Today I have two stocks I bought Friday just before close and I will explain why, at what price and when I would sell. I also briefly cover some others I am watching for next week.

MP Materials (MP)

The first stock I bought is MP Materials, ticker MP. If you have not heard of them, well as the name suggests, they are in the materials sector. They operate rare-earth element mines and processing in the US and produce many of the materials essential for magnets.

They are still relatively early in this but the tailwinds are strong:

• They have a Pentagon price-protection deal essentially underwriting their pricing which contributed 14% of Q2 revenue and it’s an area the US government is actively investing in with a 15% stake in the business.

• The US government is actively seeking to move it supply chain back on to US soil rather than China and is starting to create legislation to do this.

• Demand for magnets is likely to explode over the next few years with robotics.

Q2 revenue was $109M, up 89%, and with NdPr oxide and metal sales up 277%. Adjusted EBITDA swung to $28.5M from -$12.5M and gross margin was 68.7%. They are also opening a $1.25B new magnet campus and pretty much cover the full process from mine to magnet now.

Materials is also a bit of a forgotten sector recently, cooling off after a hard run earlier in the year and those are the sort of places I like to buy. Inside materials, MP is arguably one of the leaders too which again is what I like to buy.

They are currently have a market cap around $11B so you could argue they are quite highly valued looking at the numbers but you are buying this stock for the future potential not today.

The stock price ATH is ~$100 in October 2025 and since then it has traded between $40 and $80 as you can see below. Today, the stock trades for about $56.

Of course, as with most stocks I like to buy, the technicals look good as well as the fundamentals and the narrative around the stock.

Forgive me for the busy chart, but on the weekly chart we have strong confluence of moving averages in the mid-$50s, with the 9W, 21W and 50W all sat just under the price at $54-$55. We also have the 9W just about to form a bullish cross with the 21W after a bounce around the 200W point about a month ago and a run up since. Today, the stock closed the week just on top of these averages and I think they make a great place to open a trade around.

The daily chart shows a similar pattern too - very strong confluence of averages which the price is sat right on top of. Again these averages are right around the mid-$50 point which makes me think it could be a great support level and you can see the stock has repeatedly bounced here over the last couple of weeks. I also like the dropping volume on the red days too.

So based on all of this, I opened a position today just around close, buying at $56.

While I like the fundamentals too, I still see this as mainly a swing trade position and therefore I like to set sell criteria when I buy too to try and keep emotions out of things. I like to use two ‘stop losses’ - one as a disaster worse case that will trigger automatically in the typical stop loss fashion and then a ‘manual’ stop where I will sell myself if we close below my defined levels. I set my ‘manual’ stop to a close 0.5x ATR below my support level, in this case a daily close below $52, and I set my hard stop at $49.50 for a max loss of ~12%.

Subscribe now

Voyager Technologies (VOYG)

The second stock I bought was VOYG which is essentially a defence and space business, another sector I like lots right now.

I am not even going to go into the fundamentals here because does such a good job in his report below and I recommend you check it out, plus the rest of his writing.

Wealthmatica's Substack

Voyager Technologies: Building America’s Next Space Infrastructure Platform

Space is no longer just a frontier for exploration. It is becoming a critical layer of global infrastructure. The space economy has truly been unlocked, from supporting satellite networks, defence systems, communication, manufacturing and data processing. The U.S. Space economy alone is projected to be worth $2 trillion by 2030 growing at nearly 8% a ye…

Read more

2 months ago · 37 likes · 9 comments · Wealthmatica

In a nutshell though - they’re a strong business spanning two strong sectors with a load of potential.

The market is starting to recognise this potential and the stock is up 52% in the last month (basically that most recent big green candle) but only ~12% in the last year as it has chopped around.

Now looking at the technicals, where I can add a bit more value, this is what I like the most.

Similarly to MP above, we have confluence of several averages on the weekly chart right below today’s closing price (100W and 200W do not exist yet). It did close just below the 9W but right on top of the 21W and 50W which I care about much more. Buying strong breakouts on retests of key averages, 50W in particularly, is one of my favourite setups to buy and this is exactly it.

On the daily chart we have confluence again with the 100D and 200D sat right on top of each other and the price has crawled along them basically all week, holding beautifully. And look at that dropping volume, hardly anyone is selling here.

It is probably about as good of a setup as I could ask for…

As I said, I bought this today too around close, buying at $34. Stop losses I will approach in the same way as MP too and have set a hard stop here at $29.50 with a ‘manual’ stop on a close under $32.

What Else I am Watching

So those are the two stocks I have added to but a few others are on my watchlist too.

• AAOI - super strong growth if they execute well, lots of bullish momentum long term and sat right on its 200D average with the 50W just below, plus it is pretty near its recent low from earlier this month.

• SNDK and MU - I started buying these a couple of weeks ago and they have held up well vs other hardware names in the recent sell-off. I am bullish long term on memory and like this opportunity to keep scaling into them.

That is everything for now, thank you for reading!

Please subscribe and leave comments if you enjoyed - I would love to hear your thoughts.

JM

Leave a comment

• How the sector is doing and wider market is doing

• The narrative around the stock and sector

• The fundamentals of the business

• And how it looks technically too

I like to do this because sometimes I’m wrong about things so if I wait for multiple thing to line up, I think it reduces my chance of being wrong overall…

Well at the moment there are one, maybe two stocks that I think meet all four of the criteria above and I have been building a position in both. One of them I am writing about today and the other I am sure will feature in a future article soon.

The business I am writing about today I genuinely think has potential to become one of the biggest in the world one day.

Of course, this stock comes with risks and I could be wrong, there is no such thing as a free lunch in the market, but I like the odds. Below is why…

Read more

Below I will cover the highlights from the earnings release and call, give my thoughts on them and finally my position today.

The Numbers

• Revenue: $443M (+43.5% YoY) - beat guidance midpoint by ~5%

• Adjusted EBITDA: $92M - 20.7% margin, +170bps YoY

• GAAP Net Income: $8.2M and EPS of $0.03 - finally positive! Vs expectations of -$0.03

• Free Cash Flow: $58M (+73% YoY) - 13.1% margin

• Operating cash flow: $69M, +65%

• Super-Scaled Customers: 197 (+17%) and ARPU of $1.8M (+17%)

• Rule of 64 or 49 stripping out M&A - this does use adj. EBITDA though so take it with a grain of salt

So great results across the board although honestly I would expect nothing less. GAAP profitability is the big one for me - I had long hoped and thought they would reach it early and I am very happy to see it.

I am also very happy to see adj. EBITDA margin expansion and strong FCF. Everything is trending in the right direction.

As is almost customary with ZETA now, guidance was raised across the board:

• Q3:

  • $469-472M revenue (+39-40%)

  • Adj. EBITDA $115-116M at a 24.4-24.7% margin

• FY26

  • Revenue: $1,811-1,824M

  • Adj. EBITDA: $404-406M

  • FCF: $255M representing 55% growth

  • GAAP EPS: $0.09-0.11 up from prior guide of $0.02-0.04

That last one in particular is a big increase.

What I Liked

Well quite a lot actually!

GAAP profitable finally

I flagged in my Q1 write-up that GAAP profitability was the thing I most wanted to see and we finally got it. For some people and funds this alone would stop them buying so it’s a nice milestone to get out of the way and get to see operating leverage kicking in. Plus, a quarter earlier than I expected and they raised the full year guide substantially on the back of it.

Free cash is flowing…

They reported $58M in Q2 up 73% and the margin increased from 10.9% to 13.1%. The engine is running and it looks they are starting to become the cash flowing compounder I had hoped they would.

Super-scaled customers and ARPU both growing well

This is the flywheel in effect - more customers, each spending more and evidence that Athena is making a difference. It is the evidence that the platform is genuinely getting stickier rather than just bigger.

Adjusted EBITDA outgrew revenue

Adj. EBITDA grew 56% while revenue growth was 44%. Again, a sign operating leverage is finally kicking in.

Athena is working

After Q1 I said the Athena adoption figures were impressive but I needed to see more before I got too excited. I think we got it on the call this quarter. Athena has been live 130 days now and over 40% of super-scaled customers are monthly active users. They are accounting for the majority of revenue, they grew four times faster than everyone else and their NRR is 400bps above the company average. This shows me that Athena is a genuine game changer and everything a shareholder could want.

Oh and finally, management says clients using Athena have seen returns on marketing spend of up to 1400%.

They are buying back lots of stock

Zeta repurchased more than 75% of their FCF in shares last quarter and said they consider buying their own shares the best available use of cash at current prices. They also closed a $1B credit facility on 27th July which gives them the option to accelerate this if the stock is cheap. Given they mentioned avoiding large transformative acquisitions, this looks a possibility.

The Partnerships - Palantir, OpenAI and Snowflake

I was going to put this in the good section above but there was so much to write I am giving it its own section. This is one of the most exciting parts of this quarter’s results for me.

I’ll go through each partnership individually.

Palantir

First of all, this has moved faster than I expected. When they first announced a partnership in June they talked about a 30 to 45 day integration. Yet, on the call this week , Steinberg confirmed the data cloud was fully integrated with Foundry as of 31st July and they have already closed two deals off the back of it with a 100% hit rate. Talk about execution…

He described it as possibly the biggest opportunity the company has ever had in front of it and I can see why he would think that.

Foundry is essentially the plumbing large organisations use to make sense of messy data. If Zeta's identity graph and Athena sit on top of that, Zeta stops being a marketing tool a company buys and starts being part of the infrastructure it already runs on. It also gives Zeta an entry into Palantir's customer base which includes a lot of large advertisers they would otherwise struggle to get in front of.

And say what you want about PLTR as a stock but the business is executing fantastically. Those Q2 results were magnificent and if there’s one business you’d want ZETA to be in partnership with, I could not think of many better ones.

Finally, none of this is in guidance of course.

OpenAI

I was sceptical around this one I’ll be honest - I had assumed this was essentially a voice engine deal and people were getting carried away. And essentially that is what it started as, Athena's voice runs on OpenAI's models whilst the data cloud runs on Zeta's own models so the data stays internal.

But there's a second half to it that is much bigger. Zeta is now actively serving advertising on OpenAI's ad platform. It’s real revenue, coming from one of the world’s leading AI companies at a time when the world is concerned traditional advertising methods could be disrupted.

Snowflake

This is less glamorous but possibly the most commercially obvious. Many, many enterprises already keep their customer data in Snowflake and if Zeta can work with the data there rather than asking a client to move it, you remove one of the biggest objections to sales.

Why all three matter

The common theme is that Zeta now targets CIOs and CTOs alongside CMOs. It is a genuine strategic change.

Marketing budgets are the first thing to cut in a downturn and IT infrastructure budgets are among the last. If Zeta successfully repositions from marketing cloud to AI infrastructure, it isn't just a bigger addressable market, it's also more defensible and less cyclical. Plus, being associated with Palantir and OpenAI does a lot of the credibility work in that repositioning for free.

I have also seen them talking about ZBI now, a business intelligence layer combining Zeta and client data for things like inventory management. This is obviously not marketing at all which I think tells you where they think this is heading.

The Bad

While there was lots of good, of course there were some things that I liked less too…

Organic growth has stopped accelerating

I think this is what could have spooked some people initially. Back out M&A and political revenue and the organic growth was roughly 28% this quarter vs 29% in Q1. Ok it’s still high and not a massive drop but after four straight quarters of acceleration it’s a change.

Looking at guidance, the Q3 guide implies 23% organic and the full year 25%.

There is also an element of cyclicality with this too but it’s worth keeping an eye on.

Marigold went backwards

Recently acquired Marigold contributed $55.6M in Q1 and $48.1M in Q2, possibly made worse when people could be concerned about organic growth.

It could be seasonality and one quarter isn't a trend but again something I want to watch.

Gross margin went the wrong way

Cost of revenue came in at 41%, which is 300bps worse than a year ago. It was attributed to agencies putting more spend through social channels where the economics are worse.

I flagged this as a risk in my Q1 write-up and it has now turned up in the numbers rather than staying theoretical. The adj. EBITDA margin still expanded so the operating leverage below the gross line is compensating for it but it does put a ceiling on how good this business can eventually look if the mix doesn't shift back.

Overall it isn’t that bad

Honestly writing this, I don’t think I can look at it as anything but a great quarter. I found some things I don’t like but none are thesis breakers just things to watch.

Closing Thoughts

So yes, overall this was a good quarter with a genuine milestone in it.

Zeta is generating real cash, winning bigger customers, spending more with them, and has now crossed into GAAP profitability. The partnerships look promising and there is a real bull case forming where Zeta becomes much more than a marketing business.

There are small things to watch but none concern me particularly.

I have not touched on the valuation here but that is the next thing to consider. As I write this ZETA trades around $25 which is arguably still cheap if you consider what could be possible.

For me personally, right now I'm holding and after this recent pop it is my biggest position at somewhere around 20% of my portfolio. Simply because of the size I have high criteria to add now but would not be against it if we dip towards $20, certainly not if we go below.

I will end this here but will keep posting about Zeta so please subscribe if you would like more like this and thank you for reading.

JM

Subscribe

Back last November I published a deep dive on Shift4 Payments (FOUR) calling it the most asymmetric fintech opportunity on the market. The stock was around $70 then, down from all-time highs around $125. Today, the stock sits around $50 having spent most of the year in the $40s and dipping as low as ~$35.

This creates the question, was I wrong or just early?

That is what I want to review in this article. By cost basis, FOUR is one of my larger positions and when one goes this far against you, I see only three options: re-examine the thesis, admit you’re just hoping and got something wrong or bury your head in the sand.

This is me doing the former - a full update covering Q1 2026 results, what’s changed since November, the bull & bear case, valuation from several angles, macro, technicals and finally what I’m actually doing with my position.

Before we get into it, if you haven’t read the original deep dive, you don’t need to for this article, but it’s below if you want the full background. I will remove the paywall for a month so that everyone can see it.

Subscribe

Q1 2026: The Headline Numbers

Shift4 reported Q1 results on 7th May and operationally it was a decent quarter, broadly in line with or ahead of guidance, but not a knockout...

