Snowflake Inc. -- How the Business Works

Snowflake operates a cloud data platform built around separation of storage and compute, enabling data warehousing, data lake, data sharing, and AI/ML workloads across AWS, Azure, and GCP. The company runs a consumption-based pricing model -- customers pay per compute credit consumed, with no long-term license lock-in. Product revenue (~95.5% of total) reached $4.47B in CY2025, growing 29% YoY. Snowflake is investing heavily in AI via Cortex (LLM inference, fine-tuning) with 9,100+ accounts using AI features and a $200M OpenAI partnership. The platform serves 13,300+ customers including 740 Forbes Global 2000 companies. Net revenue retention of 125% and RPO of $9.77B (+42% YoY) demonstrate strong customer commitment. The stock trades at $151.85 with a composite score of 6.0/10 (HOLD). Key structural concern: Snowflake holds ~18-21% of cloud DW by revenue in a 5-way oligopoly -- strong position, but no dominant share.
CY2025 Free Cash Flow
$1.12B
Positive FCF despite GAAP losses
Price / Composite Score
$152 / 6.0
HOLD | RSI 35.6 near oversold
Remaining Perf. Obligations
$9.77B
+42% YoY | Strong forward visibility
Net Revenue Retention
125%
Down from 126% prior year | -1pp
How Snowflake makes money -- consumption-based cloud data platform
The Snowflake Business Model -- Revenue Engines
Product Revenue (Consumption)
95.5% of Revenue
$4.47B CY2025 | +29.2% YoY
Pay-per-credit on AWS, Azure, GCP
72% product gross margin | 13,328 customers
Professional Services & Other
4.5% of Revenue
$211.6M CY2025 | +29.1% YoY
Implementation, training, consulting
Supports platform adoption and expansion
Cortex AI / Intelligence
9,100+ Accounts
AI influenced ~50% of new logos in Q2 FY26
LLM inference, fine-tuning, Cortex Code (4,400 accts)
$200M OpenAI partnership | Anthropic + Gemini
Platform Flow -- Data Ingestion to AI-Powered Insights
Data Ingestion
Warehouse + Lake + Iceberg Tables
Compute & Analytics
Consumption scales with usage
Data Sharing / Marketplace
40% of customers use sharing
Cortex AI / Intelligence
AI on governed enterprise data
Consumption-based pricing is a double-edged sword: Snowflake wins by decoupling storage from compute, offering cross-cloud portability (AWS, Azure, GCP), and providing a unified platform for warehousing, data engineering, and AI workloads. The consumption model aligns cost with usage -- customers pay per credit, with no long-term licensing lock-in. This drives land-and-expand dynamics: 733 customers now spend >$1M annually (+26% YoY) and 56 spend >$10M (+44% YoY). However, the same model means customers can scale down easily during optimization cycles. Snowflake is a price-taker competing against hyperscaler-native offerings (Redshift, BigQuery, Fabric) that bundle pricing into broader cloud commitments. The competitive position is earned through product quality and cross-cloud neutrality, not structural lock-in.
Revenue segmentation from Snowflake earnings reports and investor presentations via Daloopa. FYE January 31. CY2025 = FY2026.
Revenue and customer metrics -- reaccelerating at $4.7B scale
Annual Revenue and Key Customer Metrics (CY = Calendar Year)
Metric CY2024 (FY25) CY2025 (FY26) YoY Change
Product Revenue $3.46B $4.47B +29.2%
Professional Services $164.0M $211.6M +29.1%
Total Revenue $3.63B $4.68B +29.2%
Product Gross Margin 71% 72% +1pp
Total Customers 11,159 13,328 +19.4%
Customers >$1M TTM Rev 580 733 +26.4%
Customers >$10M TTM Rev 39 56 +43.6%
Net Revenue Retention 126% 125% -1pp
Remaining Performance Obligations $6.9B $9.77B +41.6%
Financial data from Snowflake earnings reports via Daloopa. FYE January 31. CY2024 = FY2025, CY2025 = FY2026.
Geographic revenue -- US dominant, APAC growing fastest
Revenue by Geography (CY2025 / FY2026)
Region CY2024 Rev CY2025 Rev % of Total YoY Growth
United States $2.76B $3.52B 75.2% +27.6%
EMEA $574.7M $763.7M 16.3% +32.9%
Asia-Pacific & Japan $188.0M $271.0M 5.8% +44.1%
Other Americas $101.9M $125.3M 2.7% +22.9%
Geographic revenue from Snowflake earnings reports via Daloopa.
Competitive position -- contested 5-way oligopoly
Cloud Data Warehouse Market Share (Oligopoly Gate: BORDERLINE / CONDITIONAL PASS)
Measurement SNOW Share Redshift BigQuery Databricks Source
Adoption-Based (6sense) 20.8% 14.1% 13.5% -- 6sense 2026
Workload-Based (DataPro) ~35% ~20% ~25% -- DataPro 2025
Revenue-Based (Synergy) 12-15% -- -- -- Synergy / SparkCo
Cloud DW TAM (2025E) $12-15B (narrow) | $120-125B (AI Data Cloud, SNOW-defined) Mordor / SNOW IR
The oligopoly gate is borderline -- not a clear pass: Snowflake is the largest pure-play cloud data warehouse vendor, but market share ranges from 12-35% depending on methodology. The variance reflects a measurement challenge: Snowflake is pure-play while competitors (Redshift, BigQuery, Fabric) are bundled within hyperscaler ecosystems. The market is a genuine 5-way contest. Databricks has converged to comparable ARR (~$5B) while growing nearly twice as fast (~55% vs. ~29%), representing a credible share threat. A large enterprise migration typically takes 6-18 months -- replacement is feasible but costly. Snowflake is a price-taker: it must continuously deliver superior value to justify pricing relative to bundled hyperscaler alternatives.
Market share data from 6sense, DataPro, Synergy Research, and SparkCo. TAM estimates from Mordor Intelligence and Snowflake Investor Day.
Competitive landscape -- cloud data platform peers
Company ARR / Revenue Rev Growth EV/Revenue Positioning
Snowflake (SNOW) $4.68B +29% ~10.8x Pure-play cloud DW, cross-cloud, Cortex AI
Databricks (Private) ~$5B ARR ~55% ~20x (private) Lakehouse model | AI/ML-first | 2x valuation
AWS Redshift -- -- -- Bundled in AWS ecosystem | pricing advantage
Google BigQuery -- -- -- Serverless analytics | GCP ecosystem
Microsoft Fabric/Synapse -- -- -- Rapidly growing via Azure + Office 365 co-sell
Snowflake and Databricks ARR from company disclosures and SaaStr. Hyperscaler DW revenue not separately disclosed. Databricks valued at $100B (Sep 2025) vs. SNOW ~$50B market cap.