repeat revenue reached four times subscription revenue within two months.

🔌 Ground applies the same principle through ReBeat AI and ReBeat Arc. ReBeat AI predicts when each customer is ready to buy again, while ReBeat Arc helps turn the right one-time purchases into replenishment subscriptions. Together, they drive earned repeat revenue instead of another calendar-based blast.

Book a Demo

7. Prevent the return: Let shoppers try the product before the return happens

Apparel brands are moving virtual try-on from novelty to conversion infrastructure, helping fashion shoppers visualize a product before it arrives. The promise is not merely a more entertaining product page. It is fewer purchases made with the wrong expectations.

8. Talk to the ERP and let it do the work

Fulfil lets DTC operators query orders, sales, inventory and financials through Claude or ChatGPT in plain English. It can also create orders, apply payments, adjust inventory and post journal entries with the company’s existing permissions.

This is where AI stops describing the business and starts operating inside it.

Consumer brands using revenue agents are seeing DTC revenue climb on autopilot. Join them.

Average teams give AI a prompt. AI-native teams give it memory, proprietary data and permission to act.

The question is no longer whether your team uses AI. It is whether AI knows enough about your business to make a better decision than another blank page of copy.

🔌 Ground’s AI Revenue Agents connects Greet AI, ReCartify and ReBeat AI through Terra, creating incremental revenue across the customer lifecycle. Brands using Ground see 30-50% more first-time revenue and 2-4x higher repeat purchase revenue. Setup takes 15 minutes.

Book a Call to See Ground in Action

Curated Intelligence

• The AI operating stack for consumer brands. Why structured data and clean foundations determine whether AI creates leverage or more chaos. Read More

• How the AI adoption gap is widening. A useful framework for moving from one-off conversations to persistent projects and connected systems. Read More

• Inside Cumulus Coffee’s brand brain. A three-person team’s system for turning customer context into cross-channel creative. Read More

• What an AI-native ERP can do. How Claude and ChatGPT can read and write to live orders, inventory and financial data through MCP. Read More

  Thanks for reading Common Ground! Subscribe for free to receive new posts about AI x Commerce.

Learn More

Common Ground is Ground’s newsletter highlighting the most important consumer news, tech trends, artificial intelligence, in-depth analysis, along with a few extra thoughts from our team. Subscribe today to get the latest issue delivered to your inbox.

This issue comes from our very own customer growth strategist at Ground, Jacqueline Ngan, who’s on a mission to cut through the noise in AI x ecommerce - and bring the clearest read straight to you.

AI is getting incredibly overwhelming. Every week brings another platform, model and use case that is supposedly going to change how a brand operates.

So I decided to do a little digging (okay, a lot) into how DTC teams are actually using AI day to day, beyond writing copy and recycling the same prompts. Here are the workflows creating real leverage now.

Learn More

AI agents are best for finding patterns you can’t

Obvi, the eight-figure collagen and supplement brand, says AI saved roughly $4,000 by searching eight months of UGC. Love Me Gluten Free found that one February day caused 50% of its monthly cancellations. Built Bar reports four times more non-subscriber repeat revenue after two months of AI-timed reorders.

But here is what most people are missing: none of those wins came from asking ChatGPT to write another product description.

Copy is the entry point. Context is the advantage.

The brands pulling ahead are giving AI access to customer conversations, creative libraries, purchase behavior and live operating systems. This is not a writing shortcut. It is a new intelligent operating layer.

Here are ten ways it is already showing up...

1. Find what the best salesperson does differently

Obvi split months of pre-sale conversations into those that converted and those that did not. Its best salesperson was not the fastest or the most aggressive. She asked what the shopper was trying to solve, recommended the cheaper product when it fit and occasionally talked someone out of buying entirely.

In ecommerce, hand-holding every customer with a live salesperson is impossible-there just isn't the manpower. That's why we're seeing such strong results with Ground's Greet AI agent. It sits on top of Shopify and runs on edge AI (the same technology Tesla uses to observe its surroundings and decide what to do in real time), acting as a salesperson for your store: reading on-site behavior as it happens and stepping in with the right offer only when a visitor is ready to buy.

Book a Demo

2. Make AI argue against the weekly report

Obvi also combined Shopify, Meta, email, subscription and support data into a weekly adversarial report. It found that 18% revenue growth came from fewer customers spending more, while stronger email open rates were hiding declining click-through.