Here are some of the key numbers:

• Gross Revenue: $1.12B (+32% YoY) - ahead of guidance and consensus

• Gross Revenue Less Network Fees (GRLNF): $549M (+49% YoY) - in line with the $548M guided

• Gross profit: $370M (+54% YoY)

• Organic GRLNF growth: +11% YoY, includes a ~400bps drag from legacy revenue streams they are deliberately winding down

• Payment Volume: $56B (+24% YoY)

• Adjusted EBITDA: $234M (+39% YoY) - 43% margin, in line with guidance

• Adjusted Free Cash Flow: $88M (+26% YoY) - ahead of the $70M guided, with 98% conversion of non-GAAP EPS

• Non-GAAP EPS: $0.97 - roughly in line with consensus

• GAAP Net Income: $12M, although after $16M of preferred dividends, diluted EPS came in at -$0.01

• Buybacks: 5.5M shares repurchased for $295M - $600M of the $1B authorisation executed

The growth was broken down further below and a few things stood out to me.

International is where the growth is: payments-based RLNF grew 25% overall, but split that out and the Americas grew 15% while everything outside the Americas grew 51%. Management also said these results absorbed a $4-6M headwind from the Middle East conflict disrupting travel, which matters because the Global Blue’s tax-free shopping business is directly exposed to international travel of course.

The other thing is that organic growth… 11% is not great, less than the 18% in Q3 2025 and supports the bear thesis that much of the growth is essentially being acquired. Alone it is not necessarily a deal breaker, more a yellow flag, but if the ROI for those acquisitions isn’t good then things start to look worrying. I’ll talk about this more later.

The full year 2026 guidance was reaffirmed:

• Volume: $240-260B

• GRLNF: $2.5-2.6B (+26-31% YoY)

• Adjusted EBITDA: $1.165-1.215B (~47% margin)

• Non-GAAP EPS: $5.50-5.70

• Adjusted FCF: $490-510M

For Q2 they guided GRLNF of $615M, adjusted EBITDA of $278M and adjusted FCF of just $10M. This sounds alarming but includes ~$20M of travel disruption headwinds and the seasonality of tax-free shopping, which consumes cash in the first half of the year and generates it in the second half. Q3 guidance is back to normal with an adj. FCF range of $185-195M.

So the business did what it said it would do roughly, with some red, or at least yellow flags too.

What’s Changed Since My Deep Dive

A lot has actually happened since November when I wrote my deep dive and most of it is genuinely positive, even if the share price says otherwise.

The Up-C Collapse - A Big Governance Win

In Q1, Shift4 collapsed its multi-class share structure into a single class of common stock. Founder Jared Isaacman’s super-voting B and C shares were converted into Class A ones, Shift4 is no longer a “controlled company” under NYSE rules and an estimated $440M of potential future Tax Receivable Agreement (TRA) liabilities were eliminated. In exchange, Isaacman received around $139M of cash (from previously paid tax distributions) plus mandatory convertible preferred shares in a private placement.

I think this is significant and good news - one share class, one vote per share, no founder veto, no TRA dragging on future cash flows. The stock actually jumped over 10% when this was announced in February. Governance was a criticism of FOUR for years and gets resolved here.

Bambora

On 2nd March, Shift4 closed the acquisition of Worldline’s North American business (Bambora) for approximately $92M in cash, adding over 140,000 merchants across the US and Canada. The price looks small relative to the merchant base but this is the classic Shift4 playbook of buying payment volume cheaply and migrating it onto their own rails. Time will tell on execution, but at $92M it’s a low-risk bet compared to Global Blue.

Lots of Insider Buying

Jared Isaacman, who stepped back from the CEO role to chair NASA but remains FOUR’s largest shareholder, has been buying stock aggressively in the open market all year and he’s bought a lot

• ~$13.7M in late February around $46

• ~$2M in early March around $44

• ~$2M more in mid-March around $46

• ~$15.9M in mid-May around $41

That’s over $33M of open-market purchases this year alone, on top of the ~$16M he bought around $83 last year. He’s either going to become the king of catching the falling knife or a genius… Meanwhile the company itself has now repurchased 19.2M shares since IPO - 19% of the shares outstanding - including $600M of the current $1B authorisation.

When the largest shareholder and the company are both buying this hard while the price makes new lows it makes you look twice.

The World Cup & New Wins

The 2026 World Cup kicked off across North America in June and Shift4 has spent the last few quarters getting venues and hotels ready for it. So many of those extra visitors and that extra spending, all going through Shift4 systems…

They’ve also announced new partnerships recently with the Chicago Cubs and Inter Miami CF, a stablecoin acceptance partnership with Lydian adding USDT to their Pay with Crypto product and a voice AI integration for restaurants via Loman AI. They are still working hard and the engine keeps running.

The Organic Growth Question

Before we go further onto the bull and bear cases, I want to deal with what I see as on of the most important numbers reported and one people frequently quote: the organic growth.

Shift4 has only ever disclosed a quarterly organic GRLNF figure twice:

• Q3 2025: +18% YoY - the first disclosure, made largely to answer the “only grow by buying companies” criticism although whether it did this or made it worse is to be decided…

• Q4 2025: no clean figure - They didn’t disclose it the next quarter which was a communication misstep in my opinion and raised concerns. The nearest proxy management gave was that excluding Global Blue and Smartpay, GRLNF grew roughly 23% for the full year 2025. That’s a full-year number though, not a like-for-like quarterly one.

• Q1 2026: +11% YoY - this does include a roughly 400 basis point drag from intentionally deprecated revenue streams so add that back and you’re at ~15% but is that enough?

So the sequence is 18% → nothing → 11% (or ~15% adjusted).

Why the decline? Here are some of the reasons I can find:

1. Americas restaurant softness - Management has talked since Q3 2025 about weaker same-store sales among Americas SMBs and numbers confirm this, the Americas payments book grew 15% versus 51% worldwide ex-Americas. The mature, domestic, consumer-cyclical part is slowing.

2. Tax-free shopping is barely growing - TFS grew just 4% on a pro forma basis, hit by the travel disruption discussed above. Whether this is temporary or not is a question to be answered however probably not in Q2 either.

3. The metric itself is somewhat flawed - This is a point I have seen bears raise and I kind of agree with them. “Organic” strips out acquired revenue but a meaningful chunk of what’s left is still cross-sell into acquired merchant bases, so the organic number is still partly a function of past M&A. As the cross-sell runway from older deals matures, the reported organic rate drifts toward the true underlying rate, which is indexed to global consumer spending and is almost certainly a lower than 15%.

Ultimately, I think the deceleration is part genuine and part presentation. As a bull you could say it is “~15% adjusted, with self-inflicted noise” and as a bear “18 to 11 is a trend with the terminal rate lower than either.”

As for analysts:

• Goldman Sachs sit in the second camp, arguing underlying organic growth is really high single digits.

• Loop Capital initiated at Hold with a $40 target in June citing exactly this 11% figure.

Honestly, this is one of my biggest worries and how you view this changes a lot about the stock’s value.

The Bull Case

So let’s look at the bull side of the thesis, why do I still own FOUR?

The valuation is extreme

I’ll do this properly below, but the short version is a business guiding to 26-31% GRLNF growth this year trades at under 10x forward earnings. That is pricing for something disaster, even after the recent run up.

Diversification is real

Restaurants, hotels, stadiums, luxury retail, tax-free shopping and gaming, across 75+ countries with over 80,000 merchants outside the Americas. They are all over. Plus international payments revenue is growing 51% YoY and shows the model can work worldwide.

Deleveraging is underway

Pro forma net leverage was 3.7x at Q1 and management expects roughly half a turn of deleveraging per quarter to finish the year in the low 3s. Yes, it could be better and yes it is a lot but the direction matters most for me here. They also repriced nearly $1B of term loans at lower margins earlier this year. Should they succeed with this deleveraging, the stock will look completely different.

Alignment

Cleaner share structure, founder buying tens of millions in the open market, large buybacks. Whatever else you think about this company, incentives now point the same way as shareholders.

The cash flow target

Honestly this sort of sits in both the bullish and bearish camp and is the crux of the uncertainty around this stock.

At the February 2025 Investor Day, management set a target of $1B in adj. FCF on a Q4 2027 annualised exit-rate basis and as recently as Q3 2025 said they were on pace to exceed it. Then on the Q4 2025 call the framing shifted. CFO Chris Cruz said that given the buyback opportunity, the goal was now better viewed on a per-share basis through the lens of a long-term owner of the business.

I’ll be honest about how to read that, but it is a softening. To hit $1B annualized by Q4 2027 the company needs roughly $250M of adjusted FCF in a single quarter. Q4 2025 did $171M and 2026 is guided to ~$500M for the full year, essentially flat on 2025. So the absolute-dollar path got harder and moving the goalposts to “per share” conveniently changes the scoreboard at the same time.

That said, the buybacks are real and aggressive. The share count has already fallen from around 90M to the high-70s/low-80s and management has been clear that compounding FCF per share is how they think about durability. Even if absolute FCF lands around $740M in 2027 (roughly where the sell-side sits) on ~78M shares, that’s about $9.50 of FCF per share. At $50, you’re paying a bit over 5x that. For a business still growing GRLNF at 26-31%, that’s cheap but of course not certain.

Subscribe

The Bear Case

I always want to give the risks their due and with the stock down this much, this really deserves attention.

Honestly, if you want the most thorough bearish take on FOUR I’ve read, this article from Bearstone in December is excellent and I’d genuinely recommend it (I said so in his comments at the time):

Bearstone

Shift4: Valuation and Personal Opinion

We have three baskets for investing: yes, no, and too hard to understand…

Read more

8 months ago · 47 likes · 24 comments · Bear

The core arguments, as I’d summarise them with my responses:

The “hidden CAC” argument

Shift4 reports a tiny advertising and marketing line while growing fast. The bear view is that the real customer acquisition cost lives elsewhere, in acquisitions and in free hardware they capitalise. This means the headline margins and FCF flatter the underlying economics and the growth engine has historically been funded by debt rather than internally generated cash.

Well this links somewhat to the organic growth discussion above and is probably the strongest bear point, plus it’s partially right. Shift4’s growth model is capital-hungry, and the reported numbers do look better than a fully expensed version would.

Where I differ is on the conclusion. The acquisitions haven’t just bought revenue, they’ve also bought distribution, licences and whole verticals (Finaro, Vectron, Global Blue, now Bambora). And now, management has now explicitly committed to deleveraging rather than continued large-scale M&A.

2026 is the year where we find out if this works and if the machine can self-fund. If outflows still dwarf operating cash flow by year end, the bears win this one and the stock probably deserves to be trading where it is.

Organic growth is decelerating

Again covered above and again true really… I will not repeat myself but if this keeps falling then again, the multiple the market is applying isn’t as crazy as it looks.

The debt

Total debt was $4.52B at Q1 against $473M of cash, so net debt around $4.0B - bigger than the entire market cap.

Interest expense was $65M in Q1 alone and there’s roughly $1B+ of 6% mandatory convertible preferred stock on top. Plus, the 2027 convertible notes carry a conversion price around $115 which at ~$50 they are nowhere near converting, so realistically that gets repaid or refinanced with cash.

Leverage is one of the biggest reasons this stock trades where it does but also one of the enablers for FOUR to become the huge business it wants to be. Whether it works or not is to be determined but its a high risk strategy. Finally, in a higher-yield world (more on that below), the market is punishing levered balance sheets hard.

Governance

The 2023 Blue Orca short report raised related-party transaction and accounting concerns around the founder era. I won’t go over them all here and it is discussed in the Bearstone piece but this was a cloud hanging over the business. Today, I would argue the Up-C collapse, the end of controlled-company status and Isaacman’s repeated open-market buying are about the strongest practical responses to those concerns you could ask for so I am less concerned about this than the other reasons.

Cyclicality

Restaurants, hotels, stadiums and luxury travel retail are all consumer discretionary businesses at the end of the day and a slow down in these will impact FOUR. Management has flagged softer SMB restaurant trends in the Americas since Q3 last year and the Middle East conflict is directly hitting the travel Global Blue depends on. In a proper recession, volumes, churn and take rates will all get hit at once, exactly when exactly when a levered balance sheet is least forgiving, and there is nothing Shift4 can do about this.

My Overall Opinion

The bears have been more right than the bulls on the stock over the last eight months and partially right on organic growth. I am sitting on a big unrealised loss and many of the concerns raised have not been sorted since November. It would be naive not to recognise this even if most Substack writers tend to only publish their wins.

Where I still think (and hope) the bears are wrong is on the terminal value of the business and on management's capital allocation however this needs to be proven. This is not a low-risk setup, it's a leveraged one where the equity is a coiled spring in both directions.

Macro: A Challenging Environment

A large part of FOUR’s decline this year isn’t company-specific at all to discuss it. The macro backdrop turned hostile for a stock with their profile…

Inflation has re-accelerated: April CPI ran at an annualised 3.8%, the hottest since 2023, core PCE is back at 3.3% and the conflict in the Middle East has pushed bond yields higher. The 10-year has been above 4.5% and the 30-year above 5%, levels not seen in years. The Fed is on hold at 3.50-3.75% and the conversation has shifted from when do they cut to might they have to hike, with a new Chair now in the seat. Q1 GDP was revised down to 1.6% and consumer sentiment hit a record low in May.

Map that onto Shift4 specifically:

• Higher yields punish leverage - with ~$4B of net debt, FOUR is exactly the kind of stock that gets sold first when yields spike. This is a dominant driver of the de-rating in my view - the June low came days after the May jobs report drove the 10-year through 4.5%.

• The conflict hits travel directly - management quantified a $4-6M Q1 impact and embedded ~$20M into Q2 guidance from reduced flight capacity into Europe from the Gulf and East Asia, flowing straight into Global Blue’s tax-free shopping volumes. If this drags on, Q3 could be worse.

• A weak consumer hits Shift 4 - so far the data is holding up better than feared and management said same-store sales in restaurants and lodging came in slightly better than expected in Q1, with stable trends through April, but record-low sentiment is not a comfortable backdrop.

• One big tailwind is the World Cup - the tournament will have driven traffic through the stadiums, venues and hotels Shift4 serves. Truist explicitly flagged World Cup volumes as an incremental tailwind when raising their EPS estimates in June. It won’t transform the year alone but it lands in Q2/Q3 precisely when the travel headwinds are biting.

None of this breaks the long-term thesis but it does help to explain the price action. And note what the last three weeks demonstrated - when the debt fear was addressed somewhat, the multiple immediately started repairing itself.

The section below is for paid subscribers only and covers the valuation, what to expect from Q2 earnings and what I am doing personally. If this interests you please consider a paid subscription.

Subscribe now

Read more

Well first things first, the macro environment seems to be moving a lot of stocks together recently and how that is tomorrow will probably determine how successful a lot of these picks are realistically.

