SOFIAH NICHOLE SALIVIO News Editor
Crunchbase and Snowflake have published a report on venture capital flows in the Snowflake partner ecosystem. It found that more than USD $113 billion has gone to over 1,300 private companies in the network since 2020.
The report examines investment patterns tied to data infrastructure and software used by companies trying to move artificial intelligence projects from trials into broader use. Funding has clustered around businesses working in machine learning, data science, security, governance, observability, and analytics.
Startups in the Snowflake Partner Network raised USD $34 billion across more than 760 deals during the 2021 market boom. By 2025, that total had fallen to USD $25.5 billion and the number of deals had dropped to 260, suggesting larger sums were going to a narrower group of companies.
That shift points to a more selective venture market after the broad-based funding surge earlier in the decade. The analysis suggests investors are backing fewer companies with larger rounds as competition intensifies among startups seeking to supply AI-related data tools to enterprise customers.
"Capital is increasingly concentrated in a small number of companies raising billions, while across the broader ecosystem, investors are picking winners as competition to break through intensifies," said Gené Teare, Research Lead at Crunchbase. "AI is driving a fundamentally different investment cycle."
Funding focus
Machine Learning and Data Science accounted for 45% of all deal funding in the network since 2020, or USD $51 billion. Security, Governance and Observability, along with Business Intelligence and Analytics, each made up 24%, or USD $27 billion.
The report argues that demand for AI systems inside large organisations is shaping where capital is flowing. Investors appear to favour companies that address underlying issues of trusted data, control, and visibility rather than backing only model developers.
About 130 private unicorns valued at USD $1 billion or more are active in the ecosystem. Funding stages were relatively balanced, with 35% of companies at seed stage, 37% at early stage, and 29% at late stage.
Crunchbase also used its predictive models to assess the likelihood of further financing and exits among startups in the Snowflake ecosystem. It found that 44% of companies that have already raised at least USD $5 million are very likely or probable to raise another round, while 34% are very likely or probable to reach an acquisition or initial public offering.
Snowflake ventures
The analysis also highlighted Snowflake Ventures and the Snowflake Startup Accelerator, which back startups building products around Snowflake's platform. Annual funding raised by Snowflake Ventures portfolio companies has increased each year since 2023, reaching nearly USD $3 billion in 2025, based on Crunchbase data.
That rise is another sign that investors continue to commit substantial sums to companies linked to AI data management and enterprise software, even as the wider venture market has become tougher and more selective.
Harsha Kapre, Head of Snowflake Ventures, said the current market is being shaped by companies that combine technical development with practical business use. His comments linked the report's findings to the operational demands facing enterprises that want to deploy AI systems at scale.
"As enterprises move from AI experimentation to production, success depends on more than models alone. It requires trusted data, context and holistic governance," Kapre said. "Across the Snowflake partner ecosystem, we're seeing strong momentum behind startups that pair technical innovation with clear customer value, helping organizations scale AI on secure, governed data foundations."
The figures add to a broader picture of AI investment in which infrastructure and data management businesses are attracting sustained investor interest. Rather than funding a wide field of smaller entrants, venture firms are increasingly concentrating money on startups they believe can become core suppliers to large companies adopting AI systems.
For startups, the report suggests access to enterprise customers may depend less on AI branding alone and more on solving long-standing data problems around governance, security, and analytics. For investors, it offers further evidence that the AI market is rewarding a narrower set of companies with much larger cheques.