AI should not just summarize the numbers. It should tell your team where the story is misleading.

3. Build a brand brain

Cumulus Coffee put its mission, customer reviews, social comments, financial targets and best creative briefs into one persistent Claude project. Specialist workflows then acted as creative strategist, copywriter and channel partner for its three-person marketing team.

The output was not generic because the context was not generic.

4. Search months of UGC by meaning, not filename

Instead of manually scrubbing through eight months of videos, Obvi used Recharm and Claude to search by meaning and return the relevant clips with exact timecodes. The team estimated it avoided roughly $4,000 in new production and search work.

The creative already existed. AI made it usable.

5. Ask why customers are cancelling

Love Me Gluten Free used Stay AI’s conversational analytics and found that it had 11.5% churn. A single February day generated 50% of that month’s cancellations.

This is not churn reporting. It is churn diagnosis.

6. Give shoppers a product expert on demand

Redmond built a production commerce assistant in ten weeks with a two-person team. It now handles thousands of monthly conversations with guardrails around recommendations and customer-facing answers.

The opportunity is not removing the human element. It is giving every shopper immediate access to the brand’s best product knowledge.

7. Predict the next purchase instead of pushing another subscription

Built Bar used rePete AI to identify when individual customers were likely ready to purchase again. They report that non-subscriber repeat revenue reached four times subscription revenue within two months.

🔌 Ground applies the same principle through ReBeat AI and ReBeat Arc. ReBeat AI predicts when each customer is ready to buy again, while ReBeat Arc helps turn the right one-time purchases into replenishment subscriptions. Together, they drive earned repeat revenue instead of another calendar-based blast.

Book a Demo

8. Prevent the return: Let shoppers try the product before the return happens

Apparel brands are moving virtual try-on from novelty to conversion infrastructure, helping fashion shoppers visualize a product before it arrives. The promise is not merely a more entertaining product page. It is fewer purchases made with the wrong expectations.

9. Talk to the ERP and let it do the work

Fulfil lets DTC operators query orders, sales, inventory and financials through Claude or ChatGPT in plain English. It can also create orders, apply payments, adjust inventory and post journal entries with the company’s existing permissions.

This is where AI stops describing the business and starts operating inside it.

10. Audit claims before a regulator does

One unnamed supplement brand in Obvi’s operator network compared its site and email claims against twelve months of FDA warning letters. Claude flagged four claims for review before they became a larger problem.

Growth is not only about producing more. Sometimes it is about catching the sentence that should never ship.

Join Brands Who Are Already Stitching Operational Systems into One Intelligent Operating Layer

Average teams give AI a prompt. AI-native teams give it memory, proprietary data and permission to act.

The question is no longer whether your team uses AI. It is whether AI knows enough about your business to make a better decision than another blank page of copy.

🔌 Ground’s AI Revenue Growth Engine connects Greet AI, ReCartify and ReBeat AI through Terra, creating incremental revenue across the customer lifecycle. Brands using Ground see 30-50% more first-time revenue and 2-4x higher repeat purchase revenue. Setup takes 15 minutes.

Book a Call to See Ground in Action

Curated Intelligence

• The AI operating stack for consumer brands. Why structured data and clean foundations determine whether AI creates leverage or more chaos. Read More

• How the AI adoption gap is widening. A useful framework for moving from one-off conversations to persistent projects and connected systems. Read More

• Inside Cumulus Coffee’s brand brain. A three-person team’s system for turning customer context into cross-channel creative. Read More

• What an AI-native ERP can do. How Claude and ChatGPT can read and write to live orders, inventory and financial data through MCP. Read More

  Thanks for reading Common Ground! Subscribe for free to receive new posts about AI x Commerce.

Common Ground is Ground’s newsletter highlighting the most important consumer news, tech trends, artificial intelligence, in-depth analysis, along with a few extra thoughts from our team. Subscribe today to get the latest issue delivered to your inbox.

This issue comes from our very own customer growth strategist at Ground, Jacqueline Ngan, who’s on a mission to cut through the noise in AI x ecommerce - and bring the clearest read straight to you.

