Ben Thompson · Mar 27, 2025
Ben Thompson interviews Snowflake CEO Sridhar Ramaswamy about data and AI
Markdown · JSON · Original source
In an interview published March 27, 2025, Ben Thompson asks Snowflake CEO Sridhar Ramaswamy how the company can grow from a cloud data warehouse into a broader platform for AI. Ramaswamy argues that AI can make more business data usable and let more people query it, while Snowflake’s task is to make that use “Easy, efficient, and trusted.” He says the company is building on its data platform rather than trying to become a company. His account links that strategy to open formats, connectors and products that let customers work with structured and unstructured data. He also argues that AI could change how people use business data: instead of relying only on dashboards or specialist analysts, more users could ask questions and follow up with analysis. He stresses that AI products need ways to avoid answering questions they cannot answer reliably. Ramaswamy describes a further opportunity in combining data and software into workflows, while saying that enterprise migrations remain difficult and can take . The interview’s central tension is whether Snowflake can broaden its offering while keeping the integrated, easy-to-use product it sees as a strength. Ramaswamy’s ambitions for growth are forecasts, not outcomes; the interview does not establish how quickly these changes will deliver value or how reliably AI will handle customer work.
Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.
Original source
Ben Thompson · Article
Source details
Original source date: Mar 27, 2025 (2025-03-27T10:18:40+00:00).
Date basis: Date stated on the publisher’s page.
Added to Escape Meridian: 2026-11-14T18:53:27.323Z. First observed: 2026-11-10T05:36:01.943Z. Retrieved: 2026-11-10T05:39:35.198Z.
Referenced by
Recorded citations and public file attachments. A reference alone does not establish agreement or influence.
No public references recorded here yet.
Limits
Text as retrieved on November 10, 2026; originally published March 27, 2025. This is an interview and the account of Snowflake’s strategy and prospects is Ramaswamy’s; the interview does not independently establish future results. The interview does not establish how reliably AI products will handle customer work in practice.
Open question
What evidence will show whether Snowflake can broaden its AI and data products while maintaining the reliability and integrated experience Ramaswamy describes?
Source-stated
Ben Thompson frames the interview around whether Snowflake can expand beyond its structured data warehouse roots as competitors apply AI to unstructured data.
can Snowflake extend beyond its structured data warehouse roots before competitors like Databricks leverage AI to wrangle unstructured data into a more compelling offering?
Source-stated
Ramaswamy argues that AI could let more people get value from Snowflake data by making it easier to query complex datasets and conduct follow-up analysis.
I think it vastly expands the set of people that can get value from Snowflake data
Source-stated
Ramaswamy says Snowflake’s AI products should support trusted answers, including by avoiding questions the system should not answer.
we call it a verified query repository of positive and negative examples for Cortex Analyst
Source-stated
Ramaswamy describes combining data elements into more complex workflows as a direction for Snowflake’s products.
Where we are headed is now in how can you compose these different data elements, these agentic operators as they are, to build more complex workflows
Source-stated
Ramaswamy says Snowflake’s aspiration is to grow faster than its stated recent rate for roughly the next decade; this is his goal, not a reported outcome.
if we fulfill our mission to truly help enterprises mobilize data, we should be growing at faster than that clip for the next 10-ish years.
Source and contributor context
Ben Thompson · Publisher
The publisher is identified here. No individual author has been inferred.
Source media
Media retrieval status: unchecked.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Foundation model
A large model trained on broad data that can be adapted to many different tasks.
TIME
TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Editorial provenance
Version 1. Publication ID: 3d4935f4-1802-38a7-5361-7f36c71e52bd.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-14T18:49:17.355Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: 0e509c179744c16558294a55cc35ed9f54f3f1bf7e4358171318d5f6db2bb74f. Source text hash: 7590fcf4b09e45bd1355460b27407f3b009489456c4a31300eeac2482c23752a.
Revision and source-check history
Version 1 · 2026-11-14T18:53:27.323Z
Ben Thompson interviews Snowflake CEO Sridhar Ramaswamy about data and AI
In an interview published March 27, 2025, Ben Thompson asks Snowflake CEO Sridhar Ramaswamy how the company can grow from a cloud data warehouse into a broader platform for AI. Ramaswamy argues that AI can make more business data usable and let more people query it, while Snowflake’s task is to make that use “Easy, efficient, and trusted.” He says the company is building on its data platform rather than trying to become a company. His account links that strategy to open formats, connectors and products that let customers work with structured and unstructured data. He also argues that AI could change how people use business data: instead of relying only on dashboards or specialist analysts, more users could ask questions and follow up with analysis. He stresses that AI products need ways to avoid answering questions they cannot answer reliably. Ramaswamy describes a further opportunity in combining data and software into workflows, while saying that enterprise migrations remain difficult and can take . The interview’s central tension is whether Snowflake can broaden its offering while keeping the integrated, easy-to-use product it sees as a strength. Ramaswamy’s ambitions for growth are forecasts, not outcomes; the interview does not establish how quickly these changes will deliver value or how reliably AI will handle customer work.
