Snowflake AI Data Cloud Gains Momentum Across Multicloud and AI Workloads

Snowflake’s second-quarter results and product momentum make a strong case that the company’s “AI Data Cloud” narrative is moving from promise toward measurable commercial traction, but the climb to durable dominance in cloud analytics still runs through a crowded field of well‑capitalized incumbents and razor‑thin execution margins.

Background​

Snowflake reported second‑quarter fiscal 2026 product revenue of about $1.09 billion, a 32% year‑over‑year increase that marked an acceleration from the company’s recent trend of slowing growth. Key customer and usage metrics also moved in the right direction: net revenue retention sat at a healthy 125%, the installed base expanded to just over 12,000 customers, and Snowflake counted 654 customers generating more than $1 million in trailing‑12‑month product revenue. The business’s forward momentum was bolstered by an expanded guidance range and a material lift in remaining performance obligations — a proxy for contracted future revenues — which climbed into the multi‑billion dollar range.
These topline dynamics arrived alongside a clear product narrative pivoted to AI: Snowflake has been embedding Cortex AI SQL, Snowflake Intelligence and other AI‑centric capabilities directly into its platform, positioning the company as a data layer optimized for both analytics and model‑driven applications. The vendor is also doubling down on ecosystem partnerships — most notably with Microsoft Azure — and pursuing vertical use cases such as industrial OT/IT convergence with Siemens.
The combination of accelerating revenue growth, deepening customer spend, and explicit AI positioning is refreshing for investors and customers alike. Yet the underlying picture is nuanced: valuation remains rich, competition is intense, and the path from AI interest to predictable, long‑term monetization still needs to clear several operational and market risks.

---

Financial performance: what the numbers actually say​

Growth re‑accelerating but still measured​

• Product revenue: ~$1.09 billion, up 32% year‑over‑year — an acceleration from the company’s prior quarters of lower growth rates.
• Net revenue retention: 125%, indicating existing customers are expanding usage at a healthy clip.
• Customers: total base exceeded 12,000, a year‑over‑year increase that shows steady acquisition as well as expansion.
• Large customers: 654 accounts with >$1M in trailing 12‑month product revenue — a critical cohort for durable high‑value revenue.
• Remaining performance obligations: rose significantly, reflecting presold or committed future usage on the platform.

Operationally, Snowflake also reported improved non‑GAAP operating margin and solid product gross margins. The company is showing a movement toward operating leverage — not yet full profitability on a GAAP basis, but better unit economics as product revenue scales.

Why these metrics matter​

• Net revenue retention (NRR) at 125% is a strong indicator: it means customers already on the platform are buying more services, which is crucial for a consumption‑based vendor. Sustained NRR above 120% historically correlates with a durable enterprise franchise in the software sector.
• The >$1M customer cohort is a bellwether. Growth in this cohort suggests an expanding footprint inside large enterprises; crossing into the $1M band typically implies multi‑workload or mission‑critical adoption.
• Remaining performance obligations (RPO) growth provides forward visibility into contracted demand, which matters for forecasting revenue trajectories in a consumption model.

Taken together, these indicators show Snowflake is converting product innovation and go‑to‑market activities into measurable commercial outcomes. That’s the “proof point” investors wanted to see after months of product‑led AI positioning.

---

Product and AI strategy: built for analytics-first AI workloads​

Deepening the product stack​

Snowflake’s positioning is not just “data warehouse in the cloud” anymore. The platform includes a mix of capabilities aimed at:

• Unified storage and compute for structured and semi‑structured data
• Native SQL‑centric AI primitives (for example, Cortex AI SQL)
• New runtime and warehouse optimizations (Gen2 and related performance features)
• Postgres compatibility and transactional features to broaden workload fit
• Integration fabrics (OpenFlow/Open APIs) to enable cross‑cloud and partner workflows

These capabilities target a critical market need: enterprises want the ability to run analytics, data engineering, and AI model inference/training close to their data without constant data movement between systems.

AI as a growth lever — real or marketing?​

Snowflake’s public commentary and product releases emphasize two claims: AI is influencing new customer wins, and a non‑trivial portion of existing use cases involve AI. Reported adoption figures — thousands of accounts using AI features weekly and a substantial share of new logos citing AI as a reason to buy — suggest that AI is material to pipeline and adoption.
However, there are important nuances:

• AI workloads vary dramatically in resource intensity and predictability. Some use cases (lightweight inference, text embeddings) scale easily on a consumption model. Others (large‑model training, real‑time multimodal inference) can be cost‑intensive and sensitive to infrastructure choices.
• Snowflake’s differentiation is strongest when customers want an SQL‑native path from data to model inference or when enterprises value governance, security, and data sharing at scale. For pure model hosting or experimentation, some customers will still choose specialized model platforms or hyperscaler services.

