Is Databricks Really Winning Snowflake’s Workloads? What Buyer Signals Show

In June 2026, Databricks announced that its data warehousing business more than doubled over the past year to a $1.5 billion run rate. CEO Ali Ghodsi credited AI demand and customers switching over from other platforms. 

We put the claim to the people actually making these decisions, mining Qualitate’s 1,700+ Databricks expert discussions.

How do you test a vendor’s “we’re winning” claim against real buyers?

We ran the question through Qualitate’s Research Analyst: Databricks says it’s winning data warehousing workloads on AI demand. Is that supported by the buyer data? We asked the agent to look at where Databricks actually wins by workload, how Snowflake is defending with Cortex, and where ClickHouse fits in evaluations.

Research Analyst reasons across Qualitate’s library of structured discussions with senior technology buyers, the executives and architects who are evaluating, procuring, and running these platforms right now. Every figure below traces back to an underlying conversation you can open and read.

Is Databricks actually taking Snowflake’s warehouse workloads?

Databricks wins AI/ML, while Snowflake keeps warehousing. Databricks is winning AI/ML, data engineering, and lakehouse workloads. It is not displacing Snowflake across general-purpose data warehousing. 

Snowflake still owns SQL and BI. Buyers reach for Snowflake on SQL analytics, governed reporting, and ease of use for SQL-native teams. In regulated industries like insurance and financial services, its auditability and data sharing keep it in the lead seat.

Databricks wins on unstructured data, streaming, and built-in ML. Buyers point to one unified platform for data engineering, streaming, and ML, with native MLflow integration and production ML capabilities Snowflake couldn’t match.

Snowflake’s AI layer Cortex is showing up. Buyers cite Cortex for call-center transcript summarization, sentiment analysis, and agent orchestration. One picked it over Databricks, Airflow, and every other option for workflow orchestration. Heavy-ML buyers still call Snowpark and Cortex a step behind Databricks.

ClickHouse gets evaluations. Buyers consider it for billing and real-time analytics on cost grounds. They pass over data ingestion maturity and enterprise-scale confidence. One buyer who ran a broad evaluation summed up the field: “The main players are Snowflake and Databricks.”

Where can you see the full Databricks vs. Snowflake buyer breakdown?

The download includes the workload-by-workload breakdown for data warehouses and lakehouses, the Cortex adoption picture, the ClickHouse use cases, and the buyer quotes behind each finding.

What else can Qualitate’s Research Analyst answer?

This started as one question and returned a sourced answer in a single pass. You can point Research Analyst at any company or market in the Qualitate library and get the same depth, traceable to the discussions underneath.

When you need to go deeper than the existing library, you can commission a custom project. Qualitate’s AI Moderator runs structured voice discussions with senior buyers from our proprietary panel and delivers structured data back in 3 to 10 business days.

Connect with our sales team to transform your primary research workflow: qualitate.io/book-a-demo