Databricks Edges Snowflake on Spending Momentum

ETR Observatory for Cloud Data Warehouses  

This is a summary of the full report. Use the request form at the bottom of the article to download the complete findings.

Organizations have long relied on data warehouses to consolidate data from disparate sources into a single, governed location with a common schema, making possible the reporting and analytics work on which modern businesses run. Data lakes address a wider range of structured and unstructured data, and the lakehouse has emerged as the hybrid paradigm combining strengths of both. Across all three paradigms, storage and management have continued to shift from on-premises estates into the cloud, and the explosion of AI use cases has turned a brighter spotlight on data management strategies.

The major public cloud providers remain prominent players, pairing their warehousing offerings with tightly integrated data, analytics, governance, and security portfolios. They compete against cloud-native pure-plays such as Snowflake and Databricks, alongside vendors carrying decades of on-premises database heritage into cloud-based products. Asked what features matter most, respondents name Performance and Scalability first (54%), followed by Security and Compliance (39%), Data Integration (35%), Flexibility and Interoperability (32%), and Cost Efficiency (32%), a priority set that rewards platform vendors able to absorb more of the stack, and one where cost ranks behind four capability measures rather than ahead of them.

This Observatory features comprehensive and current data about the cloud data warehouse marketplace. The ETR Observatory data for Cloud Data Warehouses was specifically designed to capture usage and evaluation metrics across a wide swath of professionals representing the end user and evaluator buying demographic.

Databricks Overtakes Snowflake for Top Spending Momentum as the Leading Five Compress

Databricks leads the cloud data warehousing market in spending plans with a Net Score of 65%, followed by Snowflake (62%), Microsoft Azure Synapse Analytics (58%), Google BigQuery (54%), and Amazon Redshift (52%), a top tier separated by just 13 points. Net Score falls sharply after this group: SAP Datasphere (38%) and OpenText Core Analytics Database (29%) lead the second tier, while Teradata VantageCloud (-17%) and IBM Db2 Warehouse (-38%) are the only vendors in negative territory. Replacement intent stays low across the field, with Amazon Redshift's 4% the highest among the leaders, indicating that pressure on the laggards is coming through spend reduction rather than outright displacement. Among Global 2000 organizations the profile shifts: Snowflake posts the highest Net Score at 69% and the highest three-plus-year planned use at 73%, while Microsoft Azure Synapse Analytics falls to 41%.

Compared to last year's Observatory, OpenText Core Analytics Database saw the largest increase, climbing from 12% to 29%, a 17 percentage-point (ppt) jump, followed by Oracle Autonomous Data Warehouse at +9 ppts, though both come off low citation counts. The more consequential movement is at the top, where Snowflake surrendered the Net Score lead it took last year, falling 10 ppts from 72% to 62%, and Databricks (-2 ppts), Amazon Redshift (-4 ppts), and Google BigQuery (-2 ppts) also softened. Microsoft Azure Synapse Analytics is the only leader to improve, up 1 ppt. The sharpest decline in the study belongs to IBM Db2 Warehouse, down 42 ppts to -38%, where 46% of current users now plan to decrease spend over the next 12 months. Expected ROI compressed alongside spending: Databricks (70%), Google BigQuery (69%), and Snowflake (69%) lead on ROI expected within three years, but all five leaders sit below their 2025 marks of 73-76%.

Snowflake Sweeps the Write-Ins While Databricks Posts the Field's Only Positive NPS

A Net Promoter Score (NPS) for the vendors in this study simply rates which vendors respondents would recommend to their colleagues, following industry-standard measures of “promoters” (a 9 or 10 likelihood to recommend) minus “detractors” (a 6 or lower likelihood to recommend). Databricks is the only vendor in the study with a positive NPS, at +8. Snowflake (-1), Microsoft Azure Synapse Analytics (-3), Amazon Redshift (-10), ClickHouse Cloud (-16), and Google BigQuery (-16) follow, clustered near neutral, while Teradata VantageCloud posts the study's worst reading at -70.

