Three years at a startup that broke me — and what I learned from it

I want to be careful not to make this sound like a cautionary fable — a tidy arc with a clear villain and a lesson neatly extracted. What happened over those three years was more complicated and, in some ways, more ordinary than that. I made choices. The company made choices. The mythology did its job. And I came out the other end needing considerably longer to recover than I'd been told to expect.

Here's what actually happened.

What the pitch sounds like — and why it works

The CEO was charismatic in the way that startup founders tend to be: energetic, convinced, in possession of a story about the future that made the work feel meaningful before you'd shipped a single line. We were solving a real problem for real people. The hours would be intense, but it was early days — the team was small, the stakes were high, and once we hit the next milestone, things would normalise.

They did not normalise. I should tell you that upfront, because the normalisation promise is structural to the startup pitch and it is almost always false. The next milestone is perpetually the one after which things get easier. Series A will calm things down. Then it's the Series B. Then it's the pivot. There is always a reason why now is the exception and normalcy is just ahead. Normalcy is not ahead. It was never the plan. The business model, in most cases, requires the intensity to continue — and no amount of milestone achievement changes the underlying model.

I knew this, in the abstract. Most people do. The mythology works anyway, because knowing something abstractly and feeling it in the day-to-day are different things — and in year one, the day-to-day felt genuinely good.

Year one: when the mythology was also partly true

I don't want to pretend the early phase was all exploitation. It wasn't. There was genuine energy — smart people working on a hard problem, fast decisions, tight feedback loops, a clear line between what I built and what changed for users. I had more ownership than I'd had in any previous role and it produced real engagement, the kind that doesn't feel like performance because it isn't.

I was working twelve-hour days and I mostly didn't mind. Not because I was naive about the trade-off — I understood the deal — but because in year one, the fuel was real. I was interested. I was learning. The company's problems felt like mine in a way that produced motivation rather than resentment.

This is the part of the startup experience that the mythology isn't lying about. The early phase can feel genuinely meaningful in a way that slower, more structured environments often can't match. What the mythology doesn't tell you is the price attached to that feeling, and how far that price extends into the years that follow.

The transition nobody names

At some point — and I can't tell you precisely when, because it happened gradually and then suddenly — the relationship changed.

The same hours that had felt like investment started feeling like extraction. The same sense of ownership that had made problems motivating now made failures feel personal in a way I couldn't recover from quickly. The company had started needing things from me that I didn't have to give: not just hours, but enthusiasm I no longer felt, certainty about the product that I'd stopped having, performance of confidence in a roadmap that I privately knew had fundamental problems.

I kept waiting for the company to recognise that the dynamic had changed. I don't think it did, or particularly cared to. The expectation was functionally identical in year three to what it had been in year one. I was the one who had changed. My reserves had been spent, slowly, without adequate replenishment, and the company's model had no mechanism for noticing that or accounting for it. The output was still expected. The output was still produced, at first. The cost was accumulating invisibly.

The mechanisms that kept me there longer than I should have stayed

I want to be specific about this, because "startup culture" is abstract and I want to describe the actual things that were happening.

The equity conversation. Every time leaving became a serious thought, someone — a founder, a senior leader, occasionally a peer — would raise the equity. Not quite as a threat, not quite as a promise. More like a gentle reminder that I had vested a certain percentage, that the company was tracking toward something, that leaving now would mean giving up on what I'd helped build. I stayed approximately eight months longer than I should have, partly because of this. The equity was eventually worth nothing. The eight months cost considerably more than nothing.

The team obligation. Startups are small. The people you work with are people you know. When I was struggling, the thing keeping me there wasn't loyalty to the company — it was loyalty to specific colleagues who were depending on specific things I was building. This is not an accident of scale. It is the most effective form of social pressure available, and it is frequently used, consciously or not, to extract effort that equity and salary alone couldn't justify. Turning your limits into a collective obligation is the move, and it works.

The identity merger. At some point I stopped describing what I did as my job and started describing it as what I was building. The company's story had become my story. Its progress felt like mine, which meant its difficulties felt like my failures. When the product wasn't working, I was not able to hold an appropriate distance from that. Untangling the company's situation from my own worth — separating what was happening to the business from what that said about me — took longer after I left than the job itself had lasted. This is probably the most underestimated cost of the startup experience and the one that receives the least attention in post-mortems.

What broke, specifically

I'll be direct about this, because I think the general language of burnout can obscure the specific thing that actually happened.

I was leading a team of five. For approximately eighteen months I had been absorbing escalating pressure from the executive layer on one side and protecting my team from it on the other — a role that startup managers will recognise and that nobody puts in a job description. I was sleeping five hours a night, routinely, and telling myself it was temporary. The product was not performing as the narrative required, and I had been presenting optimism in all-hands meetings that I did not privately feel for the better part of a year.

The thing that made me finally understand something had broken was a planning meeting I walked into with full notes, having prepared properly, and found that I could not produce the thinking. Not in a dramatic sense — I could speak, I could reference my preparation. But the genuine engagement required to actually think through the problem in the room was absent. I was performing deliberation rather than doing it. I was simulating a functional version of myself. The tank was empty in a way I hadn't been able to fully name until that moment made it undeniable.

I took two weeks off. The company allowed it with visible reluctance — the timing wasn't great, which it never is. I came back approximately unchanged. I worked for another three months in a state I'd describe as going through the motions, adequately. Then I left.

What recovery from this actually looks like

It took me the better part of eighteen months after leaving to feel like myself again. Not because leaving was hard — it was one of the more straightforward decisions I've made — but because the depletion was deep and the identity merger had gone further than I'd recognised. Unwinding the story of what the company's outcome said about me took considerable time and, eventually, a therapist who actually understood what sustained high-pressure work does to a person.

A few things I'd tell someone earlier in this than I was:

The equity maths probably don't work. Do them honestly, including the parts that don't flatter the narrative. Do them before you make any decision about staying or going based on what might vest. I did them properly only after I left, which is when they stopped mattering.

The mission being real doesn't protect you. I was building something I genuinely believed in. That belief did not prevent the depletion. It probably extended my tolerance for conditions I should have left sooner. The value of the work and the sustainability of the environment are separate variables. Don't let one obscure the other.

Recovery from startup burnout is slow — specifically the part that involves separating your sense of worth from the company's outcomes. The fatigue recovers faster than the identity piece. Build in time for both.

None of this is an argument against startups categorically. Some people thrive in them for years. The difference between people who thrive and people who get broken is almost never about resilience or personality — it's usually about whether the actual contract (not the pitch, the actual conditions) was something the person chose knowingly and can genuinely sustain. If you go in with clear eyes about what you're exchanging and for what, the bet can be sound. The damage happens when the mythology replaces the honest evaluation.

The burnout track on the Start Here page has more on what recovery from this kind of sustained depletion tends to look like — including the timeline, which is almost always longer than you're told to expect, and the parts that are easy to miss.