Loading today’s stories
Invest1 distinct publisher3 min readPublished
Product revenue rose 30% last quarter while remaining performance obligations rose 42%. The gap suggests AI budgets are being committed to the data platform rather than routed around it.
The Investor · Invest desk
Compiled by The InvestorSomething wrong?How this is made
Backlog growing twelve points faster than revenue is the part of the quarter worth arguing about [1]. Contracted obligations now sit at 1.73 times the full-year product revenue Snowflake has guided for FY2027 [2], which means most of what customers have signed will be consumed in years the company has not yet forecast. If AI budgets were routing around the data layer, this is not the shape the paper would take.
The near-term number is more sober than the backlog. FY2027 guidance calls for 27% product revenue growth, and about a point of that comes from Observe, bought for roughly $600 million in cash and stock, so the organic guide is nearer 26% [4][5][3]. The first quarter is guided to a midpoint about 2.8% above the quarter just reported [4]. Whatever the backlog acceleration signals, management has not put it into the next twelve months.
Then margin. Non-GAAP product gross margin was 75.8% in FY2026 and is guided to 75% [9], with management attributing the pressure to newly launched AI products that carry lower margin profiles [10]. The AI dollars arriving are dilutive at the gross line, yet operating margin is guided from 10.5% to 12.5% [11][4], a two-point gain against a 0.8-point gross margin give-back [10]. The funding is visible in the comp line: stock-based compensation fell from 41% of revenue to 34% and is guided to 27%, two consecutive seven-point cuts [12][5], and Q4 net headcount additions were 37 after a reduction in force affecting about 200 people [13].
Adoption is broad and shallower than the headline count implies. More than 9,100 of the 13,300-plus accounts touch an AI offering, roughly 68% [6][7][6], while Snowflake Intelligence sits above 2,500 accounts, about 19% [7]. Net revenue retention held at 125% [16], and the committed money is concentrated at the top: 56 customers now spend above $10 million on a trailing twelve-month basis, up 56%, which implies about 36 a year earlier [8][8].
CEO Sridhar Ramaswamy told the call the company is "rapidly transforming from the platform for governing and analyzing data into the platform where customers build and run AI-native applications and workflows" [17]. The evidence for capture rather than disintermediation is the commitment itself, plus native access to third-party models inside the platform, including a $200 million expansion with OpenAI [19]. The evidence pointing the other way is cash: adjusted free cash flow margin is guided down from 25.5% to 23%, and only 1.5 points of that is attributed to Observe, leaving a point unexplained by the acquisition [15][9].
A sourcing caveat worth keeping: these are management's figures as summarized from the February 25, 2026 call by The Motley Fool [18], not audited detail on how much of the backlog is AI-specific. The disintermediation question gets answered by conversion, and conversion shows up in FY2028 disclosures, not this one.
Ranked by verification strength, evidence, and original report placement.
Remaining performance obligations were $9.77 billion, up 42% year over year, with growth acceleration noted for the second consecutive quarter.
FY2027 guidance is product revenue of $5.66 billion, 27% growth, with non-GAAP operating margin guided at 12.5%.
More than 9,100 accounts use Snowflake AI offerings, and Snowflake Intelligence is deployed by over 2,500 accounts, almost doubling sequentially from the prior quarter.
FY2026 non-GAAP product gross margin was 75.8%, with FY2027 guidance at 75%.
Q1 FY2027 guidance is product revenue of $1.262 billion to $1.267 billion, 27% growth, with non-GAAP operating margin guided at 9%.
Snowflake Q4 FY2026 product revenue was $1.23 billion, 30% year-over-year growth, driven by contributions from both the core business and AI workloads.
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Precise but single-sourced issuer disclosure
Every figure is specific, internally consistent and drawn from a primary artifact — the company's own earnings call, transcribed by the publisher. That is strong for the numbers themselves. It is weak as independent evidence: one publisher, one document, no press release, filing, customer testimony or third-party analysis in the cluster, and no disclosure of AI-specific revenue or RPO duration that would let the story's central inference be tested.
Broad disclosed adoption, no independent confirmation
Adoption signals are unusually concrete for an AI story: 9,100+ accounts on AI offerings (~68% of customers), Snowflake Intelligence at 2,500+ accounts and nearly doubling sequentially, 13,300+ total customers, 733 customers above $1 million and 56 above $10 million, 125% net revenue retention, a $400 million-plus contract and seven nine-figure deals, plus GA of Cortex Code, OpenFlow and Postgres. The discount is that all of it is vendor-reported, 'using AI offerings' is undefined, and no AI revenue is broken out.
Numbers accurate, causal reading unproven
Modestly overstated. The financial facts are reported faithfully, but the framing that backlog growth demonstrates AI budgets being committed to the data platform is an interpretation the disclosure cannot confirm: no AI revenue breakout, no contract-duration data, and RPO can inflate on longer terms or a handful of mega-deals such as the $400 million contract. Guidance also cuts against the strongest reading — 27% FY2027 growth (~26% organic) and a Q1 midpoint only ~2.8% above Q4 imply management is not translating 42% backlog growth into accelerating revenue. Gross margin and free cash flow compression further show the AI mix currently costs more than it earns.
Issuer-controlled numbers via investor-media transcript
High incentive load. Every fact originates from management on an earnings call, where the disclosure set, the non-GAAP definitions (product gross margin, adjusted free cash flow margin, SBC-excluded operating margin) and the adoption metrics are all chosen by the company; the CEO quote is explicit platform repositioning. The single distributor is a retail-investment publisher whose transcript pages sit alongside its subscription products, and it adds takeaways and a favorable summary rather than adversarial scrutiny. No counterweight source exists in the cluster.
Facts firm, interpretation thin
Confidence is moderate. The reported figures are highly reliable as statements of what management disclosed, and the arithmetic derivations follow directly from them, so the factual layer is firm. Confidence is held down by the absence of any second publisher or independent verification, a truncated transcript that omits the Q&A, no AI revenue or contract-duration disclosure, and the six-month lag between the February call date and this cluster's publication timestamp.
invest
Box's billings ran eight points ahead of revenue. The AI bill was 20 basis points.1 distinct publisher
product
Rillet's $100M reads as proof mid-market ERP is rip-and-replace, mostly at the cheap end1 distinct publisher
product
A 2x LLM bill is not a bug report: token spend is an observability problem1 distinct publisher
build
Google's legal AI bundle lands a day after a $40M model, and the connector list tells you why2 distinct publishers
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 26, 2026