Granola went from 5,000 weekly users to a $1.5B unicorn — without a meeting bot

Eighteen months ago, Granola had 5,000 weekly users and a fairly simple pitch: a meeting notes app that didn’t embarrass you in front of clients. Now, it has $192 million in total funding, a $1.5 billion valuation, and ambitions that have little to do with note-taking.

The London-based startup announced a $125 million Series C on 25 March, led by Danny Rimer at Index Ventures, with Mamoon Hamid at Kleiner Perkins also participating. Lightspeed Venture Partners, Spark Capital, and NFDG — the venture firm run by Nat Friedman and Daniel Gross — returned for the round. Rimer will join the board as an observer.

The valuation is a sixfold jump from the $250 million figure Granola carried just ten months ago after its Series B. By most measures, that’s a striking leap — even in a market that has grown comfortable with AI valuations that defy conventional logic.

The insight that started it all

Granola was founded in 2023 by Chris Pedregal and Sam Stephenson around a deceptively simple observation: professionals hate meeting bots. Not the transcription, not the summary — the visible bot. The little participant that joins your call, sits there recording, and makes your client shift uncomfortably in their seat.

Granola’s solution was to move the recording off the call entirely. The app sits on a user’s computer and captures meeting audio locally, without connecting to the meeting platform as a participant. It transcribes, generates structured notes, and makes everything searchable across a team — all without anyone in the room knowing it’s running.

That single product decision — local recording over cloud bot — turned out to be the wedge. Lawyers, salespeople, and executives adopted it precisely because it was invisible. By October 2024, when Granola raised its $20 million Series A, it had 5,000 weekly users. By May 2025, when it closed a $43 million Series B from NFDG at a $250 million valuation, the number had grown substantially. The company has not disclosed its current user count.

Beyond note-taking: the context layer play

The product Granola is selling today looks quite different from the one that attracted its early users. The company has built Granola Chat, which lets users query their full conversation history using Claude, GPT, or Gemini. It has launched Spaces, a collaborative layer for organising and searching meeting context across teams. And earlier this year, it released a Model Context Protocol (MCP) server alongside two new APIs — one for individuals, one enterprise-grade — that let external AI systems pull in meeting context directly.

That last piece is where the company’s long-term bet lives. Pedregal has argued publicly that AI meeting notes are becoming a commodity. The real moat, in his view, is not capturing the conversation — it’s making what was said accessible to other AI systems. If an agent can query every meeting a team has ever had before deciding what to do next, the reasoning goes, the value of that context layer compounds over time in a way that individual note summaries never could.

The $125 million is earmarked specifically for enterprise expansion — and the customer list already reflects that shift. Vanta, Gusto, Thumbtack, Asana, Cursor, Lovable, Decagon, and Mistral AI are among its clients. The enterprise API ships with SSO, SCIM, and consent-based data management — the infrastructure requirements that most large organisations demand before they’ll allow any tool to record employee conversations at scale.

A crowded room

The meeting intelligence market is not short of competition. Fireflies.ai has surpassed 16 million users and crossed a $1 billion valuation. Otter.ai has been in the space since 2016 and carries substantial brand recognition in the category. Glean and Mem.ai are coming at the enterprise knowledge problem from a different angle. And Microsoft, Google, and Notion are all building AI meeting features into products that already have tens of millions of users.

Granola‘s answer to this field is positioning — not as a meeting notes app competing on feature parity, but as a context infrastructure layer that other AI tools plug into. That’s a legitimate strategy, but it also raises a specific risk: Microsoft Copilot and Google Gemini are building the same context aggregation capability into environments where most enterprise employees already live. Whether Granola’s head start on agentic integration holds once the incumbents arrive with native distribution is the question the next funding cycle will probably answer.

The valuation question no one is quite answering

A $1.5 billion valuation for a company that was worth $250 million ten months ago is the kind of number that invites a follow-up. Granola has not disclosed revenue, growth rates, or retention data publicly. The AI meeting assistant market is projected to expand from roughly $3.5 billion in 2025 to over $34 billion by 2035, according to Market Research Future — which gives the category room to support several large companies simultaneously. But the gap between market projection and demonstrated revenue is still wide across this sector.

What Index Ventures and Kleiner Perkins are betting on, presumably, is the product insight that has held so far: that local recording without a visible bot unlocks a segment of professionals — in law, sales, finance, executive functions — who won’t use any alternative. That niche proved real enough to scale from 5,000 to a unicorn in under three years. Whether it’s large enough to justify the current price tag depends on how fast the enterprise pipeline converts — and how long the context-layer strategy stays defensible.

For founders watching from outside the AI meeting space, the more instructive story may be the simpler one: a two-person team in London spotted a specific friction point that every competitor had accepted as inevitable, built around it, and compounded their way to a $1.5 billion valuation before most people had heard of them.