As I said yesterday, the S&P and Nasdaq are both at key support levels and we are at a bit of a make or break point.

Should we hold and possibly see some strength though, I think the picks below could be good both as nice entries for longer term holds or as swing trades.

Micron Technology (MU)

I have to start here since I literally posted a note about buying this on Friday around close.

MU is holding the 50D EMA and memory has been extremely strong recently, both of which make me believe this is a good stock to buy here.

Personally, I have a stop loss around $880 where my breakeven price is but ATR is ~$80 so this could easily be triggered. If you are happier to take on some more risk, a stop around $790 puts you under the recent lows, outside 1.5xATR and under any support at $800.

Applied Optoelectronics (AAOI)

Anyone who read the post above will also see I bought these on Friday too when they held $100 and the 200D EMA.

After hours they dipped under the $100 mark occasionally but I think this is a solid place to buy. How the broader market reacts here will really make or break this trade but I think if we are to see some green, this could be a great one to own. To reclaim the ATH from here is over a 100% gain.

Volatility is high and ATR is ~$15 which is a bit high for me to risk really. Instead I am playing this one with a tight stop around $97 which basically means I will be out on any downwards movement however, the loss will hopefully be very small and I am ready to take part in any movement upwards - let’s see.

Marvell Technology (MRVL)

This is very similar to the last two and honestly will probably trade very similarly so you could just pick your favourite.

Holding the 100D EMA and around the level it spent a while at before the big pop in June.

Also holding on the 21W and did a good job of it all of last week.

Again, this is quite volatile with an ATR around $20 or over 10% of the current price but once again I would probably play it safe with a much tighter stop under the 100D. The low of the week was ~$178 so a stop around $175 might work.

To me this is basically a safer, less volatile version of the AAOI trade above. Personally, as a core position I would go with this but for a swing trade I chose AAOI.

Robinhood Markets (HOOD)

Moving away from the AI sector, HOOD looks interesting here. We have both the 100D and 200D basically stacking on top of each other at ~$94 and ~$92 respectively with the current price holding just above it. It also held up quite well over the last week compared to a lot of names.

Similarly, we have a couple of support points on the weekly chart with the 21 and 50W just under the current price. Retests of the 50W MA are some of my favourite points to buy and hopefully this is no exception.

If I were to buy here I would probably do so with a stop loss just under $90 - ATR is ~$6 so you might want to go slightly lower but I am being quite risk averse in this current market.

Subscribe now

That’s it for today, short and sweet but maybe some ideas for the coming week. I’ll be posting lots more in notes throughout too and updating how these ones do.

Thanks for reading

JM

As before, this covers my performance for the quarter, the winners and losers, the main position changes and the lessons I’m taking away.

Headline Performance and Thoughts

After a rough Q1 that finished at -16.5% YTD, Q2 was a very different story.

The portfolio returned +26.4% for the quarter, pulling my YTD performance to +7.1% and comfortably my strongest quarter since I started posting online.

It wasn’t a smooth ride the whole way. There was a sharp market-wide pullback mid-quarter with a lot of volatility, particularly in the high beta names that make up a big chunk of the portfolio, but the businesses I own largely delivered. And this time, the discipline of holding my winners through the noise was rewarded rather than punished.

The quarter was defined by earnings season. I went through a huge number of results across my holdings, arguably too many to keep on top of, and made real-time decisions on most of them. Some confirmed my thesis, a few tested it and one or two changed my mind.

The mid-quarter pullback was the key test. Coming off Q1, I was determined not to repeat the mistake of holding no buffer and not taking risk off the table. This time I trimmed into strength ahead of the worst of it, held more cash through the drop and was able to add to my highest-conviction names near lows or at key support.

By the end of the quarter the portfolio looked very different to how it started. What began as a fintech and insurance heavy book with a few painful laggards is now a much more deliberate portfolio with a much stronger weighting to higher momentum sectors like AI infrastructure, photonics and space.

The Portfolio as of 1st July

Last quarter I said I’d start sharing more detail on the portfolio itself so here is the full book as it stood at the end of Q2, summed up nicely by a handy graphic Claude made for me.

Let’s discuss a few of the key positions and themes…

The Core

Nearly half of the portfolio is concentrated in the top three positions and I see these as the core performance drivers. All three are stocks I am very bullish about long-term and they are also fairly uncorrelated with each other, focusing on different themes. This was shown recently when we had the day below, showing how ZETA and DLO did well while NBIS sold off heavily.

Weighting wise, the order of them swaps frequently and I would happily add to any on a pullback - I added to ZETA and NBIS in Q2.

The ‘Lagging’ Conviction Holds

FOUR and ROOT are my 4th and 5th biggest positions and both are down from my average cost, even after decent performances in Q2.

Despite wanting to sell at times, I have held onto these because I think the fundamentals justify it, even if the market was disagreeing until lately. I still think both of these businesses and stocks have huge potential and I am looking forward to this being realised eventually.

For now, I will not add to these but I may do if they start to pick up momentum and there is spare capital.

The Mid-Sized Positions

There are a few stocks with positions around mid-single digits: RDDT, UNH, RKLB and OSCR - all doing ok but not necessarily big enough to drive a lot of performance.

There’s also two separate stories here really. UNH and OSCR I would be happy to sell soon if momentum slows, particularly UNH. RDDT and RKLB though I am actively buying on pullbacks and either could be a top position eventually, especially RDDT.

The Others…

The rest of the portfolio is made of several smaller positions across a few themes, many of which I am happy to buy or sell in and out of.

AI Infrastructure & Photonics:

SOI and LPKF are the core of this theme, with AAOI and PENG as smaller, higher-beta versions. Both SOI and LPK are key suppliers to the future hardware cycle and yet because of their legacy core businesses, have not been fully recognised yet. PENG also has some exposure to the memory theme.

Energy:

I am convinced energy will be one of the biggest bottlenecks over the next few years and it’s an area I have been looking at recently.

ENPH and SEDG are two of the names I have been looking at and while their core businesses are struggling, both have potential in the data centre space. I am also considering FLNC for a similar reason. All three also look interesting technically too.

Right now though these are sized so small that they are almost meaningless. Soon I will either build a larger position in one of them or sell entirely.

Space:

Space is another area I have been interested in lately, particularly as it’s sold off after the SpaceX IPO.

In particular I’ve been buying LUNR and RKLB and like both from a fundamental and technical perspective enough to build small positions. RKLB especially I am happy to own a bit more of.

Defence & Rare Earth Minerals:

The other two areas I have been buying are rare earth materials with USAR and defence with ONDS. Both of these are themes I once again see catching a bid soon but honestly both positions are doing badly right now. My ONDS entry in particular was especially bad but I am holding for now.

Subscribe

Position Changes

This was an active quarter and around a quarter of the portfolio by weight was turned over. I was also good at keeping a nice cash buffer this quarter which allowed me to hop in and out of some shorter term trades with some success.

Sales

PGY:

I finally exited after five consecutive profitable quarters that the market completely ignored. Guidance was being raised, net income was growing and the stock kept falling. I liked the business but the market didn’t and technically it looked very bad. Add on that I did not understand the business fully, the relatively small position size and better looking opportunities and I decided it was time to sell.

MSFT:

I like Microsoft as a business but after a small bounce it seemed to have lost momentum and I felt like upside was limited. In the end I trimmed some then exited a week or so later.

MELI:

My reasons for selling MELI were similar to MSFT above - fantastic business but the market was not rewarding it and the upside, at least short term, felt limited. I also have a lot of emerging market exposure in DLO which I prefer to MELI and so selling seemed right when I needed capital.

Clear out of smaller positions from Q1:

SPGI, SAP, ORCL, NVO, PATH, LMND - all names I owned at some point in Q1 and Q2 that I don’t now. Essentially, none of them I had strong enough conviction in to justify the capital while better opportunities were in front of me.

AI Swing Trades:

GFS, SIVE, NOK, BE.

These were all more like swing trades on the AI momentum than long term holds and I have sold all now.

SIVE and GFS did very well for me and I timed selling well whereas with NOK and BE I was probably late to buy but I didn’t lose money either.

Buys

NBIS:

One of my favourite businesses and one I have been averaging up into for over a year now after first buying around $30. This quarter I added around $200 on a pullback and ultimately increasing my position by about 20%.

Sizing up a winner into weakness, rather than trimming it into strength was probably the single best decision I made this quarter.

ZETA:

Similar story to NBIS and not much else to say. Terrific business, lots of potential and I’ll keep on adding to this one. This quarter I added on the post-earnings pullback.

RDDT:

One of my new favourites and probably where the most new capital went this quarter. I bought on a 100-week moving average retest and added more on a few other retests of various support points.

LPKF:

The one I’ve written about a bit but never properly. If you don’t know them, it makes the only laser precise enough to drill the tiny holes needed to replace copper wiring with glass in AI chips. Major chipmakers are testing it but nobody has ordered at scale yet and market cap is sub $1B.

Others:

I have opened smaller postions in a few new businesses this quarter, mainly as swing trades in sectors I believe in bought when I think the technicals look good and momentum is strong.

• ENPH & SEDG for energy

• LUNR & RKLB for space

• AAOI & PENG for AI infrastructure

• USAR & ONDS for rare earths and defence

Being honest I don’t always own all these at the same time and while I think they all have potentially, long-term there are definitely some I prefer more than others.

Top Winners & Losers

As I did last time, here I want to review the top winners and losers in the portfolio, not just based on the percentage change in the stock price but on position size too.

Winners

Several of the stocks I own did very well this quarter. Of the ones I held all quarter, we had the following gains:

• NBIS ~125%

• OSCR ~100%

• UNH ~56%

• ROOT ~39%

• ZETA ~33%

• DLO ~18%

• FOUR ~15%

There were also several others traded in and out of with some success like SOI, where I sold much of the position for over 100% gains, SIVE and then RDDT where I own a core position but have traded a bit on top.

Of course sizing matters and so accounting for that, the biggest gain by far came from NBIS, which was around a 10% position at the start of the quarter and contributed for around half of the quarter’s performance. The thesis keeps strengthening with fantastic Q2 results and my only regret, as in Q1, is not owning more sooner.

ZETA, which was around a 15% position, came next, contributing mid-single digit percentages of the portfolio in returns. UNH and OSCR both also did well on a portfolio level despite me trimming both in Q1.

SOI and DLO had the next biggest impacts - SOI returning near 100% on average on a mid-sized position and DLO returning less but benefiting from a large sizing. I also did well with GFS and LPK as swing trades on the momentum around AI hardware.

Finally, whilst my positions in both are still sitting on an unrealised loss, FOUR and ROOT did perform fairly well this quarter and both drove a few percentage points of performance each due to the position size.

Losers

There hasn’t actually been that many losers this quarter which is nice to write. The few there has been have been relatively small positions too, usually more swing trades with less capital than core positions. They are also all positions I bought during the quarter too.

My biggest loss is ONDS, down ~38% from my average price. I bought this fairly high as I thought it was breaking out with strong momentum and sector tailwinds right before it turned and started dropping. In hindsight I should have followed my system like I usually do and sold when it lost key support levels, but I really believed in this theme and honestly still do. The position size was not huge but big enough to knock around 1% off the portfolio’s performance. For now, I am still holding since it is small enough that I am happy to ride it out, regardless of how it goes.

The next biggest loss and a similar story is USAR, down ~20% from my average price. Again a sector I believe in, a stock I bought on what I thought was strong momentum and one I should have probably sold earlier. This position is smaller than ONDS and less of a loss stock wise, so the overall loss it is not too bad but it is worth reviewing all the same.

Finally, while most of the actual loss was incurred in Q1, I sold PGY at a big unrealised loss in Q2 to rotate the capital elsewhere. Whilst this worked well and there are plenty of wins, PGY had a great Q2 and was up ~60%.

Lessons Learnt

Before we end I’d like to cover some things I’ve learnt or at least remembered this quarter. I think this is a good self-improvement exercise and something I recommend trying.

Let Winners Run

The biggest lesson this quarter was the value of not selling my winners. Several times I was tempted to trim NBIS or SOI to raise cash because they had ran hard and every single time the better answer was to cut a laggard or an underweight position instead.

NBIS contributed more to the quarter than the next two holdings combined, and would have contributed a fraction of that had I kept ‘taking profit’ out of nervousness.

Winners run, laggards and small positions fund the next idea - that’s how I raise capital if I need to.

Size Matters As Much As Selection

This is somewhat obvious but it’s also worth reiterating at times. OSCR was my second best stock but my fourth-best contributor. FOUR was 8.5% of my portfolio and did relatively little. Getting the call right is only half of it, the other half is having enough of it on when you’re right and not too much on when you’re wrong.

I have heard a lot of successful investors say this and it’s really true.

Having a Cash Buffer is Nice

After Q1, where my supposed ‘ballast’ in UNH failed me in the sell-off, I promised to be more disciplined with cash and risk. This quarter I actually did, trimming into strength before the pullback on some of my lower conviction names and holding enough cash to buy the dip in the best ones.

There’s also a psychological benefit, at least for me, where I feel better about potential dips or crashing knowing I have some cash ready.

Concentration Is My Biggest Risk Now

The flip side of a great quarter is my top three positions are now just under half the portfolio, up from around 43% in April. Plus much of the portfolio is high beta and moves on the same headlines.

This is the result of focusing on my best ideas and the way I would have it but it’s a risk worth being aware of.

Final Thoughts

This has been a more enjoyable review to write than the last one and I have put a lot more effort in too, both for my benefit and hopefully for readers’.

Performance was good and I am happy with it but I am also happy with how the portfolio is structured and how I am executing my ideas.

As I said last quarter, much of Q1’s pain came from decisions made the year before. In the same way, I think the discipline shown this quarter sets up the rest of the year well. The main things on my mind for Q3 navigating what is probably now quite a hot market and deciding what to do with FOUR and ROOT as the biggest laggards but ones I have conviction in.

Subscribe

Thank you for reading and I hope you found this interesting. As always, none of this is financial advice, just my own journey, mistakes and all.

If you have suggestions for future articles, please leave them in the comments.

JM

This week saw a pullback in many stocks and a huge amount of volatility but, assuming the market doesn’t continue to pullback deeper, it has also created lots of interesting looking setups - let’s look at a few!

Before we start, I’ll apologise that I’m going to keep this one brief because I don’t have much time to write today but I want to make sure I get something out.