Lately I’ve had a tough time keeping up with all the jargon in the AI E-commerce space and it’s getting a little out of hand. Revenue agents, ambient agents, AEO, GEO, agentic this, autonomous that. Half of it is the same few ideas wearing different labels. So before we get to the one that actually decides who owns your customer, here is the simple breakdown.

Learn More

AI x Ecommerce Terminology at a Glance

💡 Revenue agent: autonomous AI tied to a revenue outcome. Conversion, retention, repeat purchases, upsells, abandoned carts, customer activation. This is our world. Ground’s revenue agents drive new DTC revenue for brands like goop, Glossier, Makeup By Mario, Tower28, and MaryRuth Organics.

💡 Ambient agent: AI that works invisibly in the background. You do not prompt it. It watches for signals, notices when something matters, and acts or escalates at the right moment.

💡 AEO (Answer Engine Optimization): shaping your content to be the direct, short answer a model gives.

💡 GEO (Generative Engine Optimization): shaping how your brand gets cited and synthesized across conversational AI like ChatGPT, Gemini, Perplexity, and Claude.

Notice the split already. AEO and GEO are about getting found inside someone else’s AI. A revenue agent is about what happens on your own turf. Hold onto that difference, because it is the whole story.

The agents are here, and the market is not subtle about it. Salesforce shipped its shopper agents this month, and the executive running its commerce business did not hedge: “If you don’t have agentic live as of yesterday morning, you’re already behind.” Bain projects agentic commerce will reach $300 to $500 billion by 2030, up to a quarter of all US online retail. But brands have seen this movie twice before, and both times it ended the same way.

This Is the Third Time the Industry’s Felt a Shift

The internet disintermediates (cuts out the middle man of) the brand and its customer about once a decade, and it always arrives disguised as a convenience.

Search came first. Google made discovery effortless, and then brands spent twenty years paying rent, in SEO and SEM, to reach customers who were already looking for them.

Social came second. Meta and TikTok made attention effortless, and then brands paid rent again, in ad spend, and lost the direct line to their audience. iOS 14 was the reminder of how fragile rented reach is. One privacy update, and the targeting a lot of brands had built their business on stopped working overnight.

Agents are the third wave. LLMs are making the entire purchase effortless, from discovery to checkout. This time the rent is not just attention. It is the transaction and the first-party data too.

Every wave, the same brands won: the ones who built something they owned instead of only renting reach.

Book a Demo

What the Shopping Agent Takes

When a shopping agent stands between a brand and its customer, the sale might still happen, but the brand loses everything around it. The relationship. The first-party data. The cross-sell. The ad inventory it used to monetize. The agent keeps the shopper. The brand becomes a supplier.

This is not hypothetical. 71% of marketing leaders already say AI agents have weakened their direct connection to customers. Salesforce drew the line itself: there is a difference between shopping inside an LLM and shopping on your own site, where you keep control of the transaction and the data.

What a Revenue Agent Does

A revenue agent is the part you own, and it is ambient by nature. It works in the background on your own site, greeting the visitor who lands, learning who they are, converting them, and bringing them back. Every one of those moments does two jobs at once. It drives revenue, and it keeps the relationship and the data on your side of the table.

The winning move is not to fight the shopping agents. It is to do both: show up inside the shopping agents so you stay discoverable, and own the revenue agent that keeps the customer once they arrive.

Join Brands Who Are Already Optimizing for Revenue With Ground’s Revenue Agents

🔌 Ground’s AI works inside your Klaviyo and Shopify stack, greeting every visitor, converting the ready buyer, and bringing the lapsed customer back, all while the relationship and the data stay yours. Brands using Ground see 30-50% more first-time revenue and 2-4x higher repeat purchase revenue. Setup takes 15 minutes.

Learn More

Ground has a suite of agents, each tasked with a different ecommerce revenue KPI to optimize for. Greet AI owns the first-touch conversion. ReCartify resolves identity. ReBeat owns retention. Terra knows your repetitive tasks as a marketer and do them for you. All designed to increase the revenue you’re not capturing.

What Discovery Means Now For Consumer Brands

Every brand faces the same choice now. Feed your customers into someone else’s agent, or run one of your own.

Being discoverable inside a shopping agent is rent. Owning the agent that keeps your customer is real estate. The brands that win the agentic era will be the ones who see the difference early, the same way the winners of search and social did.

