Snowflake Just Had Its Best Quarter Ever

Most companies growing at 33% don’t also have a $9.21 billion backlog and a freshly signed $6 billion deal with AWS. Snowflake has both. And the stock is still sitting roughly 12% below its June high. That’s the part that keeps pulling me back to this one.

I want to be honest about something upfront: Snowflake has been a frustrating stock for a long time. Great business, complicated story, always one overhang away from a selloff. The bear case never fully went away. But the Q1 fiscal 2027 numbers that dropped in late May were different in a way that’s hard to dismiss.

Total revenue: $1.39 billion. Up 33% year over year. Product revenue grew 34%, beating consensus by a wide margin. Non-GAAP operating margin hit 11.9%, which is more than 300 basis points above what the company had guided entering the quarter. Non-GAAP EPS came in at $0.39, beating consensus by nearly 22%. Full-year product revenue guidance was raised to $5.84 billion — 31% growth, well ahead of the pre-quarter estimate of $5.66 billion. They also lifted the full-year non-GAAP operating margin target to 13.5%.

That’s a lot of numbers. Here’s what they actually mean: Snowflake used to be a growth story with a profitability question attached to it. That question is getting answered. Margin expansion isn’t showing up in the guidance slides anymore — it’s showing up in the actual reported figures. That’s a different thing entirely.

The stock surged roughly 38% the day after earnings. Then it pulled back. Snowflake hit a 52-week high of $284.99 on June 1, 2026, and by late June was trading near $251. That post-earnings drift lower happens a lot with high-growth names. Sometimes it’s rational. Sometimes it’s just the market cooling off after a big move. I’m not sure which this is yet.

Now the AWS deal.

This is where it gets interesting. Snowflake committed to a $6 billion, five-year infrastructure spend on Amazon Web Services — covering Graviton compute and AI services. It’s the largest AWS commitment in company history. For context: Snowflake has sold roughly $7 billion worth of its own services through AWS Marketplace since the company was founded in 2012. This one deal is nearly that entire figure, locked in for five years. That’s not a small footnote. That’s a strategic bet on where enterprise AI infrastructure is heading, and Snowflake is planting a flag in it.

CEO Sridhar Ramaswamy has been reshaping the company’s identity for about 18 months now. The old framing was data cloud. The new one is agentic enterprise. And the product momentum behind it is real: Snowflake Intelligence, the natural-language AI layer, more than doubled its account base quarter over quarter. Cortex Code — CoCo internally — is already live across more than 7,100 accounts. Management described them as the fastest-adopted products in company history. Whether that holds at scale is worth watching, but the early adoption curve is hard to argue with.

Slight tangent, but it matters: there’s been a running debate for two years about whether Snowflake’s consumption model is a feature or a bug. The bear case says consumption revenue is inherently volatile — customers dial it up and down based on budget cycles, and when enterprise spending tightens, it shows up fast. That’s a real concern. It’s not wrong. But the net revenue retention rate sitting at 126% suggests existing customers are still expanding, not contracting. 126% means the average customer spent 26% more than they did a year ago. That’s the model working.

Remaining performance obligations — contracted future revenue that hasn’t been recognized yet — came in at $9.21 billion, up 38% year over year. That’s a visibility number. It doesn’t move the income statement today, but it tells you how much is already committed and waiting to flow through. Revenue grew from $592 million in fiscal 2021 to $4.68 billion in fiscal 2026. Nearly eightfold in five years. The $9.21 billion RPO is essentially signaling the next chapter of that same compounding.

The competition is real and I won’t pretend otherwise. Databricks is growing sales north of 80% and targeting the same enterprise data buyer from a different architectural angle. AWS, Azure, and GCP are all bundling data tools that eat into Snowflake’s core use cases. The hyperscaler bundling threat in particular isn’t going away — if anything, it’s getting louder as cloud providers get more aggressive on enterprise contracts.

But Snowflake’s AWS deal cuts through some of that. You don’t commit $6 billion to a platform you’re trying to replace. That’s a partnership, not just a vendor relationship.

51 analysts currently covering the stock rate it a consensus Strong Buy, with an average 12-month price target around $292. The stock is near $251. That implied upside is real, even if analyst targets are often more directional than precise.

What I keep coming back to: the business has de-risked in ways the stock price doesn’t fully reflect. The profitability story is no longer theoretical. The AI product layer is in production. The backlog is massive. And the AWS relationship just got significantly deeper.

Whether that translates into the stock moving in the next two quarters or two years — that’s the actual question. I don’t have a clean answer to it. But the picture looks different than it did 18 months ago, and that’s worth paying attention to.