Respondents were asked to provide write-in responses stating the vendors they saw as most desired among cloud data warehouses and most innovative. For most desired, this meant a vendor a respondent would choose to prioritize if given the opportunity to completely rebuild their organization's data stack. Snowflake leads most desired at 24%, narrowly ahead of Microsoft/Azure/Fabric (23%) and Databricks (20%), with Amazon/AWS (13%) and Google/GCP (10%) following. Among many attributes, what makes a product desired is a tool's completeness, or ability to do what is expected of it, as well as its ability to integrate with an existing technical ecosystem. Snowflake (83%), Databricks (80%), and Amazon Redshift (77%) lead on doing everything expected of a cloud data warehouse, while Microsoft Azure Synapse Analytics (88%), Snowflake (81%), and Amazon Redshift (79%) hold the top three spots for ease of. Microsoft's 88% is the single highest strength reading recorded for any vendor on any measure in this study.

In the write-in question for who respondents view as the most innovative cloud data warehouse, Snowflake again tops the list at 26%, followed by Databricks (23%), Microsoft/Azure/Fabric (16%), Amazon/AWS (14%), and Google/GCP (11%). Snowflake's sweep reverses the 2025 split, when Microsoft led most desired and Snowflake led most innovative. As a corollary to the innovation rankings, Databricks (83%) and Snowflake (83%) have the highest rates of agreement to the statement, “this product has an innovative technical roadmap,” followed by Amazon Redshift (77%) and Google BigQuery (73%); Microsoft Azure Synapse Analytics places fifth at 69% despite leading on ecosystem integration, the clearest split in the study between being embedded and being seen as forward-looking. The same two pure-plays also top agreement that updates are executed well (Snowflake 82%, Databricks 80%).

Conclusion – A Compressed Top Tier and a Widening Gap Behind It

The cloud data warehouse market has settled into a five-vendor top tier that is converging on every measure that matters. Databricks, Snowflake, Microsoft Azure Synapse Analytics, Google BigQuery, and Amazon Redshift sit within 13 points on Net Score, within four points on expected ROI, and within five points on talent availability. The hyperscalers continue to leverage bundled portfolios and ecosystem integration while the pure-plays hold the strategic high ground, taking a combined 44% of most-desired and 49% of most-innovative write-in votes. Behind that tier, the long-established and on-premises legacy vendors are diverging sharply: IBM Db2 Warehouse posts a -38% Net Score with 46% of users cutting spend, and Teradata VantageCloud sits at -17% with the study's worst NPS and its shortest planned use. Yet these tools remain hard to dislodge — IBM Db2 Warehouse records the field's second-highest “difficult to replace” agreement at 60%, and its three-plus-year commitment improved from 30% to 58% year-over-year. The gap is widening on spend, not on presence.

The open question heading into 2027 is who captures the AI workload, and this cycle's data suggests it will be settled on economics rather than capability. Respondents rank Performance and Scalability (54%) and Security and Compliance (39%) well ahead of Cost Efficiency (32%) when naming what matters, yet consumption pricing — which 52% call the most attractive model, more than double the 23% who prefer per-user licensing — converts every incremental AI-driven query into an incremental bill. That tension already shows in the data: expected ROI fell for all five leaders, and Snowflake carries a 23-point spread between agreement that it does everything expected of a warehouse (83%) and agreement that it is good value for the money (60%).

To dive deeper into the ETR Observatory's insights and uncover the full competitive landscape, use the form below to check out the complete data set.

About ETR

Enterprise Technology Research (ETR), a Futurum company, is an enterprise technology market research firm that delivers actionable, transparent, and unbiased insights to institutional investors and a trusted community of technology leaders, empowering them to make smarter, faster decisions. ETR’s proprietary approach is grounded in their vision to reinvent technology market research so that business leaders can strategically position their organizations to outperform the competition. In fact, no other firm harnesses the same scale and makeup of their vetted community to quickly deliver the unbiased data and analysis that financial organizations need to achieve better outcomes.

We use our core surveys to collect data and insights directly from the ETR Community. Use this data and insights to navigate the complex enterprise technology landscape and our proprietary visualizations and models to mine insights and unearth predictors of enterprise technology performance.

Beyond our core surveys, we also offer custom market research surveys. These can be commissioned with a targeted group and are guided by our expert content team to determine the best audience, topics, and questions. Additionally, the target group cannot only be based on their organization size, sector, and title, but also on our proprietary research around a firm’s spending intentions and technology stack.

Methodology
If you’re interested in the process behind the data, you’ll find it in the methodology section on our website.