Intuitive Machines (LUNR)

I spent a lot of last week looking at the space sector and had my eyes on this stock. It hadn’t quite retested any support then but now it has and it looks very bullish.

First, we have a retest and hold of the 50W EMA (yellow line below).

But, not only do we have that, we have also had a successful test and bounce at the 200D SMA and with increased volume, a sign buyers are stepping up at this point.

I am bullish on this sector and assuming the market in general holds up next week, I think buying LUNR anywhere around $20 could be a fantastic buy. The ATH from here is over a 100% gain too should it reclaim it.

Personally, I already bought on Friday near close but I’ll be buying more on Monday to create a sizeable position. Since it will be quite a big position I’ll manage it with a stop loss around $16.50 or a daily close below the 200D at $18.50.

Subscribe

LPKF Laser & Electronics

I already wrote a note about this one on Friday because I did not want to wait until now to post about it, so I will start by quoting what I wrote there:

One name I am watching right now: LPKF

I’m not even going to wait until the weekend to write about this one. I’m buying today.

They make the only laser precise enough to drill the holes needed to replace copper wiring with glass in AI chips.

Every major chipmaker is testing their tech, but no one has ordered at scale yet.

The stock has ran a lot recently and been very volatilebut now it’s sitting right on its 50D moving average, exactly where it bounced just weeks ago.

I am buying here at €20.90 and betting on it holding again. Last time it did it ran up 50% over the next couple of weeks.

And just as I had hoped, the support did hold and I did buy at €20.90. I also had a limit order at €20.50 but unfortunately that did not fill so I will likely buy more on Monday.

I think this is one of the few AI hardware stocks still reasonably priced and at sub $1B market cap, I see a lot of potential. Bear in mind though that small caps like this come with a lot of volatility and reduced liquidity. I will likely sell on a close under the 50D but only if by a couple of €s since the price does not update as often here and when it does it can jump around.

Applied Optoelectronics (AAOI)

Those of you who read last week’s article will know I talked about this one there when it held its 50D EMA.

What I didn’t account for was a market wide pullback and that trade ended up closing for a small loss early in the week.

This is ok, it happens and as I mentioned back then, I sized it smaller because I expected volatility.

However, with that market pullback comes another opportunity for AAOI… this time holding at the 100D EMA instead, again with strong volume.

On top of this, we have a second support line around the same price, in the 21W EMA.

Now this isn’t a sure thing and like I said before it is a very volatile stock, not to mention it held the 50D before losing that, but there is definitely potential for a bounce here depending what the wider market does.

I am less bullish than I am about the other two and will size it as such but the potential upside could make up for that.

Other Honourable Mentions…

There are a few other stocks I am watching but not necessarily planning to buy on Monday which I’ll briefly cover here.

Reddit (RDDT)

Another stock I mentioned last week and of all the ones I’ve discussed, this is one of my favourites as a long term position.

Last week I bought based on it holding its 50W EMA. This week it dipped below mid-week but reclaimed and closed above it on Friday, keeping the thesis intact.

Below is last week’s article if you want to read more about it, including a link to my longer thesis article too.

JM’s Top Stocks of the Week: 22nd June 2026

If I had more spare capital and a smaller position I would be buying more.

Enphase Energy (ENPH)

I have been looking into the energy sector recently because I see it as a key future bottleneck and this is one name I am watching.

Strong narrative, good looking fundamentals and now an attractive technical picture too.

So far it is holding at its 100D EMA and briefly tested it on Thursday.

It also recently tested and held both its 21W and 50W EMAs.

All these moving averages coming together in one place is a good sign and one I’ll be watching closely. This is likely the stock I would buy next after the ones I have already talked about, maybe even next week.

Subscribe

That’s it for this week, thanks for reading!

JM

It’s been a while since I have written one of these articles, or really had time to, but they are something I enjoy and hopefully they give people some ideas to look into.

Today I have three stocks I am planning to buy tomorrow at open. All three I actually already have positions in but two are very small and the other I am very bullish on.

I will be honest, I am not going to focus too much on the fundamentals today, plenty of people already do that for these stocks - instead I want to focus on the technical setups, why I am buying now and when I might sell.

Let’s get into it.

Reddit (RDDT)

Why It’s a Buy

If you follow me closely, this should be no surprise, I literally posted about it yesterday, but it really deserved a proper mention here.

Firstly, I think the fundamentals are great and I have written about it below already - if you are remotely interested I really recommend checking it out and it isn’t particularly long:

Reddit (RDDT) Q1 2026 Results & Why This Could Be a Special Business

Add on top of this some huge insider buys recently and you already have the makings for a potentially great buy.

But there is more, because I also think the technical picture is looking great.

If we look at the weekly chart we can see the 9, 21 and 50 week EMAs coming together at almost exactly the same point - that’s about as good of a support point as you could hope for. In fact we already had a retest of this point last week and bounced right off it. That’s a buy signal for me.

Not only that but, if the price holds above these averages, soon there is going to be a double golden cross, or whatever you would call that - the 9W (purple) will slice up through both the 21W (blue) and 50W (yellow) and then the 21W will cross the 50W too, stacking all the averages in order.

Once that’s done there’s one nearby resistance point to watch, the 50W SMA at about $187. This is the point where the stock found resistance recently. Break that and the next resistance is the highs from September 2025 and January, around $280 and $260. To $260 from $175 if you buy around there tomorrow would be about a 40% gain.

The final thing I find interesting is the volume - the stock has been moving upwards recently on declining volume. This can show a lack of interest and momentum but also shows there’s not many sellers here. Even a low volume of buyers is enough to push the price up.

What I am Doing

So first of all, I already own some RDDT - not a huge amount, about 3-4% of the portfolio, but some. I want a lot more though.

My plan is to buy tomorrow, probably an hour or so after open, assuming the price holds above the moving averages around $167.

If it does dip below the averages, I may still buy anyway but I might wait until later in the week to try and get a better idea where it might close the week.

In terms of selling, since I believe in this stock long term, I’m in no rush to sell and will keep adding when I think it is attractive to. The sell signal for this ‘trade’ though would be:

• A weekly close below the averages, say under $163

• Or, a big intra-week drop, say over 1.5xATR which would be around $135

I would not normally want to risk such a big drop to close a trade but because I like this business so much and believe in it long term, I am happy to here.

Subscribe

Applied Optoelectronics (AAOI)

So, as I said in the intro, I’m not going to dive into the fundamentals much here, lots of people have already done that. If you have never heard of them though… basically AAOI is one of the major players in photonics, basically the next big thing in AI hardware. Search the ticker on Substack and you’ll find lots I’m sure.

It has already ran a lot and might look expensive but that depends who you ask and what you believe. Short term though at least, I think we have some potential here.

Why I am Buying

What I will focus on here is the technicals. My thesis isn’t long or complicated but I do not think it has to be.

It is the following: if you look at the daily chart below, we have just had a retest and successful hold of the 50D EMA with increased volume, after several days of declining volume. This suggests to me that buyers/algorithms stepped up when they needed to and defended this level. This level was also tested at the beginning of June and held then too.

Ultimately, I believe there is a strong chance this support holds again and we get a bounce. The stock is very volatile and still seems to have relatively strong momentum behind it which makes me believe there is a reasonable amount of upside. Finally, the narrative and fundamentals are still strong - it passes all my checks.

What I am Doing

Nice and simple, I’ll buy around $162 if I can and sell if there is a daily close below that $161 moving average. If we break it and have an intraday dip over below $140, which is over 1xATR and below previous lows, I will also sell.

I will note though that since this is a volatile stock and the ATR is high, I will size accordingly, it won’t be a huge position.

Rocket Lab Corporation (RKLB)

Final one for today and I’ll be brief again…

Why I am Buying

Fundamentally, there’s huge potential but it’s expensive - how expensive and if it is worth it, all depends what they pull off, that’s basically it in a nutshell.

Narrative wise, SpaceX has kind of stolen the limelight recently and capital has probably moved from Rocket Lab to them as a result, but I think this gives us the opportunity.

Technically, we have a price that is starting to consolidate right on the 50D EMA. There’s been a couple of closes below it but it’s holding pretty well, well enough that I am going to look at it as support. My theory is that capital will start to rotate back and this will serve as a bottom in the price. Again, not a sophisticated or complicated thesis but I believe it is worth taking a chance on.

Upside is sky high really with this stock if the narrative carries on.

What I am Doing

Very similar to with AAOI, I will buy around $106 if I can and hold until either a close below the moving average or a drop below $90 which is about 1.5xATR. Again, it won’t be sized huge, to limit total losses, but the upside potential is enough to still offer meaningful gains.

That’s it for now and thank you for reading. If you enjoy this sort of article please let me know and I will endeavour to do them more frequently.

JMIt

Back in January I published my deep dive on dLocal, the emerging market payments processor, and concluded it was a potential long-term compounder hiding behind messy optics. This article will be an update on that five months on.

If you have not read the deep dive, you can find it below. This update will assume you haven’t, but the deep dive covers the business model and moat in far more detail.

https://open.substack.com/pub/jminvestments/p/dlocal-dlo-deep-dive

Since then we have had two sets of results - Q4 and FY25 in March and Q1 26 in May. We have also had the first regular dividend payment, a new $300M buyback and a stock price that has done absolutely nothing to reward shareholders. DLO is down around 12% year-to-date and currently sits around $12.50, a fair bit below its 52-week high of $16.78.

So… with the business seemingly delivering and the stock falling, I thought it was the perfect time to revisit the thesis: what is going right, what is going wrong, what I think it is worth and what I am doing about it.

Subscribe

The Headline Numbers

dLocal reported Q1 2026 on 14th May and on the surface the growth was exceptional:

TPV: $14.1B (+73% YoY) - surpassing $14B for the first time and the sixth consecutive quarter of 50%+ growth. Even in constant currency this was +63%. Fantastic to see.

Revenue: $335.9M (+55% YoY) - although broadly flat vs Q4’s $337.9M due to a less favourable payment method mix and narrower FX spreads.

Gross Profit: $118.7M (+40% YoY) - a new record, up 2% QoQ.

Gross Profit/TPV (net take rate): 0.84% - down from 1.05% a year ago and 0.88% in Q4.

Operating Profit: $52.8M reported, or $57.2M (+25% YoY) excluding a $4.4M one-off opex adjustment (part of a wider $9.7M prior-period tax item, the rest of which sits in the tax line below operating profit).

Net Income: $41.9M ($0.14 diluted EPS), down 10% YoY as reported. Excluding the tax one-off it was $51.6M ($0.17), up 11% YoY.

Adjusted FCF: just $14.7M, down 63% YoY on temporary working capital effects which management expects to reverse. Not great.

Corporate cash: $451.8M, up $27.3M in the quarter even after ~$10M of buybacks.

Full-year 2026 guidance was left unchanged: TPV growth of 50-60%, gross profit growth of 22.5-27.5% and operating profit growth of 27.5-32.5%.

The market’s reaction? The stock fell around 4-5% on the day and has drifted lower since.

I’m not particularly surprised, the reported numbers were a bit messy - a GAAP EPS miss caused by the tax one-off, operating expenses up 58% YoY, and a very weak FCF print. Top-line looked great but bottom-line less so.

But as I explained in the deep dive, dLocal’s quarters often look messier than the underlying business actually is. So let’s separate the signal from the noise…

Thesis Update: What Is Going Right

The Growth Machine Keeps Running

For me, TPV remains the single most important metric at this stage of dLocal’s life and it is hard to find anything wrong with it. $14.1B in a quarter, 73% growth, six straight quarters above 50%. For context, the full year 2025 came in at $40.8B (+60% YoY) with revenue crossing $1B for the first time, net income up 63% to $197M and adjusted free cash flow more than doubling to $191M.

On the Q1 call, management noted that pay-ins surpassed $10B for the first time while pay-outs reached $3.9B, with growth broad-based across on-demand delivery, ride-hailing, SaaS, streaming and financial services. Newer verticals like travel and gaming are also ramping, with gaming flagged as a potentially higher take-rate segment.

To me, this is exactly what the thesis needs: volume growing relentlessly, across more products, more verticals and more geographies.

Diversification Is Improving

One of the risks I flagged in the deep dive was concentration - by merchant and by country. The direction of travel here is encouraging though. dLocal now operates in over 60 markets (the company says this gives access to 70%+ of the world’s population), adding Algeria, Qatar, Kuwait and Oman recently, and holds 38 licences and authorisations across 26 markets.

More importantly, the growth is no longer a LatAm-only story. Africa and Asia grew gross profit 16% quarter-on-quarter with notable contributions from Nigeria, Mozambique and Vietnam. Argentina recovered strongly after the election-related FX volatility that hit Q4 and Brazil’s gross profit more than doubled year-over-year (the sequential dip there was seasonal after a huge e-commerce Q4).

Retention Remains Best-in-Class

Revenue retention has now exceeded 140% for four consecutive quarters. As I said in the deep dive, this is the metric that proves merchants not only stay but keep expanding on the platform and it remains exceptional. The moat - one API, one contract, dozens of hard to operate markets - appears fully intact.

Management even quantified part of the value proposition this quarter: dLocal delivers up to 20 percentage points of uplift in conversion rates versus international acquiring. That is an enormous benefit for a merchant and explains the level of retention better than anything else.

Capital Returns Have Arrived

This is a genuine change since the deep dive, where I noted DLO preferred dividends because the small float made buybacks awkward. They have now done both:

• The first regular dividend under the new policy (30% of prior-year free cash flow) was declared in March - $57.2M, around $0.1939 per share and was paid on 10th June. At ~$13 that is roughly a 1.5% yield.

• The board also authorised a $300M share repurchase programme running to March 2027 at the latest. Against a market cap of roughly $3.8B, that is potentially close to 8% of the company.

Around $10M was already repurchased in Q1. Management’s stated logic was that the business will generate significant cash beyond its liquidity needs and dividend commitments and they would rather not hoard it. For a business operating in volatile emerging markets, returning capital while still growing TPV at 70% is a statement of confidence and it directly supports my deep dive point that DLO does not need outside capital to grow.

Subscribe

What I Am Watching: The Cons

I would be doing a disservice if I only covered the good and to be honest, there are some real things to watch here, I can kind of see why the stock price is where it is.

Take-Rate Compression Accelerated

This is the big one and essentially what the bears keep coming back to.

Gross profit over TPV fell from 1.05% a year ago to 0.84% this quarter. I modelled take-rate compression in the deep dive and argued volume growth would more than offset it, which it did with gross profit still up 40%, but a fall of that size in twelve months is faster than I would like and had modelled, at least in my base case.