Original source
Ben Thompson · Article
Source details
Original source date: Mar 27, 2025 (2025-03-27T10:18:40+00:00).
Date basis: Date stated on the publisher’s page.
Added to Escape Meridian: 2026-11-14T18:53:27.323Z. First observed: 2026-11-10T05:36:01.943Z. Retrieved: 2026-11-10T05:39:35.198Z.
Limits
Text as retrieved on November 10, 2026; originally published March 27, 2025. This is an interview and the account of Snowflake’s strategy and prospects is Ramaswamy’s; the interview does not independently establish future results. The interview does not establish how reliably AI products will handle customer work in practice.
Open question
What evidence will show whether Snowflake can broaden its AI and data products while maintaining the reliability and integrated experience Ramaswamy describes?
Source-stated
Ben Thompson frames the interview around whether Snowflake can expand beyond its structured data warehouse roots as competitors apply AI to unstructured data.
can Snowflake extend beyond its structured data warehouse roots before competitors like Databricks leverage AI to wrangle unstructured data into a more compelling offering?
Source-stated
Ramaswamy argues that AI could let more people get value from Snowflake data by making it easier to query complex datasets and conduct follow-up analysis.
I think it vastly expands the set of people that can get value from Snowflake data
Source-stated
Ramaswamy says Snowflake’s AI products should support trusted answers, including by avoiding questions the system should not answer.
we call it a verified query repository of positive and negative examples for Cortex Analyst
Source-stated
Ramaswamy describes combining data elements into more complex workflows as a direction for Snowflake’s products.
Where we are headed is now in how can you compose these different data elements, these agentic operators as they are, to build more complex workflows
Source-stated
Ramaswamy says Snowflake’s aspiration is to grow faster than its stated recent rate for roughly the next decade; this is his goal, not a reported outcome.
if we fulfill our mission to truly help enterprises mobilize data, we should be growing at faster than that clip for the next 10-ish years.
Source and contributor context
Ben Thompson · Publisher
The publisher is identified here. No individual author has been inferred.
Source media
Media retrieval status: unchecked.
Terms and references
Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.
Foundation model
A large model trained on broad data that can be adapted to many different tasks.
TIME
TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Editorial provenance
Version 1. Publication ID: 3d4935f4-1802-38a7-5361-7f36c71e52bd.
Prepared by openai/gpt-6-luna; grounding review by anthropic/claude-opus-5.5.
Source check: uncertain by anthropic/claude-opus-5.5 on 2026-11-14T18:49:17.355Z. This checks fidelity to the source, not whether its claims are independently true.
Editorial selection: accept; reasons: bridge_admitted_by_model, second_opinion_categorical.
Observation: 0e509c179744c16558294a55cc35ed9f54f3f1bf7e4358171318d5f6db2bb74f. Source text hash: 7590fcf4b09e45bd1355460b27407f3b009489456c4a31300eeac2482c23752a.
uncertain · 2026-11-14T18:49:17.355Z · anthropic/claude-opus-5.5
Publication: 3d4935f4-1802-38a7-5361-7f36c71e52bd
attribution — pass: The reading correctly identifies this as a single Stratechery interview by Ben Thompson with Snowflake CEO Sridhar Ramaswamy. It keeps Thompson's framing of the competitive question with Thompson and the strategy claims with Ramaswamy.
This week’s Stratechery Interview is with Snowflake CEO Sridhar Ramaswamy.
summary — uncertain: The summary is faithful and keeps Ramaswamy's voice. Its central tension depends on terms left unexplained for a general reader, though: "structured data warehouse", "structured and unstructured data", "open formats" and "connectors". Only "foundation model" is defined beside the reading. Smallest repair: add term-library tooltips for structured/unstructured data and open formats, or a short plain clause such as tidy database tables versus documents like PDFs. The source illustrates the distinction with extracting structured data from a PDF.
your ability to analyze a PDF document let’s say, and extract structured data from it just got a whole lot easier because of things like multimodal models
date — pass: March 27, 2025 matches the UTC calendar day of evidence.originalPublished and the source dateline. This is a publisher page, not an archive copy. The growth aspiration is framed as a forecast. The trailing related-post titles are not used, and there is no hindsight framing.
Thursday, March 27, 2025
support — pass: Each claim is grounded in Ramaswamy's or Thompson's words and stays attributed. This covers: expanding who can use the data; the "Easy, efficient, and trusted" watchword; the verified query repository with negative examples; composing agentic operators into workflows; difficult migrations; the integrated-product bet; and growth faster than 30% for about 10 years as an aspiration. The stated unknowns are appropriately limited, and the reading adds no unsupported benefits or costs.
our watchword for AI on Snowflake is “Easy, efficient, and trusted”
media — pass: No media was selected and no verified media candidates were supplied, so there is no media identity or relevance issue.
Challenge or correct this reading
Send a correction or an opposing source with this reading’s URL. Disagreement and uncertainty can remain unresolved.
A large model trained on broad data that can be adapted to many different tasks.
Definition referenceTIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Explore TIMEA large model trained on broad data that can be adapted to many different tasks.
Definition referenceTIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.
Explore TIME