When an agent finally makes the sale, the only question that matters is who keeps the customer.

Book a Call to See Ground in Action

Curated Intelligence

• How Salesforce is positioning for the agentic landscape. Shopper agents, and the line between selling inside an LLM and selling on your own site. Digital Commerce 360, Jul 2026. Read More

• Who gets to transact. A sharp map of the agentic commerce stack and where the value is actually accruing. Activant Capital. Read More

• Agentic commerce is an engagement problem, not just a transaction one. Why brands risk losing the direct relationship, and what to do about it. Braze. Read More

• AI agents for ecommerce: what is real and what is hype. A clear-eyed taxonomy of the agent categories brands are being sold. DTC Skills. Read More

• The agency-led ecommerce model is changing. Why continuous agentic execution is outpacing campaign-cycle agency work. CommerceIQ / MartechView, Jun 2026. Read More

  Thanks for reading Common Ground! Subscribe for free to receive new posts about AI x Commerce.

Learn More

This issue comes from our very own customer growth manager at Ground, Noopur Luthra, who’s breaking down how hims and hers grew when nobody was expecting it.

I almost didn’t write about this one because the headline made it sound like a disaster…

“Hims & Hers posts $92 million loss.” Stock down 15% in a day. Missed Wall Street’s estimates. If you only read the headline, it sounds like a cautionary tale. Basically like the kind of story you’d use to warn a founder about growing “too fast.”

But the thing is, I’ve spent enough time now on both sides of this - building my own brand, and now working with dozens of consumer companies on growth - to know that a scary-looking number in a single quarter and a bad decision are not the same thing. In fact, often times they’re the opposite. And the more I read into what actually happened at Hims & Hers this year, the more I think this is one of those times.

Before we get into the loss, there’s a genuinely wild growth story here, so let’s take a moment to understand both halves.

How Hims & Hers scaled to $1.4B in the first place

Hims & Hers didn’t get big by selling a better product. They got big by building a better system. They stopped treating themselves as just another healthcare company and started treating themselves as the next best health & wellness DTC brand.

A few things stood out to me:

1. Their website’s first impression is a quiz, not a store. No product grid, no ten-paragraph pitch. Just “what do you actually need help with?,” then a five-minute quiz that asks about you before it asks for your email. By the time they ask for your information, you’ve already invested a few minutes, and you genuinely want the personalized answer on the other side. That’s not an accident. That’s founder-level understanding of what actually gets someone to convert, and I have a lot of respect for how simple they’ve kept it.

2. Content isn’t a blog, it’s an acquisition channel. Over a million monthly visitors come through organic content, built on a topic-cluster strategy - one core topic, dozens of subtopics underneath it, and every piece medically reviewed for credibility (huge plus!). It’s the kind of unglamorous, compounding work that doesn’t show up in a highlight reel but quietly outperforms almost every paid channel over time.

3. They test relentlessly, then go all in on what works. Over a thousand live ad variants running at once. Millions spent on video in a single week when they find an angle that converts. Most brands test three ads and call it a strategy. Hims & Hers treats testing like a full-time department (which let’s be real, it probably is), and only scales once something’s actually proven. They bet small on everything, then bet big on the winner.

4. They own the infrastructure underneath the brand. Their own pharmacies. Their own manufacturing. That’s not a marketing decision. That’s a margin and control decision, and it’s why they’ve been able to personalize at a scale most DTC brands can’t even touch.

5. Subscriptions do the heavy lifting. 98% of revenue is recurring, and they’ve really hit the jackpot with this. Not because subscriptions are trendy, but because predictable revenue is what allows them to keep reinvesting in all of the above.

Put together, that’s not really a healthcare story. It’s a growth operating system that happens to be wrapped around healthcare.

And shameless plug, but this the same principle behind what we’ve built at Ground. Retention isn’t just a follow-up email - it’s understanding a customer well enough, in real time, to give them the next right thing before they think to ask for it. That’s what ReBeat does for consumer brands without needing a MedMatch-sized R&D budget to pull it off. If you’re curious about how Ground’s AI intelligence layer can add value to your DTC brand, let’s chat.

Book a Call

The Q1 2026 loss, and why it may not actually be worst thing for them

In Q1 of 2026, Hims & Hers reported a $92 million loss, compared to a $49.5 million profit the same quarter a year before. Revenue was up 4% to $608 million, but that still missed analyst expectations, and the stock dropped accordingly.

