In this interview, Christopher Pedregal, founder of Granola, reflects on the company’s early growth trajectory and the challenges of scaling a startup with a very small team. He explains how the initial launch involved only four to six people, yet AI‑assisted coding allowed them to achieve more than traditional small teams could. Despite an organic surge in users (about 500 installs on launch day) that eventually revealed product‑market fit, the team underestimated hiring needs and were six months late in expanding staff, especially in London. Pedregal discusses the tension between rapid product decisions—such as expanding to Windows, iOS, and defining a business model—and the need for thoughtful, parallel decision‑making. He emphasizes that AI currently serves as an augmentation tool that speeds up execution, but strategic choices still require human insight. He concludes that even with today’s AI advances, he would not change his past decisions, underscoring the importance of timely hiring and deliberate strategic planning in early‑stage growth.
Chapters
are we scaling quickly or not scaling quickly, it's interesting, right? There's so many dimensions to that. On one hand, we were 4 people when Granola launched. Granola's like— Granola launched a year and a half ago, right? We were 4 people. It's really— and then we were 4 or 5, maybe 6 people for way too long after that. And on one hand, I think that's testament to what you can do with small teams today, what you can do with AI coding. So it's like a small group of people can do a whole lot more than they could before, right? So that's great. When we started hiring, like, I made the mistake to not hire fast enough, right? Like, I come from a world where it's like you have a few really, really talented, really aligned people can accomplish great things. And like, that is true, right? But to keep up with the demand and scale that we had to, like, I completely underestimated that, right? I was like 6 months too late. —6 months. —6 months.
—For launchers, right?
That's very, very long time. What happened is we launched Granola. And on launch day, I think we had 500 people install the app, which is exactly what I expected, right? Like most— again, most of the time you build a startup, no one cares. Like you launch something, no one cares. So we didn't realize we had product market fit. And what happened is it just started growing organically, little by little by little. And then all of a sudden, when we looked around, it was too late. There were already too many people trying to pay for it, too many people with bugs, too many people— and at that point, it was like, ah, crap. And then hiring in London was— I mean, it's like a mixed bag, right? But it's taken us a little while to find our muscle there. I think we're there now, but it did take us a—
And that point was about a year ago? Yeah. Roughly, were you at that point? Yeah. What you've seen in terms of progress of AI since the last 12 months, would that be different today or would you still hit the same point and still say, we still need to scale faster?
Still hit it. Yeah, yeah, yeah. Because I guess like, I think we can do crystal ball, like where AI is going to go, et cetera, right? But like right now, what it does is just, if you're using AI well, it just augments a person's ability so you can do more faster. That's how I view it. There's still a lot of critical decisions that need to be made. And for us, it's like, oh, OK, well, do we expand to Windows? Do we have to build this on iOS? What's our business model? How do we charge for this? There are all these questions that need really thoughtful people. And I guess I thought we could kind of stagger those. And actually, we had to do all of those in parallel. So I think I would go— if I can go back, or using today's technology back then, Wouldn't change my decision at all.
