Simile hits $2B valuation with $200M Series B

Simile, the synthetic-user startup founded by Stanford PhD graduate Joon Sung Park, closed a $200 million Series B at a $2 billion valuation on July 30. The round came just five months after the company emerged from stealth with a $100 million Series A led by Index Ventures.

Greenoaks led the Series B, with participation from Index, Hanabi, Bain Capital Ventures, A*, Factory, Definition, and CVS Health Ventures. CVS is also a Simile customer, making this a strategic bet as well as a financial one.

What Simile actually does

Simile creates simulated users for marketing and product research. Instead of recruiting panels or running focus groups, companies can test messaging, product concepts, and user flows against AI-generated populations.

Park's academic work foreshadowed this. His dissertation project, called "Smallville," built AI agents that carried on simulated human lives, complete with social interactions and parties. The research explored how LLM-powered agents could model realistic human behavior at scale.

The company's stated mission is to simulate "all eight billion people on earth, accurately and honestly." That claim invites skepticism. The entire premise of market research exists because humans are unpredictable, driven by shifting emotions, context, and reasoning that defies easy modeling.

The competitive landscape

Simile is not alone in chasing synthetic-user research. Aaru, another startup in the space, raised a Series A in December 2025 at a reported $1 billion valuation. The category is attracting serious capital because the potential efficiency gains are obvious: faster iteration cycles, lower research costs, and the ability to test concepts at scales that would be impossible with human participants.

The risk is equally obvious. Simulated users can only reflect patterns present in their training data. They cannot surface genuinely novel preferences or behaviors. For certain research questions, especially those probing edge cases or cultural shifts, synthetic respondents may produce confident but misleading signals.

Why CVS matters

CVS Health Ventures appearing on the cap table signals enterprise traction in healthcare, a sector with expensive, compliance-heavy research requirements. If Simile can help CVS test pharmacy app flows or insurance messaging without recruiting actual patients, the time savings alone justify experimentation.

Healthcare and financial services are natural beachheads for synthetic-user tools. Both industries face regulatory constraints on how they can use real consumer data, and both spend heavily on research. A synthetic approach could sidestep some compliance burdens while accelerating product cycles.

What this signals for AI startups

A $2 billion valuation five months after a Series A is aggressive even by AI standards. It suggests either remarkable revenue growth that Simile has not disclosed, or a venture market willing to pay up for category leadership before fundamentals are proven.

Park's academic pedigree and the "Smallville" research provide technical credibility. But credibility is not defensibility. The core technology relies on LLMs that competitors can also access. Simile's moat, if one exists, will come from proprietary data, customer integrations, or a feedback loop that improves simulation quality faster than rivals.

For founders watching this space, the question is whether synthetic-user tools become a standalone category or a feature that HubSpot, Salesforce, or the major survey platforms absorb. If Simile can establish itself as the infrastructure layer before incumbents catch up, the valuation may prove justified. If not, $2 billion is a lot of expectation to carry.

Source: Startups | TechCrunch / Julie Bort