Overall, Snowflake is positioning itself as an AI data plane — a place where data lives, is governed, and powers AI workflows. The commercial impact will depend on how many customers move beyond proof‑of‑concept AI experiments to sustained, productionized model usage that drives ongoing consumption.

---

Ecosystem and partnerships: multicloud as a strategic lever​

Microsoft Azure and other cloud partners​

Snowflake’s multicloud strategy allows enterprises to run Snowflake on AWS, Azure, and other clouds. Recent quarters show:

• Microsoft Azure was a standout growth channel for Snowflake, with customer spend on Azure growing substantially faster than other clouds in the reported period.
• Product‑level integrations (analytics + BI + OneLake/Power BI interactions) and stronger field alignment with Microsoft are cited as drivers.

The multicloud architecture is a two‑edged sword: it widens addressable market and reduces single‑vendor risk for customers, but it also increases engineering and go‑to‑market complexity. Maintaining tight platform parity, performance, and integration with each cloud provider is resource‑intensive.

Vertical partnerships: Siemens and OT/IT convergence​

A significant strategic move is Snowflake’s collaboration with industrial incumbents to bridge operational technology (OT) at the edge with IT and cloud analytics. By integrating Siemens Industrial Edge with Snowflake’s data platform, manufacturers can stream shop‑floor telemetry into governable, queryable datasets for AI‑driven analytics and automation.
This kind of verticalization matters because:

• It creates differentiated use cases that are harder for general cloud providers to displace.
• OT/IT projects often involve long sales cycles and high switching costs once implemented.
• Industrial AI workloads can be stickier and more mission‑critical, potentially increasing the lifetime value of customers.

The Siemens tie‑up is one example; similar vertical plays across healthcare, financial services, and retail will be essential for broadening Snowflake’s enterprise moat.

---

Competitive landscape: big incumbents and specialized challengers​

Snowflake competes in a crowded field. The main dynamics:

• Hyperscalers (Google Cloud, AWS, Microsoft Azure) are aggressively building integrated analytics + AI stacks. Google’s BigQuery, Microsoft Fabric/Synapse, and Amazon Redshift/SageMaker portfolios combine data management and model services, sometimes at scale and at price points that can be compelling for large customers.
• Observability and monitoring vendors (Datadog and others) are extending from telemetry to analytics and security, which creates adjacency competition for certain workloads.
• Newer AI infrastructure vendors and model platforms offer model hosting, feature stores, and MLOps capabilities that enterprises may pair with or substitute for Snowflake depending on needs.

Notably, Google Cloud’s analytics growth and profitability improvements have made BigQuery a credible alternative for enterprises that prefer a hyperscaler‑native approach. Likewise, Datadog’s expansion into security, observability, and AI agents creates overlapping value propositions for customers seeking unified operational analytics.
The competitive reality: Snowflake’s best defense is product differentiation (governance, cross‑cloud portability, data sharing), vertical use cases (OT/IT, manufacturing, finance), and partnerships that deepen integration with customers’ strategic platforms.

---

Valuation and market reaction​

Snowflake’s stock performance has been strong year‑to‑date, significantly outperforming broader tech indices in the reporting period. That market reaction reflects:

• Investor enthusiasm for AI narratives, especially when tied to revenue acceleration.
• Durable monetization signals: rising NRR, expansion of million‑dollar customers, and elevated RPO.
• An improved path to non‑GAAP profitability driven by operating leverage.

At the same time, Snowflake is trading at a premium multiple relative to the Internet software group based on forward price‑to‑sales metrics. Valuation models from different data providers show Snowflake at a high forward P/S ratio — meaning future revenue expectations are richly priced into the shares. The company’s fundamental story must therefore deliver consistent execution and margin improvement to justify current valuation levels.
Two important caveats about valuation:

• Multiples are volatile and sensitive to near‑term growth beats or misses; a single quarter of underperformance could compress the multiple meaningfully.
• Third‑party model scores and “value” grades are methodology‑dependent. Proprietary ratings or value scores should be interpreted as one input among many, not a sole determinant for investment decisions.

