Snowflake’s product revenue hit $1.09 billion in its fiscal 2026 second quarter, a 32% year-over-year jump that marks an acceleration from recent quarters — and a good chunk of that momentum came straight from Microsoft Azure. While the data cloud company’s entire narrative has pivoted toward AI, the Azure piece is the one that matters most to Windows-centric organizations and the IT professionals who run them. According to Snowflake’s latest figures, Azure was its fastest-growing cloud platform, with customer spend climbing 40% year-over-year, underscoring a deepening partnership that brings Snowflake’s analytics and AI tools directly into the Microsoft ecosystem that already powers so many enterprise Windows workloads.

What actually changed this quarter

Snowflake isn’t just collecting logos; its existing customers are spending more. Net revenue retention sat at 125%, meaning the typical customer expanded consumption by a quarter compared to the same period last year. The total customer count breached 12,000, and 654 accounts now generate more than $1 million in trailing 12-month product revenue — a figure that signals the platform is becoming embedded in large-scale, mission-critical operations. Remaining performance obligations, a proxy for contracted future revenue, also shot up by billions, providing better visibility into the company’s near-term trajectory.

The product portfolio grew alongside the top line. Snowflake pushed deeper into AI with Cortex AI SQL — an offering that lets users run machine learning models and large language model inference directly from SQL queries — and introduced Snowflake Intelligence, which surfaces AI-generated insights inside the data cloud. At the infrastructure level, Gen2 runtime optimizations and expanded Postgres compatibility made the platform friendlier for transactional and mixed workloads, not just pure analytics. The company also detailed a collaboration with Siemens to bridge operational technology (OT) data from factory floors into Snowflake’s governed AI Data Cloud, aiming to turn industrial edge telemetry into actionable intelligence.

All of this aligns with Snowflake’s stated goal: becoming the default layer where enterprises store, govern, and activate data for AI. But for Windows and Azure shops, the standout number was the 40% Azure growth. It signals that Microsoft and Snowflake are no longer just coexisting — they’re increasingly complementary in the field.

What the Azure surge means for Windows users and IT admins

If your organization runs Windows Server workloads, manages hybrid infrastructure through Azure Arc, or relies on Active Directory for identity, the Snowflake-Azure alignment matters in concrete ways:

  • Direct data connectivity: Snowflake runs natively on Azure and supports Azure Data Lake Storage, simplifying ingestion from Windows-based applications and SQL Server databases without complex ETL pipelines or cross-cloud data egress fees.
  • Unified identity and governance: Azure Active Directory integration allows single sign-on and role-based access control that maps to existing Windows authentication policies. For regulated industries, this reduces the compliance burden of adding a new analytics platform.
  • Power BI synergy: Many Windows-centric BI teams already use Power BI. Snowflake’s connectors for Power BI are mature, and the performance improvements in recent releases make live querying against large Snowflake datasets feel snappier than ever. According to Snowflake’s reports, Azure customers are driving a disproportionate share of that Power BI-Snowflake usage growth.
  • Multicloud flexibility without lock-in: Admins who manage both on-premises Windows Server environments and Azure cloud projects value the ability to run the same Snowflake interface across AWS and Azure. That portability hedges against single-vendor risk — a recurring concern in boardroom discussions about hyperscaler dependency.

For developers, the Cortex AI SQL offering is especially relevant. It lets data engineers and analysts who are fluent in T-SQL or Snowflake SQL invoke AI models without learning Python or spinning up separate model-hosting services. That lowers the barrier for Windows shops that want to add natural language processing or predictive analytics features to internal applications without hiring specialized ML engineers.

How we got here: Snowflake’s multicloud evolution

Snowflake launched a decade ago as a cloud data warehouse designed to decouple storage from compute. Its initial bets on AWS gave way to a deliberate multicloud strategy, with Azure support arriving in 2018 and Google Cloud Platform following later. In the years since, the company has expanded beyond warehousing into data engineering, data sharing, and now AI.

