Snowflake director sells EUR 16.4m while Datadog keeps the richer multiple

Snowflake sits in a market that still has money for cloud infrastructure, but not for every cloud story. The backdrop is unusually specific right now. The four largest cloud providers reported combined backlogs above $2.3 trillion in the second quarter of 2026, and projected 2026 hyperscaler capex is above $860 billion, according to the cited research. That matters because Snowflake is not a generic software name trading on hope. It is a pure-play data platform provider, so when enterprise buyers keep funding AI workloads and cloud data warehousing, Snowflake gets pulled into the same spending current.

Datadog is the cleaner comparison than some random software peer because the market keeps pricing both names as beneficiaries of the same enterprise cloud spend, but with different valuation temperaments. Datadog still trades at a richer multiple, while commentary in the cited research says Snowflake’s forward EV/revenue multiple is more compressed. That is the tension. Snowflake has the stronger recent share price run, Datadog has the premium, and both live inside a market that is still rewarding growth, just not equally.

InsiderTrades data says the relevant historical bucket, director-level buys at mega-cap names, has a 90 day win rate of 54.9% and an average return of 3.9% over that horizon, with a 365 day average return of 58.07% across 3,030 cases. That is historical cohort data, not a forecast for Snowflake, and it is not a promise that this filing will behave the same way. It does, however, tell you that director activity in large names is not random noise in our sample.

Michael Speiser’s sale, and why the size matters less than the timing

Snowflake Inc. director Michael L. Speiser reported sales of 50,741 shares of common stock on August 7, 2026, at a weighted-average price of $324.06 per share for a total of approximately EUR 16.44 million, euro-normalised at ingest. The filing was submitted on August 11. The sale was executed under a pre-established 10b5-1 trading plan, which is the first thing to get straight before anyone starts pretending this is a fresh judgment call on the business.

Still, the timing is not nothing. Snowflake closed at $334.14 on August 11, after ending July at $293.28. That is a clean move higher in a short span, and Speiser’s sale came into that strength rather than into weakness. The filing value itself is tiny relative to the company, under 0.01% of market value, and InsiderTrades data puts the transaction at a negligible fraction of Snowflake’s roughly EUR 99.24 billion market cap. So this is not a balance sheet event. It is not a capital structure event. It is a director taking liquidity while the stock is working.

The market usually overreacts to the word “sale” and underreacts to the context around it. Here the context is doing most of the work. A 10b5-1 plan limits the interpretation, but it does not erase the fact that the sale arrived after a strong run and while the stock was sitting near the top of its recent range. If you own Snowflake, you care less about the mechanical label and more about whether the company is still getting the benefit of the doubt from the market. Right now, it is.

Datadog’s premium, Snowflake’s compression, and what the market is paying for

Datadog remains the cleaner example of a market that still pays up for cloud software with visible growth, but it does so selectively. The cited comparison notes Datadog’s higher multiple and Snowflake’s more compressed forward EV/revenue multiple. That gap matters because it tells you where expectations sit. Datadog is still priced like a premium operating asset. Snowflake is priced more like a growth name that has already had to prove itself once and is now being asked to prove it again.

That is why the peer frame matters more than the filing headline. If Snowflake were trading like a broken software name, a director sale would read one way. If it were trading like a fully loved momentum name, it would read another. Instead, it is somewhere in between. The stock is up 52.32 percent year to date through August 11, 2026, versus 12.89 percent for the S&P 500 over the same period, so the market has already paid for a lot of the AI and cloud narrative. Yet analysts have still been lifting targets, including moves to $360 and $340 in recent weeks, according to the cited market coverage.

That combination is what makes Snowflake interesting against Datadog. Datadog’s premium says the market still wants quality growth. Snowflake’s run says the market is willing to re-rate the data platform story when AI capex is the dominant macro theme. But the insider sale says the stock is no longer cheap enough to ignore the instinct to take some chips off the table. You do not need to make a drama out of that. You just need to notice it.

The cluster is the useful part, not the headline sale

InsiderTrades data flags the filing as part of a cluster. There were 2 distinct insiders and 12 recent declarations in the same name, including additional August 11 sales by Christian Kleinerman and multiple filings for Speiser. That is the part that deserves attention, more than the single director sale on its own. One director filing under a 10b5-1 plan can be routine. A cluster of recent sales from more than one insider is a different read, even when each trade has a mechanical explanation.

