Despite their promises of innovation, AI startups face a familiar and fundamental question: how do they know when they have achieved the holy grail of product-market fit? This concept has been studied for years, with entire books written on how to master it. However, artificial intelligence is now upending these established practices.
According to Ann Bordetsky, a partner at New Enterprise Associates, the approach for AI companies could not be more different from traditional tech playbooks. She described it as a completely different ball game, noting that the breakneck pace of change in the AI world is a primary factor because the technology itself is not static.
Even with these challenges, there are methods for founders to evaluate their progress. Murali Joshi, a partner at Iconiq, advises monitoring the durability of spend. Since AI is still in early adoption at many companies, spending is often focused on experimentation. A key signal of product-market fit is a shift from experimental AI budgets to core company budgets, indicating a solution is seen as essential rather than just a test.
Joshi also emphasized classic metrics like daily, weekly, and monthly active users to gauge how frequently customers engage with a product they are paying for.
Bordetsky agreed with the importance of these metrics but added that qualitative data from customer interviews provides crucial nuance. Speaking directly to users can reveal whether they are likely to stick with a product, a insight that numbers alone might not confirm.
Interviewing executives is another valuable tactic. Joshi suggests asking where a product sits in the company’s tech stack and focusing on how to make the product sticky by integrating it into core workflows.
Finally, Bordetsky highlighted that product-market fit should be viewed as a continuum, not a single point in time. It begins with a small amount of fit in a specific area and must be strengthened and expanded over time.