The Azure partnership didn’t become a headline item overnight. Early Azure deployments lagged behind AWS in both performance and feature parity, but Snowflake invested heavily to close the gap. The 40% Azure growth figure reported in the most recent quarter suggests that investment is paying off. Part of the acceleration owes to Microsoft’s own push: Azure sales teams increasingly position Snowflake as a preferred analytics partner for large migrations from legacy on-premises data warehouses like Microsoft APS or IBM Netezza. The co-sell motion — where Microsoft account executives bring in Snowflake during Azure deals — has strengthened, particularly in industries like manufacturing and financial services that run heavy Windows-based enterprise resource planning systems.

Simultaneously, Snowflake embedded AI capabilities directly into its platform rather than relying on external model connectors. Cortex AI SQL, released in public preview earlier this year, brings large language model inference inside the Snowflake execution environment. Snowflake Intelligence, meanwhile, acts as an AI assistant that can summarize tables, generate SQL queries, and explain data lineage. These features reflect the company’s conviction that AI will be consumed primarily where data resides — not in separate model-hosting services — and that the data platform with the strongest governance and security posture will win the lion’s share of AI spend.

What to do now: evaluating Snowflake in a Windows/Azure environment

For IT decision-makers and architects in organizations running Windows stacks, the news warrants a pragmatic reassessment of your analytics and AI roadmaps. Here are four actionable steps:

  1. Audit data movement patterns. Identify datasets that currently hop between SQL Server, Azure Data Lake, and maybe a third-party analytics tool. Calculate the latency and egress costs. Snowflake’s ability to sit natively on Azure and read from Data Lake Storage can collapse that chain, but only if the workload justifies the switch. For small datasets and simple reporting, Power BI on top of SQL Server might remain more cost-effective.

  2. Run a total-cost-of-ownership model with AI included. Snowflake’s consumption-based pricing is predictable for steady query workloads, but AI usage can spike unpredictably. Use the Azure Pricing Calculator to model storage and compute for a hypothetical year, adding a buffer for Cortex AI SQL experiments. Compare against building a similar stack with Microsoft Fabric, Synapse Analytics, and Azure OpenAI Service. Remember that Snowflake’s cross-cloud portability adds a layer of fiscal safety if your organization later decides to shift workloads to AWS.

  3. Pilot an AI workload with governance constraints. Choose a use case that touches sensitive data — customer support ticket analysis, internal document summarization, or predictive maintenance for Windows Server-managed OT devices. Implement it within Snowflake’s governance framework (row-level security, dynamic masking, object tagging) to see if the AI features operate cleanly within existing compliance rules. This pilot will surface not only performance characteristics but also how Snowflake’s identity integration with Azure Active Directory holds up at scale.

  4. Leverage vertical accelerators if you’re in manufacturing or heavy industry. Snowflake’s Siemens collaboration is a tangible template. If your factory floor runs Windows-based industrial PCs or gateways, inquire whether the Siemens Industrial Edge integration can funnel OT telemetry into Snowflake without building custom connectors. The goal is to reduce the time from sensor to analytics dashboard from months to days.

Keep an eye on the competitive landscape, too. Microsoft’s own Fabric platform is maturing rapidly, and for organizations that are all-in on Azure and Power BI, Fabric’s tighter integration with Teams, Office 365, and Copilot for Microsoft 365 may become difficult to ignore. Datadog similarly has extended from infrastructure monitoring into analytics and security, though its focus is more ops-centric than data-warehousing. The decision ultimately comes down to whether you need a cross-cloud, governance-first analytics platform optimized for AI (Snowflake) or a single-vendor stack with deep productivity-app alignment (Microsoft Fabric).

Outlook: what to watch next

The next six months will test whether Snowflake’s Azure-fueled growth is sustainable. Key indicators include sequential product revenue trends, the rate of adoption for Cortex AI SQL and Snowflake Intelligence (especially among existing Azure customers), and any announcements of deeper product-level integrations with Microsoft’s Fabric or Copilot ecosystems. On the industrial side, success stories from the Siemens partnership — such as a Fortune 500 manufacturer bringing a predictive quality use case into production — would validate the vertical strategy. For Windows IT pros, the signal to monitor is whether Snowflake’s Azure performance and feature set continue to keep pace with what Microsoft builds natively. Right now, the numbers suggest Snowflake has earned a seat at the table; the burden of proof lies in turning that seat into a recurring, mission-critical subscription.