The recent activity matters because it shows the selling is not isolated. On August 3, EVP Christian Kleinerman sold 10,000 shares at prices around $294 to $300 under a separate 10b5-1 plan, according to the cited filing coverage. Then Speiser sold on August 7. Then the August 11 filings landed. That sequence does not prove a view on the business. It does show that multiple insiders were monetizing into the same window. In a stock that has already moved sharply higher, that is enough to make the filing worth more than a passing glance.

You should also keep the scale in perspective. The company is still a mega-cap by our internal classification, with a market value near EUR 99.24 billion. A EUR 16.44 million sale is real money for a human being, but it is not a statement about the company’s capital structure or near-term operating path. The cluster matters because it tells you how insiders are behaving around the stock, not because it changes the revenue model. That distinction is easy to lose when a headline says “director sells.”

Snowflake’s operating backdrop is better than its score, and that is the point

InsiderTrades data gives Snowflake a fundamental score of 34, with a quality score of 32 and a value score of 36. The rank sits at 21,356 out of 28,276. Those are not flattering numbers, and they are not meant to be. They tell you that the company is not screening as a pristine fundamental bargain. Yet the stock has still rallied hard because the market is looking through the current scorecard and paying for the strategic position inside AI-linked enterprise spending.

That is where the comparison with Datadog helps. Datadog’s premium multiple says the market is still willing to pay for software with durable growth and strong execution. Snowflake’s more compressed multiple says the market is more cautious about how much of that future is already in the price. The insider sale does not settle that argument. It sits on top of it. If anything, it reminds you that insiders can see a stock that has already done a lot and decide to realize some gains without making a grand statement about the company’s prospects.

The broader sector backdrop is still constructive. The cloud backlog and capex figures in the research are not abstract macro wallpaper, they are the reason Snowflake can keep trading like a beneficiary of the AI buildout. When hyperscalers keep spending, enterprise data platforms stay in the conversation. That does not mean every rally is sustainable. It means the market has a reason to keep giving Snowflake a seat at the table while it debates valuation against names like Datadog.

What the filing adds when the stock is already up 52.32 percent

Snowflake’s year-to-date gain of 52.32 percent through August 11 is the real backdrop to the filing. A stock that has already outpaced the S&P 500 by a wide margin can absorb insider selling better than a stock that is rolling over. That is why the price action matters first. The filing then tells you how insiders are behaving inside that price action. Speiser sold into strength. Kleinerman sold into strength. The cluster says the behavior is shared, at least across a small set of insiders.

The market has also been helped by the analyst tape, with recent target hikes to $360 and $340. That matters because it shows the sell-side is still willing to lean into the name even after the run. So you have a stock with a strong chart, a supportive macro backdrop, and a valuation debate that is still alive. Against that, a director sale under a 10b5-1 plan is not a thesis breaker. It is a reminder that the stock is no longer in the cheap part of its cycle.

If you want the practical comparison with Datadog, it is this. Datadog still carries the richer valuation, so the market is paying for more certainty there. Snowflake has the stronger recent share performance and the more obvious AI infrastructure tailwind, but it is also the name where insiders have been more willing to sell into the move. That does not make Datadog safer or Snowflake weaker by itself. It does tell you where the market is asking for proof.

The next print matters more than the filing

The filing is useful because it shows how insiders are acting while the stock is strong. It is less useful as a standalone forecast. Snowflake still has to keep converting the AI and cloud spending backdrop into actual operating progress, and the market will keep comparing that progress with peers like Datadog. If Snowflake keeps holding near the highs while the peer multiple debate stays open, the insider sales will fade into the background. If the stock stalls, the same filings will look more like a warning that insiders were happy to sell into a good window.

For now, the cleanest way to read it is against the price and the peer set. Snowflake is up sharply, the sector backdrop is supportive, Datadog still commands the richer multiple, and a director just sold EUR 16.44 million worth of stock under a 10b5-1 plan while a cluster of recent insider sales was already on the tape. That is enough to keep the name on your screen, not enough to force a conclusion before the next operating update.

This is not investment advice.