Snowflake (NYSE: SNOW) reported second-quarter fiscal 2027 earnings after the close on September 2, and the market’s verdict was immediate: shares surged nearly 23% in after-hours trading, jumping from a $305.84 regular-session close to $375.51 in overnight trading.
That reversal is the story in miniature. Snowflake had actually sold off nearly 4.4% during the regular session heading into the print, with shares sliding under both their 50-day and 200-day moving averages as investors grew nervous ahead of a report that needed to clear an unusually high bar.
What it delivered was acceleration, not deceleration — product revenue growth actually sped up for a third consecutive quarter, defying the gravity that usually pulls high-multiple growth stocks back to earth as they scale. The report is a useful case study in how perception and fundamentals can drift apart within a single trading day, only to snap back into alignment the moment the numbers hit the tape.
Snowflake Earnings Beat Estimates as Revenue Growth Accelerates
Total revenue came in at $1.55 billion, up 35% year-over-year and comfortably ahead of the $1.49 billion Wall Street expected. Product revenue — the figure investors watch most closely, since it reflects actual platform consumption rather than one-time services work — grew 37% year-over-year to $1.49 billion. That marked a seven-percentage-point acceleration from the 30% growth rate Snowflake exited fiscal 2026 with, and the third straight quarter in which growth sped up rather than slowed.
Adjusted earnings per share landed at 62 cents, well above the consensus estimate of 46 cents and Snowflake’s fifth consecutive quarterly beat. On a GAAP basis, the company narrowed its net loss to $191.7 million, or $0.55 per share, down from a $297.9 million loss a year earlier. Profitability improved alongside growth: non-GAAP operating margin expanded to roughly 15%, up from 11% in the year-ago quarter, and management raised its full-year margin outlook to 14.5% from 13.5%.
The perception-versus-fundamentals tension around Snowflake has often centered on a simple question: is growth coming from genuine expansion, or from a shrinking pool of easy comparisons? This quarter answered it.
Net revenue retention held at 126%, meaning existing customers are spending well beyond their prior-year contracts. The company added 692 net new customers, up 32% year-over-year, bringing its total base to 14,554. Customers spending more than $1 million annually grew 27% year-over-year to 828, with 48 net additions in the quarter alone. Snowflake now also counts 829 Forbes Global 2000 customers, representing more than 41% penetration among the world’s largest enterprises.
Remaining performance obligations — a proxy for contracted future business — rose 30% year-over-year to $9.0 billion, with 54% expected to convert to revenue over the next twelve months. That combination of accelerating current revenue and growing forward-looking commitments is the piece that tends to move a stock double digits: it tells investors the growth isn’t just holding, it’s compounding.
Snowflake AI Growth Is Accelerating as Customers Expand Usage
Every software company now claims an AI story, and the market has grown skeptical of narratives without numbers attached. Snowflake’s AI products offered something more concrete. CoCo, its AI coding agent, closed the quarter with 9,100 accounts, up more than 2,000 from the prior quarter, while CoWork reached 5,800 accounts.
CEO Sridhar Ramaswamy described AI as compounding the company’s advantages, creating what he called a flywheel effect that pulls new workloads onto the platform and lifts overall consumption. Executives on the call attributed roughly half of the growth acceleration to AI-driven usage, spanning tools from AI functions and document processing to Cortex Code and machine learning notebooks. Production use cases climbed 89% year-over-year, a sign that AI experimentation is converting into workloads customers actually depend on rather than staying stuck in pilot mode.
That momentum came with a trade-off. Non-GAAP product gross margin came in at 75% for the quarter, down a point from a year earlier, and management’s full-year guidance for the metric moved lower still, to 74%, since AI workloads currently carry lower contribution margins than the core data platform.
Management framed this as a deliberate sequencing choice — prioritizing adoption and platform consumption now, with unit economics expected to improve as AI workloads scale. It’s the one line in the report that argues against a purely one-sided read: the AI flywheel is real, but it isn’t yet as profitable, workload for workload, as the business it’s layered onto.
Snowflake Raises Fiscal 2027 Guidance After Strong Q2 Earnings
Beats matter, but guidance is what typically decides whether a stock jumps 5% or 25%. Snowflake raised its full-year fiscal 2027 product revenue outlook to $6.07 billion, up from $5.84 billion previously, implying 36% annual growth. For the third quarter, management guided to product revenue growth of 37% to 38% year-over-year, above the $1.50 billion analysts had modeled in dollar terms.
Operating margin guidance moved higher too, even as product gross margin guidance moved lower — a combination that tells investors management is comfortable trading a few points of gross margin for faster adoption, while still finding operating leverage elsewhere in the business. That’s a more nuanced signal than a simple beat-and-raise, and the market treated it as credible rather than as a red flag.
SNOW Stock Surge Highlights the Gap Between Price and Fundamentals
Going into this report, the bear case on Snowflake was visible in real time: shares fell nearly 4.4% in the hours before the print, a sign that the market was bracing for disappointment after a year of gains that left little room for error. The fundamentals told a different story once the release hit. Growth didn’t decelerate — it accelerated, for the third quarter running, while margins expanded and forward bookings grew even faster than revenue.
When a company’s actual trajectory turns out to be stronger than the market’s cautious same-day pricing implied, the reaction tends to be sharp and immediate, which is exactly what a same-session round trip from a 4.4% decline to a near-23% afterhours pop represents. The question for Snowflake now isn’t whether the AI narrative is real — Wednesday’s numbers answered that — but whether the company can keep compounding acceleration on top of acceleration without eventually running into the same law of large numbers the bears were originally worried about.