Management attributes it to scaling volume with established merchants and expanding into new payment methods, products and countries, which is the standard payment processor dynamic. The full-year guidance of 22.5-27.5% gross profit growth on 50-60% TPV growth tells us management expects this compression to continue through 2026. The thesis survives this, but if gross profit growth ever stalls while TPV keeps flying, that is the warning sign I described in the deep dive.

For now, I’m still willing to accept some margin compression to keep scaling volumes well but this needs watching.

The Cost Base Has Ballooned (For Now)

Operating expenses rose 58% YoY. Even normalised for the $4.4M of one-off tax adjustments in opex and a further $3.8M of one-off effects, opex was about $57.7M and up roughly 9% QoQ. This is the carry-over from the 2025 investment cycle, which ramped mostly late last year and management has been explicit that operating leverage should improve in the second half of 2026 helped by the end of the investment cycle, automation-driven headcount reductions, corrective opex actions and lower share-based payments.

The operating profit to gross profit ratio of 48% (ex one-offs) versus 54% a year ago is the cost of all that. I am willing to give them the second half of the year to prove it, 2025 played out exactly this way after the 2024 investment dip, but this is now a “show me” item.

Free Cash Flow Was Ugly

No denying this. $14.7M of adjusted FCF against ~$52M of net income is a poor quarter of cash conversion for a business I praised for converting consistently. The explanation is timing of tax credit and higher receivables from their advancement operations. This is plausible, they expect it to normalise in coming quarters, and corporate cash still grew $27M in the quarter. But cash conversion was a core quality pillar of my thesis so I want to see this revert in Q2 and Q3.

One-Offs and Optics

The $9.7M prior-period tax adjustment knocked reported EPS to $0.14 versus consensus around $0.16, and headlines reported a miss (excluding it was $0.17, slightly ahead). Management says they do not expect comparable items going forward. Fine, but tax complexity producing surprise adjustments is exactly the kind of risk & bad optics that keeps the stock cheap. For me, it is fair for the market to apply some discount for it.

The Float Is Still Tiny

Nothing has changed here since the deep dive - insiders and early investors own the majority of the company, the freely traded float is small, and the stock remains volatile around results as a consequence. The buyback will shrink the float further which cuts both ways - it amplifies drawdowns and it will amplify any re-rating too.

The Macro Picture

I briefly want to touch on the bigger macro picture: dLocal is very much a bet on emerging markets and the macro setup is the usual mix of structural tailwind and cyclical noise.

The structural story has not changed and is the core reason I own the thesis: emerging markets are earlier in the shift from cash to digital payments, have younger and faster growing populations and merchants increasingly want to convert international card processing into local processing - a trend management calls out as a secular tailwind across their markets. dLocal grew TPV 60% in 2025 in markets where most global payment companies cannot or will not operate. That asymmetry is the whole point.

The cyclical risks are real though and management was quite specific about them in their guidance commentary. Their key exposures right now are: the evolving Brazilian tax environment, Argentine FX, tariff sensitivity, electoral uncertainty across the region and broader EM currency risk.

We saw an example in Q4 2025, when Argentina’s election related FX and rate volatility pushed dLocal’s funding costs up and dented gross profit there. Then we watched it fully reverse in Q1 as conditions normalised and Argentina returned to strong growth. Brazil also holds general elections in October 2026, so expect more headline volatility from one of DLO’s biggest markets in the second half of the year.

My view from the deep dive stands: this volatility is not a bug, it is the moat. It is why competitors stay away, why take rates are higher than developed market processors and why the stock trades at a discount. With 60+ markets, the portfolio effect means single country shocks have historically been absorbed - the risk to respect is a simultaneous, multi-market shock, which no amount of diversification will fully remove.

Thoughts on Valuation

Now the part that matters and I will look at this a few ways.

Given the stock’s volatility I will assume a share price of $13 per share for this section.

Earnings Multiples

On trailing earnings of about $0.66 per share, DLO trades around 20x. Cheap-ish, but not screaming.

The forward view is where it gets interesting. Guidance implies roughly $280-290M of operating profit this year at the midpoint. Applying dLocal’s normalised ~16% tax rate and a modest net financial gain in line with recent quarters gets me to roughly $240-250M of net income for 2026 or about $0.80 per share. That puts the stock at around 16x forward earnings for a business guiding to ~30% operating profit growth - a PEG of roughly 0.5. That seems very reasonable for a business growing profits at this rate unless you believe the growth is about to stop.

EV / Gross Profit

Because the take rate moves around, I think gross profit is the cleanest ‘true top line’ for DLO.

Stripping out the $452M of corporate cash (using corporate cash, not the $816M total, since some of that relates to merchant funds) and adding back ~$107M of liabilities gives an enterprise value of roughly $3.5B.

Against guided 2026 gross profit of approximately $493-514M, that is about 7x forward EV/gross profit for gross profit compounding at 20-30%+. For reference, the business converted 69% of gross profit into adjusted EBITDA in 2025, so this is high-quality gross profit, not empty top line.

Cash Returns

On 2025’s $191M of adjusted free cash flow, the trailing FCF yield is about 5%. Add the ~1.5% dividend yield and a buyback authorisation worth up to ~8% of the market cap and you have a shareholder yield that probably should not exist alongside 50%+ volume growth unless you believe it is about to stop suddenly.

Yes, Q1’s FCF was weak and if it does not normalise, this pillar weakens, but management has been clear it is timing and I believe them. This has historically been a very cash generative business.

Scenario Valuation

Rather than try to forecast emerging market payments to a decimal place, here is a rough scenario framework off my ~$0.80 of estimated 2026 EPS, discounting back at 12%:

• Bear: growth fades to ~10% a year through 2030 and the market pays just 12x. That gets me to roughly $9 today.

• Base: earnings compound at ~18% and the multiple recovers to a still-modest 16x. I get roughly $16 today plus dividends along the way.

• Bull: growth holds around 25% as operating leverage kicks in and the market pays 20x. Roughly $24-25 today.

Put differently, if I hold my base-case 16x exit multiple, a $13 price only requires around 12- 13% annual earnings growth from here, for a company guiding ~30% operating profit growth this year. The market is still pricing in a meaningful slowdown that nothing in the operating numbers currently supports.

For what it’s worth, analyst consensus broadly agrees with my base case - average targets sit in the $15-17.50 area, with a range of roughly $14.50 to $21 and a consensus buy rating.

My DCF Model

Finally, while it’s somewhat like fortune telling… my own DCF, even knocking down the take rates and profitability to reflect some normalisation from this latest quarter, gives a price around low to mid $20s.

So by every method I look at, the market is essentially pricing dLocal as if the growth is largely over. Given six straight quarters of 50%+ TPV growth, 140%+ retention and unchanged guidance, I struggle to agree.

The bet you are making at $13 is that take-rate compression and EM chaos do not fully consume the gross profit growth - the same bet as in January but now with a dividend and buyback attached.

Technical Analysis: No Momentum To Speak Of

Fundamentally I think this stock is cheap but being honest, technically, it has not looked great recently.

DLO started the year around $14, rallied after the well-received Q4 results in March, but the Q1 print in May broke the trend and the stock has slid steadily since, now sitting around $12.50.

Longer term the stock price has been trending downward since August 2025, making lower highs and lows, most of the time trapped between the 200W and 100W moving averages above and below respectively (red and orange lines below).

Crucially, it is trading below key averages like its 200D (around $13.40) and 50W (around $13.14). From a technical point only, I would want a breakout above these with successful retest before I add heavily.

The other area I would consider is the $10-11 zone. The stock spent late 2024 and the first half of 2025 around there before the big Q2 2025 re-rating, so there is plenty of price memory around there and it’s held as a floor since. $10 is the obvious round-number level to hold and below that and the chart opens up uncomfortably towards the 2024 lows.

The next major catalyst is Q2 earnings on 13th August, and given this stock’s history of double-digit moves on results (it rose over 30% after Q2 25), I would not be surprised if the next big move, in either direction, happens there.

Based on fundamentals alone, I think anything around $13 today is a good long-term price and the closer to $10 the better. But technically there is no momentum, and you could easily wait months while nothing happens. If you are ok with that, then now could be a great time to buy but if not, waiting for confirmation of an uptrend might be better.

Basically, on fundamentals, I would buy now but on technicals I would wait.

My Final Thoughts

So we have a business executing well, a valuation that looks cheap but a chart that gives no reason to rush.

I still think dLocal remains misunderstood somewhat but I can kind of see why too. The market sees a falling take rate, inconsistent quarters, the emerging market risk and a one-off charges. I see a business compounding payment volume at 70%+, retaining merchants at 140%+, self-funding its expansion across 60+ countries and now returning hundreds of millions to shareholders, priced attractively.

The things I’m watching out for: gross profit growth stalling despite strong TPV, retention slipping meaningfully below current levels, the promised 2H operating leverage failing to show up and FCF conversion not normalising over the next couple of quarters. Not whether the headline EPS beats by a penny or if the take rate decreases slightly though.

Until the above happens I think the stock is a good buy with the caveat that it might take some time for returns to materialise.

If you have already read some of my content you will know I already own DLO as a fairly large position and I will continue to hold for now.

I will revisit after Q2 results on 13th August and keep you updated along the way.

Thanks for reading,

JM

Please consider subscribing below if you found this interesting, I really appreciate any support.

Subscribe

I have kept this article completely free to view but some, including the full dLocal deep dive, contain paywalls.

JM Investments holds a position in DLO at an average price of $11.05/share.

Below is what they reported and why I continue to believe in them. Despite the frustrating price action this year, the business itself has kept delivering and Q1 2026 was no exception.

Subscribe

The Headline Numbers

I’ll start with the headline numbers:

• Revenue: $396M (+50% YoY) - beat guidance midpoint by $26M or 7%

• Organic Revenue (ex-Marigold & political): $341M (+29% YoY) - fourth consecutive quarter of organic growth acceleration

• Adjusted EBITDA: $66.1M (+42% YoY) - 16.7% margin

• GAAP Net Loss: $13.2M - narrowed from $21.6M a year ago

• EPS (GAAP): -$0.06 - ahead of expectations

• Free Cash Flow: $41.7M - FCF conversion of 63%

I am pleased to see organic revenue growth accelerating, they are not just acquiring their way into growth.

Guidance:

• Q2:

  • $420M revenue

  • $86.6M adj EBITDA (20.6% margin)

• Full Year 2026:

  • $1.785B revenue (raised by $30M, +37% YoY)

  • Adj. EBITDA Guidance: $397M (22.3% margin)

  • FCF: $235M

  • Guiding to positive GAAP Net Income

This was their 19th consecutive beat and raise quarter. At this point it stops being luck or favourable conditions and the norm. You could argue it is a statement about the quality of the management team and business model, or that they just sandbag guidance every quarter. Regardless, it’s impressive that they can keep on guiding higher.

The Rule of 67 and What It Means

Management highlighted their ‘Rule of 67’ performance this quarter, combining revenue growth rate (50%) with adjusted EBITDA margin (16.7%) to reach 66.7. This is a fairly common metric for software companies and is essentially a way of framing growth efficiency. It shows you are not just growing fast but doing it profitably. A Rule of 40 is generally considered good so Zeta running at 67 is very good and puts them in a select group of companies.

What makes this even more encouraging is the direction of travel. Adj. EBITDA margins are guided to expand throughout 2026 - Q2 is guided at 20.6% and the full year average comes out around 22.3%. The business is scaling well and the operating leverage is becoming increasingly clear. Combine this with their fantastic revenue growth and things can very quickly look very good.

Athena: The Game Changer?

The most interesting thing this quarter was the launch and early data on Athena. Athena has long been discussed as a potential game changer for Zeta and was launched for general availability on 24th March 2026, essentially just days before the quarter closed.

Despite the small window of availability, management reported that in its first week, Athena drove:

• Over 7x more agent interactions than Zeta’s previous AI platform usage

• More than 60% of all AI platform activity

I would argue those are not the kind of numbers you see from a feature nobody wants. That is explosive adoption right out of the gate. My only concern is that it could be skewed by people testing it because it’s new and those numbers drop later - it is a possibility so we will have to wait for Q2 for confirmation.

What actually is Athena? Basically a conversational AI agent for enterprise marketing, built on OpenAI’s models and Zeta’s proprietary data cloud. Marketers can interact with it through natural language to get instant analytics, campaign recommendations, audience insights and direct execution across channels. Effectively it should turn Zeta’s platform from a tool you operate into one that thinks and acts alongside you.

If it works well it could have enormous potential. First, it deepens the moat. The more customers use Athena, the more it learns from their data, their decisions and their outcomes. This creates a flywheel that gets harder to replicate over time, particularly because Athena is built on Zeta’s own proprietary data. Second, it expands wallet share. Super-scaled customer ARPU grew 21% YoY this quarter, the strongest rate in over a year. Management explicitly linked this to Athena driving more use cases and deeper platform adoption. The idea is that as Athena proves its value, customers naturally do more with the platform, spend more and become stickier.

The Core Thesis: Why I Own ZETA

I will be upfront - ZETA is one of my largest positions and one I first started buying over a year ago now. Not just focussing on this quarter, here is why:

Proprietary Data at Scale

Zeta’s platform is powered by a proprietary data cloud incorporating trillions of consumer signals. This is not rented or licensed data, it is theirs, built from years of interactions across their platform. This data also gets smarter with every customer interaction, which is the key input that hopefully makes Athena actually powerful rather than just a chatbot bolted on top. Data moats of this kind are hard to replicate.

Enterprise customers choosing Zeta are not just buying software, they are plugging into this intelligence layer and in theory every interaction makes it more valuable.

Marketing Cloud Replacement

Zeta is positioning itself as the AI-native replacement for legacy marketing clouds like Adobe, Salesforce and older tools that enterprises have used for years but which are increasingly struggling to compete in an AI- first world. This is a huge addressable market and the cycle of replacement is just beginning. How successful this is, is a large part of how good of an investment ZETA becomes.

Management described this quarter as evidence of enterprises “consolidating vendors into a single, unified operating model” and that vendor is increasingly Zeta. We need more time to confirm this but it looks positive. Organic revenue acceleration for four consecutive quarters shows this is not just M&A flattering the numbers.

Valuation

Valuing ZETA is tricky because it is still posting GAAP losses which rules out a straightforward earnings multiple. Instead I think the most useful way to look at it right now is on a forward revenue and EBITDA basis.