---

Opportunities: where Snowflake can win​

• AI‑driven workload monetization: If Snowflake can convert AI experimentation into sustained production workloads, consumption will scale. Features that make it easy to run inference, manage embeddings, and operate agents inside governed data contexts are high‑leverage.
• Multicloud portability: Enterprises with a multicloud strategy will value a cloud‑agnostic data layer that reduces vendor lock‑in and centralizes governance.
• Vertical depth: Industrial, healthcare, and financial services use cases — especially where OT/IT integration or regulated data governance is required — provide higher switching costs and stronger revenue durability.
• Ecosystem partnerships: Deep integrations with BI, MLOps, and cloud platform partners can turn Snowflake into the orchestration layer for complex enterprise AI stacks.

---

Risks and headwinds​

• Intense competition: Hyperscalers continue to integrate data, analytics, and AI offerings; their pricing power and scale can put pressure on Snowflake’s growth and margins.
• Execution complexity: Maintaining feature parity and performance across multiple clouds while expanding AI capabilities is costly and operationally difficult.
• Consumption model variability: Snowflake’s revenue depends on customer usage. AI workloads are unpredictable; spikes and troughs in usage can make revenue lumpy.
• Valuation sensitivity: Given the premium multiple, any slowdown in growth or unexpected margin erosion could result in material share price downside.
• Macro and budget cycles: Enterprise IT spending is still subject to macro shifts and budget cycles; customers may delay large AI projects in uncertain conditions.

Flagged uncertainty: some forward‑looking projections, third‑party “value” scores, and consensus EPS figures are derived from analyst models that change frequently. These are useful for context but should be treated as estimates rather than certainties.

---

Practical checklist for enterprise buyers and IT leaders​

• Map workloads: Identify which analytics and AI workloads require cross‑cloud portability versus hyperscaler‑native features.
• Evaluate total cost of ownership: Model not only software costs but also data egress, storage, and inference compute across clouds.
• Prioritize governance: If data sharing, compliance, or enterprise lineage are key, Snowflake’s centralized data governance capabilities are differentiators.
• Pilot AI at scale: Move beyond single‑model POCs to evaluate operational costs, latency, and retraining regimes in production.
• Consider vertical accelerators: Industrial or healthcare integrations can speed time‑to‑value and reduce integration risk.

---

Investment considerations: a balanced view​

• Bull case: Snowflake successfully converts AI interest into recurring, high‑value consumption; multicloud momentum and vertical partnerships drive stickier, higher‑margin revenue; operating leverage sustains margin expansion.
• Bear case: Hyperscalers neutralize Snowflake’s differentiation through tighter integrations and pricing pressure; AI spending proves episodic and fails to generate predictable consumption growth; valuation re‑rates lower as growth decelerates.

For long‑term investors, the company’s core strengths — platform breadth, governance, and a growing cohort of large customers — argue for a structural opportunity. For shorter‑term investors, the premium valuation and sensitivity to execution and macro cycles counsel caution.

---

Outlook and what to watch next​

Key near‑term indicators that will signal whether Snowflake’s momentum is durable:

• Sequential product revenue growth and guidance trajectory in the coming quarters.
• Net revenue retention trends — sustained NRR above 120% would be a clear positive.
• Growth and profitability of the >$1M customer cohort — expansion here signals entrenchment.
• Adoption and billed usage trends tied to AI features (frequency of AI use, productionized AI workloads).
• Execution of multicloud integrations, and measurable outcomes from strategic partnerships (for example, Azure momentum and vertical wins like Siemens).

If Snowflake continues to raise guidance and show consistent customer expansion tied to AI workloads, the growth thesis strengthens materially. If usage flickers or hyperscalers capture the bulk of AI‑native production workloads, Snowflake will need to lean more heavily on verticals and ecosystem depth to sustain its premium valuation.

---

Conclusion​

Snowflake’s recent results are a clear, tangible step in the company’s evolution from a cloud data warehouse provider to a platform for enterprise AI workloads. Revenue acceleration, healthy net revenue retention, rapid growth in high‑value customers, and a slate of AI‑centric product enhancements paint a promising picture for the AI Data Cloud narrative.
Nevertheless, the road ahead is not without friction. Competition from hyperscalers and observability platform vendors, heavy engineering requirements for multicloud parity, and the challenge of turning AI interest into consistent, long‑term consumption are realistic constraints that temper enthusiasm. Valuation already prices in a strong execution outcome; Snowflake must therefore deliver on product adoption, operational efficiency, and sustainable expansion within large accounts to justify the optimism.
In short: Snowflake’s growth thesis looks stronger today than it did a year ago, driven by AI‑led adoption and strategic partnerships, but the company still faces a demanding market test — converting rapid interest into durable, margin‑accretive revenue at scale.

---

Source: The Globe and Mail SNOW Expands in Cloud Analytics: Is the Growth Thesis Strengthening?