Here’s the part that actually matters: this wasn’t the business breaking. It was the business walking away from a model it knew wouldn’t last.

For a while, a meaningful chunk of Hims & Hers’ weight loss revenue came from compounded GLP-1 medications — a workaround that existed because branded drugs like Wegovy and Ozempic were in shortage. That workaround was never going to be permanent, and to their credit, they didn’t wait for a regulator to force the issue. Instead, they struck a deal with Novo Nordisk to offer branded GLP-1 products directly (and wildly, Novo dropped its lawsuit against the company as part of it). The Q1 loss included $33 million in restructuring costs tied to writing down a compounding supply chain that, in their own words on the earnings call, was already at risk of becoming obsolete.

So yes, gross margin dropped from 73% to 65%. Revenue per subscriber dipped from $85 to $80. Those are real costs. But subscribers still grew to 2.6 million, and within weeks of launching the Wegovy pill and pen, the company said it was on track to add over 100,000 new weight-loss subscribers a month (just absorb that for a second, because that’s truly an insane number). They’d also fulfilled more than 125,000 Wegovy shipments within a mere six weeks of getting direct access.

If you only look at one quarter, this looks like a stumble; but if you look at what they were protecting (a business built on trust and long-term subscriber relationships, not on a temporary supply loophole), this looks like exactly the kind of short-term pain a company should be willing to take. Personally, I have respect for a brand that writes down $33 million on purpose than one that gets blindsided by a regulatory shift they saw coming and ignored.

The AI layer with Hims & Hers that nobody is talking about

The part I think is getting buried under all of the earnings headlines: Hims & Hers has quietly built a real AI layer into the product, and it’s doing more than the baseline marketing tech usually does.

Their system, MedMatch, analyzes patient data to recommend treatments and reportedly hits a 93% match rate on those recommendations while cutting wait times by roughly 40%. That’s not an AI chatbot bolted onto a website for the sake of a press release. That’s AI doing the actual clinical-adjacent work of getting someone to the right treatment faster, which is probably why retention numbers look the way they do: 82% of subscribers renew past the first three months, and 85% are still active two years in.

They just extended that thesis with Labs AI, an assistant built to help interpret biomarker lab results as they expand into diagnostics. The pattern is consistent - they’re not using AI to generate more content or answer more support tickets, but instead using it to make the core product experience smarter, which is a much harder (and more valuable) thing to get right.

And shameless plug, but this the same principle behind what we’ve built at Ground. Retention isn’t just a follow-up email - it’s understanding a customer well enough, in real time, to give them the next right thing before they think to ask for it. That’s what ReBeat does for consumer brands without needing a MedMatch-sized R&D budget to pull it off. If you’re curious about how Ground’s AI intelligence layer can add value to your DTC brand, let’s chat.

Book a Call

The takeaway from Hims & Hers decision making

The lesson I keep taking from this one isn’t really about healthcare, or even about AI. It’s something I learned the hard way building my own brand: the brands that last aren’t the ones that avoid taking a hit. They’re the ones that take the hit on their own terms, before someone else forces it on them.

A $92 million loss is a scary headline. But a healthcare company voluntarily walking away from $200 million in supply-chain-shortage revenue, before a regulator or a competitor made that decision for them? That’s not the story of a company in trouble. That’s the story of a company that read the room correctly and was willing to eat a bad quarter to be right for the next five years.

Curated Intelligence

• The $3.7 Billion Subscription Machine: How Hims & Hers Disrupted American Healthcare: Watch Here

• Hims & Hers posts $92M loss in Q1 as it shifts to branded GLP-1 medications: Read More

• Lessons from Hims & Hers: Key Success Factors in DTC Telehealth: Read More

💡 Prompt Smarter

Did you know: you can ask an AI model to role-play as a skeptical investor reading your last earnings update or performance recap, and ask it to poke holes in the story. It’s a fast way to pressure-test whether your “growth story” actually holds up, or whether you’re just telling yourself the version you want to be true.

If you’re curious to learn more about how Ground’s AI capabilities can help take your DTC brand to the next level, let’s chat.

Finding Common Ground

The Ground community comes together sharing different stories, but common ground.