At around $18-19 per share, the market cap is roughly $4.5B. With $1.785B full year revenue guidance that is approximately 2.5x forward revenue. For a business growing at 37% with accelerating organic growth, expanding margins and a clear path to GAAP profitability, that is genuinely cheap. The average SaaS business growing at half this rate trades at a higher multiple. At its all-time highs ZETA was trading somewhere closer to 10-15x forward revenue.

On EBITDA, $4.5B against the $397M full year guidance gives roughly 11x forward EBITDA. Again, for 37% revenue growth with margins expanding throughout the year, that isn’t a stretched valuation. FCF of $235M guided for the full year puts the forward FCF yield at around 5% which I think is attractive for a high growth business.

The analyst consensus sits in the $25-$30 range with some as high as $44. Most have a Buy or Strong Buy rating. Even the more conservative targets imply 30-50% upside from here. My own personal DCF based on their ZETA 2028 targets also puts a fair price over $30.

By whatever metric I look, this business seems cheap for what you are getting. It’s not stinking cheap, deep value but it is very cheap for such a high quality business. If this business works out how I expect I really think it could be a multi-bagger in the new few years.

Subscribe

The Risks

I have discussed why I own ZETA but there are still risks of course. There are a few things I am watching closely that I would be doing a disservice not to mention.

GAAP profitability is still a question mark for now. The full year is guided to be GAAP net income positive, but right now they are still posting GAAP losses and the Marigold integration costs are adding noise. I believe it will happen but until it does and the operating leverage really kicks in, there will always be some risk.

Speaking of Marigold, while they contributed $55.6M this quarter and are expected to contribute around $198M for the full year, integrations of this size always carry risk. They can distract management, delay product development and sometimes disappoint. So far the early signs are good but it is worth keeping in mind.

Another thing, Zeta is a great business but realistically it is competing against significantly larger ones that aren’t just going to roll over and give up. These businesses have enormous existing customer relationships and sales teams and they are all trying to adapt to this new AI world too.

Finally, people often mention SBC and it was high but I’m comfortable with the direction and speed it has been trending downwards over the last couple of years.

Technicals: Struggling, But With Signs of Life

So fundamentally I like this business but technically, things don’t look great to be honest when looking at the weekly chart below.

ZETA was affected by the SaaS sell of this year and after hitting a high in January, has chopped lower most of the year. At the moment it seems to be somewhat range bound between $15 and $20. It is trading below most of its moving averages and seems to be lacking any upwards momentum. Brief pops such as after earnings are quickly sold back off. In fact after the Q1 earnings it rose to ~$20 before falling back into the $15s and $16s a couple of weeks later.

I would not be surprised if this continues for a while longer until a major catalyst is announced. That could be Q2 earnings, it could be Q1 2027, I’m not sure.

The remaining moving average that it’s still holding above, and did even at its lows, is the 200W around $14.15 (red line on the weekly chart above) At the other end, the 100W sits around $19 (orange line) and I think it needs a solid break and hold above there to get out of the range it’s stuck in.

Closing Thoughts

Well I clearly like ZETA and I think this has been another great quarter in their story.

Q1 was another reminder of why I have conviction in Zeta as a long-term holding despite the underwhelming price action.

The business is accelerating, Athena is showing early signs of being transformational and management is guiding to full year GAAP profitability.

The market has been brutal on software this year and Zeta has not been spared but the fundamentals have remained strong throughout and it’s very hard to argue the business is broken.

Based on fundamentals only, honestly anything under $20 seems a good deal to me but the closer to $14 you can get the better. However, technically, I’d be waiting for that break of the 100W to add and capture some momentum - before then you could be waiting for months while nothing happens. Personally, I’ll probably do both - add more around that 200W MA if we get there and hold but I’ll also add on the way up as the stock starts to breakout.

This remains one of my largest positions and one I intend to hold.

Thanks for reading,

JM

Comment

Reddit dropped their Q1 2026 results on April 30th and they were fantastic. The stock jumped over 13% on the day and is still sitting around $165, which is actually still down around 36% from its all-time high.

Given what they reported, I think they are a worth paying attention to.

I hold a small position in RDDT, bought just before earnings but I am strongly considering adding more on the back of these results.

Below is a summary of the results, my thoughts on them and why I am long RDDT.

The Headline Numbers

Let’s start with the numbers that matter most:

• Revenue: $663M (+69% YoY) - beating analyst estimates of ~$614M by around 8%

  • Ad Revenue: $625M (+74% YoY)

  • Data Licensing / Other Revenue: $39M (+15% YoY)

• Net Income: $204M (31% net margin) - up 680% year-over-year

• Adjusted EBITDA: $266M (40% margin) - up 131% YoY

• EPS: $1.01 vs $0.62 estimate, a 63% beat

• Free Cash Flow: $311M (47% margin)

• CapEx: $1M - that’s not a typo

• Gross Margin: 91.5% - up 100bps YoY

• Daily Active Uniques (DAUq): 126.8M (+17% YoY)

• Weekly Active Uniques (WAUq): 493.1M (+23% YoY)

• Global ARPU: $5.23 - beat $4.81 estimate

• US ARPU: $9.63 - beat $8.53 estimate

• $2.77B cash & equivalents

I have highlighted a few numbers in bold above to emphasise how good these results really are.

69% revenue growth and also their seventh consecutive quarter of revenue growth above 60%. 47% FCF margins with $1M spent on CapEx, that is staggering. Revenue is growing at a blistering pace and nearly half of it is flowing straight into cash. Operating margins also improved and active users are growing steadily. There’s lots to like.

It does not finish with Q1 either - Q2 guidance was strong too with $715M to $725M revenue, ahead of the $712M consensus and adjusted EBITDA of $285M to $295M, ahead of the $276M estimate.

Subscribe

Why I Think Reddit is Special

The results are obviously great, but I want to talk more about why I think Reddit is worth paying attention to, because the numbers alone don’t tell the full story.

The Content Moat

Reddit has over 25 billion posts and comments on its platform, accumulated over more than 20 years. CEO Steve Huffman described it on the earnings call as being like “oil for the modern internet”.

Every month, Reddit’s communities generate in new content, the equivalent of Wikipedia’s entire library. That is a ridiculous amount of human-generated data and in a world where AI-generated content is becoming widespread, that authentic, real human conversation is arguably becoming more scarce and valuable.

This is a moat that simply cannot be replicated overnight. You cannot just build a competitor with 20 years of community discussion and I do not think there is anything close. I used to think X could be but these days AI content is everywhere there and even without it, there’s not the quality and depth of discussion Reddit has.

The AI Data Licensing Angle

This leads into one of the most interesting aspects of Reddit’s business and a big part of why I am paying attention. As AI models need more and more high-quality human data to train on, Reddit sits on one of the largest and richest reserves of exactly that.

Google and OpenAI are already two of Reddit’s biggest data licensing partners. The “Other Revenue” category, which is primarily this data licensing, grew 15% YoY to $39M this quarter. Ok, that’s not huge yet, but the direction is clear and the strategic value is massive. Every major AI lab in the world needs what Reddit has and no one else can replicate.

I think as the AI landscape develops further, this licensing revenue has the potential to become far more significant than the market currently gives it credit for. Reddit is a great business on advertising alone but this certainly forms part of the bull case.

The International Monetisation Gap

One of the things that jumped out when I looked at the Q1 numbers was the ARPU gap - US ARPU came in at $9.63, while International ARPU was just $2.02.

That is a big gap. It tells me that Reddit is not being properly monetised outside of the US yet. International DAUq grew 26% YoY, far faster than the 7% US DAU growth, and yet those users are currently generating just a fraction of the revenue per user of their American counterparts.

As Reddit continues to build its international ad infrastructure, localises its product (they now translate content into 30+ languages), and grows its advertiser base globally, I see this as a very clear path to substantial ARPU and revenue growth. It’s almost a second growth wave sitting right in plain sight.

The Capital-Light Model

Let’s just come back to that CapEx number: $1 million on $663 million in revenue. That is 0.2% of revenue to run and grow this business, something I have never seen before. It means that essentially all of their free cash flow converts directly to value.

The $311M free cash flow this quarter was nearly 47% of revenue and they now have $2.77 billion in cash and securities, a $1 billion share buyback authorised and are essentially printing money.

Capital light, high growth and generating tonnes of cash is exactly the sort of elite business that makes a long term compounder.

The Risks

I don’t like to talk about a business without also acknowledging the risks too, so here are the things I am keeping an eye on:

• Ad concentration - Advertising accounts for roughly 94% of revenue. This means Reddit is heavily exposed to any macro downturn that causes advertisers to pull back spend. The broader ad market has been strong recently, Meta and Alphabet both reported their fastest growth in years this cycle, but that won’t last forever.

• US DAU growth is slow - While international is surging, US DAUq only grew 7% YoY. Reddit has stated their goal is to reach 100 million daily US users (they are at around 53M US DAUq right now). That is a lot of growth to find in a market where most people who want to use Reddit probably already do. I will also acknowledge here that while a slowdown in top-line revenue is expected, the Q2 guidance is for mid-40% growth vs the 69% this quarter.

• AI search disruption - There is a risk that as AI-powered search improves, fewer people navigate through Reddit the traditional way, reducing organic traffic. Reddit is building their own AI-powered search to address this and it could even become a tailwind as there becomes less human answers online the data, plus licensing deals mean they actually benefit from AI growth too, but it’s a risk worth watching.

• Valuation - At around $165 per share, RDDT trades at roughly 47x trailing earnings which isn’t exactly cheap. The business growing at 69% and generating 47% FCF margins does justify a premium, but the question is how much of one.

Thoughts on the Valuation

Reddit is not a business that I think can be easily valued with something like a DCF model - the growth trajectory is difficult to predict and the results swing estimates dramatically. Instead, I think the most useful lens is to look at where this business could be in a few years.

The business is clearly at an inflection point. Margins are expanding fast, revenue growth is remarkable and they have several distinct growth drivers still in relatively early stages:

• International monetisation,

• Data licensing,

• Continued ad product improvements

Analyst consensus sits around $225-$230, implying around 35-40% upside from here. All-time highs are around $280, reached in September 2025. Given the consistent beats and the growth drivers outlined above, I do not think those prices are unreasonable. At 47% FCF margins and still-accelerating revenue, Reddit should be worth considerably more than it trades for today if it simply continues executing.

That said, at 47x earnings, if growth slows or the ad market turns, there is meaningful downside too. This is not necessarily a low risk valuation.

Technical Analysis & My Positioning

RDDT has been in a pretty strong downtrend this year and was down around 50% YTD when it bottomed towards the end of March around $120. Notably this was below all the key moving averages.

I did actually buy RDDT earlier in the year when it briefly held on its 100W moving around $140 and wrote about it below however I ended up selling shortly after as the average was lost.

https://open.substack.com/pub/jminvestments/p/jms-top-stocks-for-the-week-16th

More recently the trend has been more positive and the price reached mid $160s in April before retesting and holding the 100W and 50D moving averages at around $145. Based on this and the fundamentals above, I opened a position here.

The weekly chart below shows the 100W SMA in orange.

Today the stock is back up to the $160s after a pop on large volume after earnings results. While I normally use simple moving averages for anything bigger than 50 days, it has just popped about its 50W EMA and seems to be using that as support. I added more here and will likely do at least one more buy should these levels hold.

Below is the daily chart as of 1st May.

Closing Thoughts

Reddit is a genuinely interesting business. The combination of a strong moat, an opportunity in AI data licensing still in its early innings, a massive international monetisation gap and a capital-light model producing exceptional FCF puts it in a rare category.

It is not cheap but exceptional businesses rarely are. The key question is whether the growth runway is long enough to justify the current price, which based on what I see, I think it likely is.

I will be watching this one closely to see how it plays out and keep you updated along the way.

Thanks for reading,

JM

Please consider subscribing below if you found this interesting, I really appreciate any support.

Subscribe

I have kept this article completely free to view but some do contain paywalls.

• My performance that quarter and YTD

• Any big wins or losses

• Any reflections or lessons learnt

My hope is that this might be interesting for some people but will also serve as a great learning exercise for myself and a way of capturing my thoughts over time.

I will keep these posts for paid subscribers only.

Subscribe

Headline Performance and Thoughts

My YTD performance as of 31/03/26 was -16.5%.

Using Vanguard’s S&P 500 (VOO) as a comparison, that finished Q1 at -4.7% YTD. Vanguard FTSE All-World (VWRP) did even better at -2.4%.

Clearly I have underperformed and by some margin.

Q1 was tough and as I will touch on soon, I made what I now think were a few mistakes which cost me. The silver lining is I think I have learnt from these and I also do not have to perform well every quarter, long term outperformance is my goal.

Quarter in Review

The year started extremely well for me, as it did for many people. The portfolio reached ~10% YTD by mid-January, driven by strong performance in big positions like ZETA, and several successful trades while upwards momentum was strong.

Performance declined sharply from there until the end of February where it stood at -20%. This was driven largely by bad performance in most of the top holdings, while holding relatively little cash or hedges.

Performance improved somewhat over the next few weeks before another low around -20% near the end of March. Again, this was driven by poor performance in the top holdings, most of which were even lower than in February. However, overall losses were reduced vs February’s low by opportunistically raising cash on the small pumps throughout the month and several successful trades.

On the last day of March the market swung back upwards and managed to bring performance up to the -16.5%.

As somewhat of a silver lining, despite the poor performance, I feel the portfolio was in a much better place at the end of the quarter than the start of the year. Several of the laggards were cut and many great businesses were bought at what I believe are great prices, setting the portfolio up well for the rest of the year. I will talk more about these shortly.

Top Winners & Losers

Truthfully, when I look deeper into the top wins and losses over the last quarter it is almost surprising that the portfolio performance is not lower. Several large positions sustained very large losses and on the contrary relatively few achieved large gains.

Top Losses

ROOT

The worst performing stock in my portfolio over the last quarter has been ROOT, down ~64% from the end of 2025 to end of March. At its peak, this was a high single-digit position so the impact was significant. I will discuss it in the lessons learnt section next but in hindsight I should have handled this position differently to limit this loss.

Ironically, it remains one of the positions I believe has the highest potential in my portfolio. I wrote about it most recently in February in the article below and my thoughts have not changed much since:

FOUR

Up next is FOUR down ~44% over the quarter. This is a position I believe in long-term although it has become somewhat of a ‘prove to me’ stock after their last results. I am not particularly surprised by the price action here but again, see it as one with huge potential. At its peak this was a mid-teens percentage position and on a pure capital basis, was responsible for the largest loss in the portfolio this quarter.

