On hiring too early

We raised $5M and immediately started hiring. That's what you do, right? Raise money, build a team. The investors expect it. The roadmap demands it. More people means you go faster.

Looking back, that instinct cost us more than almost any other mistake we made. But not for the reasons you'd think.

The conventional warnings about hiring too early focus on burn rate and communication overhead. Those are real, but they're not the main problem. The main problem is that hiring insulates you from learning. And pre-product-market fit, learning is literally the only thing that matters.

Before PMF, a startup is a learning machine. Every conversation with a customer, every failed experiment, every feature that doesn't land the way you expected – that's all data. The founders' job is to absorb that data as directly as possible and convert it into better decisions. Revenue doesn't matter yet. Org structure doesn't matter yet. The only question is: are you learning fast enough to find the thing that works? And that learning isn't just about product – it's about who you're selling to, how you talk about it, and whether the whole package fits together.

Hiring disrupts that process in ways that feel like progress but aren't.

When you hire someone to do a job you haven't done yourself, you're outsourcing your learning. You're putting someone between you and the reality of the problem. They'll do the work, give you updates, and present results. But the raw, messy understanding that comes from doing it with your own hands – the kind that changes how you think about the whole business – that stays with them. Or worse, it doesn't develop at all, because they're executing on your assumptions rather than pressure-testing them.

I see this pattern often. A technical founder raises money and immediately hires a head of sales because "I'm not a sales person." You might not be, but if you haven't sold your own product to at least a handful of customers, you don't know what the sales motion looks like. You don't know which objections are real. You don't know how people describe the problem in their own words. You're hiring someone to figure all of that out, and you won't have the context to evaluate whether they're getting it right.

Worse, a good salesperson will sell. They'll bring in revenue. And revenue is the most convincing form of false progress there is. You'll look at the numbers going up and feel like you're on your way – when what's actually happening is you're paying someone to push a product that hasn't found its fit yet. The revenue masks the learning you still haven't done. It lets you avoid the hard, uncomfortable work of figuring out whether you're building the right thing for the right people.

The same applies to marketing, customer success, product management – any function where understanding the problem deeply matters more than raw throughput. And pre-PMF, that's every function.

Every person you hire adds distance between you and the problem. The feedback loop gets longer. The signal gets filtered through someone else's interpretation. You start managing people instead of talking to customers. The org chart looks real, things are happening – but you can't tell whether the right things are happening if you've never done the work yourself.

The strongest teams I've seen were built by founders who hired late – who pushed through the discomfort of doing unfamiliar work themselves until they'd learned what they needed to learn, and only then brought in someone to take it further. The delay is the investment. What you learn in that window shapes every hiring decision, every management conversation, and every strategic call that follows.

Early on, hiring and learning are often in tension. And most of the time, learning should win.