UNH

Rounding out the losses, and the one I was most unhappy about, is UNH down around 22% for the quarter while it was a low teens position. I believed, and still largely do, that they are undervalued and will be a great long term hold but there is too much uncertainty influencing the stock recently and I do not think it is the best hold right now.

Crucially, I saw UNH as a ballast like position for the portfolio, which would hold relatively steady in a risk-off scenario like the one we had. This was not the case, perhaps usually it would be, but with the news around health insurance this year and ultimately a big downwards move after earnings, it dropped heavily when I counted on it not doing so.

Top Winners

Being totally honest, there were not many big winners in Q1. Obviously, performance was bad and while there were some wins to offset the losses, these were made up of several, smaller trades rather than single large positions.

The only large position which performed particularly well in Q1 was NBIS. This gained around 19% and was a high single-digit percentage of the portfolio. I am very bullish about them and my only regret is not buying more sooner.

The other ‘wins’ to note in my opinion are not gains made in Q1, but the buys I was able to make which I believe will set me up extremely well for the future. I will discuss these in the next section.

Position Changes

I trade in and out of positions fairly regularly so will just cover the main ones here.

Sells

UNH, MOH & CNC

I reduced my UNH position by ~60% based on my comments above. I only sold for a small loss overall but after the 20% loss I talked about. I did this because I do not feel the stock and business is what I initially thought it was. I am still keeping a smaller position though because I do think it is a good business at a cheap price.

I also sold my other health insurer stocks, MOH and CNC, based on the same reasoning. MOH was also sold around breakeven but CNC was for around a 50% gain. Between them they were about 8% of the portfolio.

PATH

I reduced PATH from about ~5% to ~1% of the portfolio in several tranches throughout the quarter. Earnings did not show me the bull case I had hoped for and it is clearly in a strong downtrend technically. Claude is bringing out new features almost weekly and I feel that despite PATH having its advantages, these could be dwindling.

The fundamentals still look fairly good at the current price, which is why I am holding the 1% moonshot, but I do not want a large position in a laggard stock with a cloud hanging over it.

Small Positions - LMND, OSCR

I trimmed some of the smaller, lower conviction positions in the portfolio as the market started to turn and I wanted to reduce risk somewhat. This turned out to be a good move, allowing me to hold a larger cash ballast through much of the drop and then buy some great businesses recently.

Buys

MSFT

My biggest new position in Q1 was MSFT which ended up ~5% of the portfolio. I covered this in one of my Top Stocks of the Week articles, which you can find below.

I see Microsoft as one of the most long-term resilient software companies during this recent sell-off and technically it was reaching its 50 month moving average, a level not broken in over 10 years. It seemed like a fantastic business at a good price.

In the end, it did actually close March just below this average but I was happy to hold still.

MELI

MELI is a business I have wanted to own for a while but I have been put off by the price until recently. Again, I believe the fundamentals are great and technically it was approaching support at its 200W moving average. Basically, another fantastic business at a good price. I also spent much of Q1 in LatAm and speaking to people there further strengthened my conviction.

This became ~4% of the portfolio and I also wrote about it below:

SPGI

After its Q1 earnings SPGI sold off heavily, crashing through its 200W moving average and dipping as low as the $380s. This seemed a good opportunity to buy a great, stable business at a good price and particularly given the risk off environment, I felt it could add some ballast to the portfolio.

Unfortunately, by the time I decided to buy it was already around $400 but it became a ~3% position and is one I am happy to have been able to buy and hold.

NOW

Finally, following a similar theme to the other buys, NOW offered another opportunity for a great business at a good price when sold off heavily in February. Strong fundamentals and it was approaching its 100 month moving average, an eight year trend line.

I also wrote about this one in the same Top Stocks of the Week article as MSFT above.

Lessons Learnt

This was the first bad market I have been caught out by since I started investing in individual stocks and it has taught, or at least reminded me, of several things.

In 2020, I owned a lot of ETFs which faired better in the initial sell-off and I quickly sold most and moved to cash when it was clear we had issues. This let me buy several stocks at the bottom and come out very well by 2021. Then in 2022, I felt things were getting a bit crazy and took a lot of risk off the table early - I missed some upside but was shielded well from the downside that came.

In 2026, I did not have a strong buffer like the ETFs in 2020 and I did not take risk off the table like in 2022 which ultimately cost me.

Here are some of the key things I have taken away from the quarter.

Don’t Fight Strong Trends

Despite what people might say, the market likes to follow trends and stocks that are cheap can and do get much cheaper, even if they already seem ludicrously cheap. I knew this but several times in Q1 I fell into the trap of not wanting to miss a bargain.

Using FOUR as an example, I thought it was cheap at $80 and bought there, so did the founder and Chairman, but now it’s in the $40s… It probably was cheap then but now it is much cheaper and those buys are down nearly 50%. Similar story with ROOT.

Now I prefer to buy on the upwards move from a bottom, not on the way down. Yes, this might mean I miss the very bottom and some upside, but it also means you’re not throwing capital at a laggard for months or buying nowhere near the bottom. I wrote about this strategy more in the article below:

Going forward, I am also not against selling stocks at small losses if they break key moving averages and start longer term downtrends.

Similarly but in the other direction, I missed many strong performing stocks in big uptrends because I was worried about buying at what I perceived high values. Stocks like SNDK in the $400s back in January which nearly doubled in Q1.

Finally, on a positive note, even during Q1, following this advice allowed me to avoid buying several stocks until near their bottoms and got me some great entries.

Have Some Sort of Buffer

Ok, so there seems to be a lot of people leveraging into a few stocks and smashing out triple digit returns and that’s fantastic for them, but personally I like to have some sort of buffer in my portfolio, even if it might drag on gains. I’m not saying sit at 50% cash but even if it’s just a relatively defensive, stable position or regular deposits of fresh capital, I think having something to cushion a big downturn and potentially buy any dips is a good idea.

For me this was supposed to be UNH and the other health insurers which I have already mentioned failed that role. Now, I’m more cautious with my cash balance and ‘defensive’ positions, plus I am able to add a lot of fresh capital to the portfolio this year.

Final Thoughts

I will end here with some final thoughts on the quarter.

Overall, I have mixed feelings: the headline performance was not good and I’m disappointed, however I am happy with the moves made during the quarter and the way it was handled. Much of this quarter’s underperformance came from actions taken last year and in the same way, I am confident that actions taken this quarter will lead to strong performance for the rest.

In next quarter’s review I will share a full breakdown of the whole portfolio and more detailed performance so make sure to stick around if you would like to see that.

Thank you for reading and I hope you found this interesting. If you have any suggestions for future articles please feel free to leave them in the comments.

JM

With news about the US-Iran war looking bearish this weekend, they may already open lower and we can be sure there will be volatility but these are two of the stocks I will be watching.

SAP SE (SAP)

The first stock I’m looking at is SAP, a German software giant, entrenched in many of the largest businesses in the world.

Just like nearly every other SaaS stock, they too have been sold off heavily over the last few months and are now trading at a price where I think the risk/reward has become extremely attractive. The stock trades on the NYSE in dollars and the German Xetra in Euros where it’s down 38% in the last year. Here I will focus on the XETR listing.

Fundamentals

Total revenue was up 8% for FY25 with cloud revenue growing 23% and making up ~57% of the total. The business transitioning to a cloud-centric model and they’re guiding for this growth to accelerate in the coming years. It’s not hyper growth but it’s a strong business growing well.

Margins are strong with a FCF margin on ~22% and evidence of operating leverage as profits grow faster than revenue.

They have also announced a new share repurchase program of up to €10B from February 2026 to the end of 2027 which will represent around 5% of shares outstanding at current prices and is roughly equal to 2026’s guided FCF.

From a valuation perspective, a conservative DCF model I made gave a fair price around €250, giving a huge margin of safety from here.

The elephant in the room of course is potential AI disruption - that is why we have this opportunity. Nearly all SaaS businesses have sold off over the last few months because of this. The obvious risk is complete replacement of the business from AI but even if that doesn’t happen, disruption could come from lower sales as fewer people are needed for jobs and this could flow through into lower margins in a race to the bottom for sales. I believe some businesses will suffer but the ones that don’t, could represent fantastic opportunities.

My belief is that SAP will be one of the businesses more likely to handle this AI risk. They are incredibly deeply entrenched into many of the world’s largest businesses and provide mission critical systems. I have personally worked with many SAP’s systems and there is so much to them I cannot imagine them being disrupted without huge difficulty.

Whether I am correct, we will have to see but in the mean time, when combining the fundamentals with the technicals, I think they make an attractive trade opportunity and I’ll discuss that more next.

Technicals

As I said, SAP has been sold off heavily over the last few months and is down 49% from its all time high in February 2025.

On the daily and weekly chart, the key moving averages have all been lost but the monthly chart is what has gotten my attention. SAP is now trading almost exactly at its 100 month moving average, a trend line with eight years of data. The price hit this almost exactly last week and has held above it since.

This line was broken in the 2022 bear market which you can see on the above chart, so it’s not unbreakable, but even then it held just below it before running around 250% two and a half years.

Monthly RSI is around 39, again very similar to that 2022 low.

My Plan

I plan to open a position tomorrow at open with a limit price of €140 - I am hopeful it will fill lower but this is my upper limit.

Stop losses can be hard to judge with longer term averages since we could drop under it mid-month but still close the month above it. The monthly ATR is ~€25 so a €115 stop should cover that but does leave us vulnerable to around an 18% drawdown in the worst case. This could be used but would limit my position size.

A better middle ground for me though is to use a stop more like 0.5x ATR, at around €127. This gives some room to swing under the moving average and the round number support of €130 but limits the drawdown to around 9% so I can make the position a larger size.

The other sell condition for me would be a monthly close under the average. Obviously, we have just seen that it did this in 2022 but I do not want to risk holding a dead stock or a large drawdown. Instead, if it were to do what it did in 2022, I would prefer to breakout back above the average even if it means missing some upside.

If you find this article valuable or even just interesting, please consider subscribing for more!

Subscribe

Novo Nordisk (NVO)

The second stock today is probably one which needs no introduction to most people, European pharma giant Novo Nordisk, most recently famous or perhaps infamous for being one of the market leaders on GLP-1s and then perhaps squandering that position.

This is a stock I have owned before, in fact it was once about 10% of my portfolio before I realised the thesis I had in mind was not true and sold, but at its current price I once again am becoming interested.

Fundamentals

When it comes to fundamentals, it’s getting very hard now to argue NVO is not cheap. It is now priced around where it was before they even launched their weight loss drug, Wegovy. Despite its market leader, or even 2nd place position now being in dispute, and sales forecast being slashed recently, it is still a business generating a lot of cash with a very strong future pipeline.

I will not dive too deeply into the fundamentals here since, to be honest, that’s done enough by others already. Below is a recent update on NVO from one of my favourite writers on Substack and not only have they written this, they covers them in detail if you check their page.

Hated Moats

Novo Nordisk (NVO): Updated DCF Valuation

Date of Analysis: March 2 - 6, 2026…

Read more

6 months ago · 69 likes · 17 comments · HatedMoats

Doing my own rough and even more conservative DCF, I got a price $55 per share. This was assuming FCF does not reach the underlying levels it did last year, entirely writing off the massive manufacturing investments they’ve recently made.

Plus, while not a huge deal to me, it’s worth mentioning the dividend yield is now around 5%

Basically, I see it as the business is struggling, things don’t look great, but now it’s almost too cheap.

Technicals

So fundamentally it looks cheap but so do lots of stocks, the technical situation is what has gotten my attention recently.

Both this month and last NVO has been pretty much sitting on its 200 month moving average, that’s a 16 year trend line!

In all of its history, going back to 1981, this has not happened.

I do not believe the fact that the stock appears to have bottomed here is a coincidence and I am of the opinion that this trend line is acting as support. This is the short of place that institutions load up. With this in mind, only now am I happy to repurchase NVO despite it appearing cheap for months.

My Plan

I plan to start a position in NVO next week as long as it stays under $40 but ideally I will be able to buy around $38.

Similarly to with SAP above, I will use a stop loss at around $34.50 to prevent downside in a big flush but allow some room for the stock to temporarily dip below the 200M as it has recently. So far the 52W low is $35.10 and I am hoping the $35 point acts as a strong support level.

Also similarly to with SAP, I will also sell if we close below the monthly moving average.

Closing Thoughts

The macro environment is tough right now and I am still being cautious. I am buying some positions but keeping a healthy cash pile and managing risk strictly.

Usually I prefer to stick to smaller businesses but right now I see some advantages to holding beaten down mega caps like these.

I have a third and maybe even fourth stock I’m watching this week which I will post about later - this time one in a strong uptrend with secular tailwinds.

I will keep readers updated on these trades so make sure to subscribe and follow along if interested.

Thanks for reading.

JM

Subscribe

Before we get into the article though, if you are completely new to NBIS then I recommend you check out the deep dive below. I have not covered them on my Substack before but M.V. Cunha has the best coverage I have seen and was also one of the first.

M. V. Cunha's Substack

Why Nebius Group (NBIS) Is My Largest Position: A Deep Dive

Nebius Group (NBIS) is my largest position — and it’s finally starting to catch the attention of retail investors…

Read more

a year ago · 99 likes · 14 comments · M. V. Cunha

The Headline Results

First, I will start with the financial highlights…

Honestly, no surprises here - all very encouraging and basically as guided:

• Revenue is up and growing at a blistering pace still

• Adj. EBITDA is improving and was positive in Q4 - obviously there are a lot of adjustments but it is good to see that operating leverage is already taking effect and margins are improving

• The cash position is great and that plus some of the deals announced in the last month make me confident in Nebius’ ability to keep growing

• Operating cash flow is strong too but we all know this is an incredibly capex heavy business right now and net loss increased substantially too

Revenue was actually slightly below analyst expectations and some people were reporting a miss but honestly I think this is because they are looking at the wrong thing. More important than revenue is the ARR and power being brought online and results and guidance for these were very impressive just as they were in Q3.

FY 2025 ARR came in above their guidance, which was already very impressive, and their 2026 guidance is huge. They also ended 2025 with 170MW of active power instead of the 100MW targeted, beating their guidance on what is, in my opinion, one of the key constraints to the business, by 70%. Plus they sold out of all their capacity again in Q4, as they did in Q3 and Q1 was effectively already fully booked when earnings were released.

The pace they are delivering and growing is simply staggering.

They also announced their planned capacity expansion to support these new power goals and I am glad to see the majority of the new locations will be owned data centres.

Ultimately, this reinforces how huge the demand is and there have been new deals since the Q4 results, reinforcing this further.

It is extremely clear that NBIS is a business in demand and the tailwinds show no sign of slowing. The critical part in the thesis for me is if/when they become net profitable.

Expenses

Speaking of profitability, there is no hiding the huge expenses. These of course increased in Q4, but crucially they are becoming lower percentages of revenue across the board.

On expenses, they have guided for $16B to $20B in capex for 2026. This sounds huge and it is but against the results they are achieving and the demand we are seeing it is less scary.

Crucially, they have said ~60% of this is pre-funded from cash on hand, operating cash flows and upfront payments from long-term contracts. The remaining 40% will be financed with a combination of corporate debt, asset-backed financing and equity.

Hyperscaler Deals

They also updated on their MSFT and META deals, confirming that both were going successfully.

Meta’s full capacity was delivered in early February 2026 and Microsoft’s first tranche was delivered on time in November 2025.

Remaining tranches are scheduled throughout 2026, with the majority in H2, and management confirmed the supply chain for those components was secured in 2025 at pre-price-increase costs.

Subsidiaries Update

TripleTen

Wholly owned by Nebius, TripleTen grew revenue 88% YoY and continues to perform well.

Avride

Avride is continuing to grow, attract investments and has plans to grow substantially. While the financial contribution is low right now, I feel there is a lot of potential here and it’s exciting that an investment in NBIS gives you this ‘moonshot’ optionality too.

ClickHouse

ClickHouse continues to grow at a fantastic pace and is starting to look like a fantastic business on its own. As of Jan 2026 it was valued at $15B and Nebius continue to own around a 25% stake. I see it as the highest potential subsidiary Nebius has.

Toloka

NBIS no longer holds voting control but continues to have a financial stake in Toloka. Not much was said about this but it is potentially a source of future funding still.

Subscribe now

The New Deals Since Q4 Results

Since the Q4 results, Nebius has added more big deals that both validate their build‑out and derisk the thesis even more. First, a $2B investment and partnership with Nvidia and then a new agreement with Meta worth up to $27B.

NVDA Partnership

On 11th March NVDA announced it will invest $2 billion in Nebius as part of a partnership to scale a full‑stack AI cloud platform and target more than 5GW of NVDA systems deployed by 2030 with early access to their next-gen Vera Rubin infrastructure.

In practice, I see this doing three things:

• It secures priority access to state‑of‑the‑art GPUs

• It validates Nebius as a business to attract investment from Nvidia

• It partly de‑risks the 2026 capex by aligning NBIS with the most important vendor in the ecosystem

META Supply Agreement

Five days later, NBIS announced another deal, a new five‑year supply agreement with Meta with a value of up to $27B.

Nebius will provide $12B of dedicated capacity based on one of the first large deployments of the NVDA Vera Rubin platform with delivery starting in early 2027.

Additionally, Meta has committed to purchase up to $15B of additional available capacity over the next five years but crucially, with Nebius intending to sell that capacity first to third‑party AI cloud customers, with Meta effectively acting as a backstop buyer for any remaining capacity.

This is huge a huge amount of revenue and reduces utilisation risk on future capacity but also preserves upside from higher‑margin demand - the fact META would agree to this shows just how high the demand is.

What These Show

These two announcements reinforce and de-risk two key pillars to the thesis:

1. Nebius’ is guaranteed access to cutting‑edge NVIDIA systems at multi‑gigawatt scale

2. Demand is huge and underwritten by long‑dated, multi‑billion‑dollar contracts, not just for dedicated capacity but also spillover.

Execution risk on filling future power and GPU requirements goes down while a larger share of that future revenue base is now contracted rather than merely hoped for. The stock rallied strongly on this news and I believe this is one of the key reasons why.

Valuing NBIS

Right now I believe the main factor limiting Nebius is the speed at which they get power active. Whatever they have, they generally sell, but building the data centres they need is not easy and takes time. Plus, there is also an element of speculation at this stage of course. They are not cash flow positive yet so you are buying under the assumption this changes and they can start making money off their huge amounts of revenue.

With this in mind, I believe the best way to value NBIS right now is to model the power ramp and apply certain multiples. These multiples are speculative but at this stage, I think it is the best we have.

Power Ramp

Let’s start with the latest guidance and targets for YE ‘26:

• Contracted power >3GW

• Connected power of 800MW to 1000MW

Active power, which is actually generating revenue, will be less than connected however I expect by year end, the connected power guide will have been increased such that active power ends up around the 1GW mark. Nevertheless, keeping conservative, let’s assume it isn’t and that active power is around 80% of the lower end of the connected power guidance. They previously guided for 220MW connected by YE ‘25 and delivered 170MW active so that ratio seems about right.

For my model I will assume 650MW active power by YE 2026 for my base case.

For 2027, we have to be more speculative because they have not provided guidance but these new deals can help us. To hit Nvidia’s 5GW target by 2030, NBIS needs to add ~1GW of capacity annually plus another $12B for Meta starting in 2027. Given they are planning to add at least 600MW of connected power in 2026, at least a similar increase in 2027 feels reasonable.

With this in mind, an active power of 1200MW by YE 2027 feels reasonable for a base case, possibly conservative.

For both years, I’ll provide a range of options to account for performing under or over my expectations.

ARR Per Megawatt

Next I need a ARR per MW to convert the power to revenue.

At year-end 2025 we have:

• ARR ~$1.2B

• Active Power = 170MW

• ARR/MW ~$7M/MW

For 2026, this metric will likely increase due to pricing, especially as new Nvidia Vera Rubin systems come online. In fact, using the lowest 2026 ARR guidance of $7B and that $7M/MW gives 1GW of power active. Since their target in the latest earnings is 800MW to 1000MW connected by year end, not necessarily active, this suggests they themselves are expecting higher prices than this $7M/MW.

This means an increase in ARR/MW is almost certain but to keep my valuations conservative, I will stick with the $7M/MW anyway.

Price/ARR Multiple

The next bit of the puzzle is a price/ARR multiple. I am using this in my model to get from a projected ARR number to a market cap.

At the time of writing, NBIS is ~$22.4B market cap and year end ARR was $1.25B giving a price/ARR of ~18. This is about as low as it has traded since the Q4 earnings.

Looking at some other points for reference:

• The lowest valuation YTD was ~$18.7B in February with a guided $1B ARR - a P/ARR of ~19x

• The highest valuation YTD was $32.7B mid-March (post Q4 earnings) giving a P/ARR ~26x

• However, using CoreWeave (CRWV) as a competitor example, their ARR from Q4 earnings was ~$6.3B and the lowest and highest market caps since are ~$33.0B and ~$46.0B - this gives a P/ARR of 5.2x to 7.3x

At first glance, this makes Nebius look expensive. However, CoreWeave funded its backlog by taking on over $14B in debt, whereas Nebius operates with net-cash. Because Nebius has a cleaner balance sheet, it can command a higher P/ARR multiple than CRWV.

With this in mind, for 2026 I will consider a range of multiples from 6x to 14x - I believe this captures largely disappointing execution while still conservatively capping the maximum multiple well below the lows of this year. I do expect some multiple compression as NBIS grows so I think this is reasonable.

For 2027 I will bring the range down by two to cover 4x to 12x.

Shares Outstanding

The final thing is shares outstanding. The recent NVDA deal has allowed them to purchase up to 21 million shares and I would not be surprised by further dilution despite them funding much of their 2026 capex already.

In fact, on 19th March, while writing this article, NBIS has issued another $4.3B in convertible notes which will add approximately 24M shares if fully converted.

By my calculations, as of April 2026, there is roughly the following shares outstanding:

• Q4 diluted: 286.6M

• NVDA deal: 21M

• New convertible notes: 23.8M

• Total shares outstanding ~331M

For 2027 I will assume a further 10% dilution.

Subsidiaries

For my valuations I am going to ignore any value from the subsidiaries because I will assume that they are all used for future funding. This and 10% dilution is very conservative but it helps me feel better about the large amounts of capex we know NBIS will require.

Final Values

Putting all of the above together, for 2026 I get the following per share values for year end.

These translate to the following returns from $100/share with negative returns highlighted red and ones under 20% highlighted amber.

So, this says that for NBIS to return 20% by year end, were you to buy at $100, it must keep a higher multiple than CRWV and let’s be realistic, if they miss guidance that might not happen. This for me rules out the higher multiples in the top row and makes the second row ones unlikely too. Based on this I think the more likely scenarios are those below:

The options I see are:

1. NBIS cannot meet guidance, the market punishes them heavily with a low multiple - you lose money

2. NBIS meets or maybe even beats guidance but the market punishes them with a low multiple, perhaps from a risk-off scenario - here you do OK

3. NBIS meets or beats guidance and the market is happy to assign a multiple close but still slightly less than today - you do very well

For me personally, already holding a decent NBIS position, I will add again at around $75 - here even most of the downside scenarios are small.

For 2027, we get a similar story too. Fall below guidance and you are probably going to lose some money but beat it and you could be rewarded very well.

This time the percentages in the second table are the CAGR you would get between today and year end 2027.

Wrapping Up

So in summary, Nebius had another great quarter and is continuing to execute beyond what expectations.

Growth is strong and demand still looks huge. We are starting to see some operating leverage and despite big capex numbers, much of it is already secured with several options to raise the rest. Deals are still coming in.

For me, the success of Nebius as a business and as an investment hinges on two things:

• Can they swing from their current state of high spending and losses to one where they are cash flow positive?

• Can they continue to execute and grow at the pace they expect?

So far, the answer to both of these seems more likely yes than no.

If it is yes, based on my rough valuations, I think they could be a fantastic investment. And remember, the valuations above were tilted very conservatively and used the more negative scenario nearly everywhere they could.

With all the optionality of the subsidiaries too, I think it’s more likely NBIS makes a good investment than does not.

Subscribe now

Thank you for reading and please subscribe if you would like to see more.

JM

If you have not seen yet, make sure to check out the past articles below too, they are still relevant!

Now let’s get straight to it.

dLocal (DLO)

dLocal is perhaps a lesser known stock but has long been one of my favourites and one I recently wrote a deep dive on.

Basically, they process payments across emerging markets to let people and businesses receive and make payments with local methods and currencies.

For example, say you are Uber - you are operating across Latin America and want to accept payments in their local currencies or via payment methods different to the US. There’s tens of countries & currencies here, even more payment methods and tonnes of local regulation on top and you also want to pay people as well as accepting payments. You don’t want to set this up yourself, that’s not your business. Instead you route everything through a central platform (dLocal), save all the headache and costs and pay a very small fee on the payment volume. Scale this either further across the world and you can see how one single platform can be attractive to global businesses.

dLocal are essentially a toll booth in emerging market payments.

If you would like more information, check out the deep dive below:

Fundamentals

I will not go too deep here since I have covered this thoroughly in my deep dive but I’ll list the highlights:

• Top line growth is consistently growing 30% YoY

• They are net profitable with fantastic operating margins and extremely asset light

• They have a formidable moat due to the combination of their global scale and local expertise - other global players can’t match the local expertise without lots of investment, local players can’t offer a global platform like large global businesses want

• They have a solid balance sheet and with the FCF they generate, they are able to self-fund growth and investment

• Emerging markets are going to experience huge growth over the next decade and have huge runways to keep growing - all of which flows through DLO.

Valuation wise, in my base case

• I calculate a fair value of ~$21, ~40% margin of safety from today’s value around $12

• I expect a return of >20% CAGR for the next five years.

Basically, they are a great business at a good price.

Technicals

Now this is where things get really nice because the technicals are lining up too.

DLO has not been immune to the recent sell off and today is trading around prices not seen since August 2025 when it rocketed from the $10s to the $15s after fantastic earnings.

As I write this, it closed at $11.60 last Friday.

Crucially it is now very close to the 100 week moving-average at $11.22. An average it has not revisited since crossing it back in August.

We also have confluence of support with the 21 month moving average at $11.27, although this is around 91 weeks so it’s not too surprising to be similar to the 100W.

Ultimately, I believe the $11.20 area will act as an area of strong support for DLO.

My Plan

Personally, I already have a big position in DLO and which makes me less likely to add more but even so, I am very interested here. Under $12 was always my target fundamentally and now that it is lining up with a key support level, it seems like a great time to buy.

With everything happening in the world right now, it is hard to say what the market will do but if you can buy under $12, I think you have a nice entry. If you can get closer to $11.20 then even better.

Personally, I will set a limit order around $11.40 and keep an eye on the market. If I did not already have a big position I would probably just buy at market open.

If I were to use a stop loss, I would go for around $10.90 so there is the additional whole number support of $11. At a worst possible buy price of $12, you would be risking about 9%. Daily ATR is ~$0.56 so you should be safe from daily volatility but weekly ATR is ~$1.36 so I would monitor the position throughout the week.

Personally, I will not use a stop loss though, since I am already holding DLO long term with a lower average.

Overall, I think this is a brilliant opportunity, one of my favourite in a while.

Subscribe

Others

In terms of other stocks I am watching, I am actually still watching, all three that I wrote about in my 16th February top stocks article:

• Reddit (RDDT)

• Ero Copper (ERO)

• RocketLab (RKLB)

RDDT I already bought and would potentially add to but it closed just under support last week so I want to see how it performs over the next few days first.

RKLB and ERO I bought but sold with a stop losses. I do not like to rush back in and re-buy when that happens but it’s been long enough now, and I am happy enough with the fundamentals, that I will consider it.

I will not go into detail on the thesis behind each one here, to avoid repeating myself, but I recommend checking out the article below:

Here are the key levels I am watching for ERO and RKLB:

ERO

• Buy: under $27.50

• Support: 100D ~$27.15 and 21W ~$26.95

• Stop loss: $24.90, max loss of 10%

RKLB

• Buy: under $70

• Support: 100D ~$66.40 and 21W ~$66.70

• Stop loss: $62.90, max loss of 10%

That’s it for this one. Please subscribe if you found it useful or interesting and thanks for reading.

JM

If you missed last week’s article, you can check …

Read more

Root release their Q4 earnings next week, on Wednesday 25th Feb, after market close. Here is what I am expecting based on information already available.

Read more

Assuming they don’t open at drastically different pri…

Read more

Read more

For many investors, including myself, this seemed overdone & repres…

Read more