The burnout that pushed me to go freelance — and why I haven't looked back

I want to be honest about why I went freelance, because the honest version is less flattering than the one I told people at the time.

I told people I was going freelance because I wanted autonomy, flexibility, to take control of my career. All of that was true. But the truer thing — the thing underneath — was that I had nothing left for the job and no idea how to fix that without leaving, and freelancing was the exit that felt least like admitting defeat. I could frame it as a move rather than a retreat. That distinction mattered more to me than it should have.

What had actually happened over the previous eighteen months is that I had burned out slowly and then comprehensively inside a role that looked fine from the outside — a well-compensated backend engineering job at a scale-up with a strong product and a reasonable team. There was nothing dramatically wrong with the company. The problem was subtler and more personal: I had stopped caring about the work at a level I couldn't recover from through willpower or better habits, and continuing to show up as if I hadn't was costing me more each month than I was prepared to admit.

What the burnout actually looked like

By the time I started seriously thinking about freelancing, I was spending roughly three hours of every working day performing the appearance of engagement without producing much of actual value. I don't mean I was slacking — I was showing up to every meeting, reviewing every pull request, delivering on every commitment. But the thinking behind the work had gone thin. The solutions I was reaching for were competent and conservative and entirely lacking the kind of original thinking I used to be capable of and that my senior role nominally required.

Nobody had noticed yet. That was almost the most frightening part. The slow decline in the quality of my thinking was invisible to the people around me because I was still delivering, still hitting the bar, still present enough that the absence underneath the performance wasn't visible. But I could feel it. I knew exactly what my best thinking felt like, and I knew I hadn't produced any of it in months.

What I needed, eventually, was to stop. Not a holiday — I'd had two in that period and come back from both exactly as depleted as I'd left. I needed to change the fundamental conditions. And freelancing, in my head, was the version of changing the conditions that preserved the most optionality.

The fear that nearly stopped me

The fear wasn't primarily financial, though I dressed it in financial language to make it feel more rational. I had eight months of expenses saved. My skills were marketable. I knew people at companies who had mentioned wanting to work with me. The financial case for trying was reasonable.

The fear was simpler and more personal: what if I went freelance and discovered that the depletion hadn't been caused by the job? What if the problem was me? Leaving the job would make that undeniable in a way that staying never quite would. As long as I was still employed, there was a plausible structural explanation for everything that was wrong. Going freelance meant finding out.

This is not a unique fear. I've heard versions of it from almost everyone who has made a significant career move coming out of burnout. The burnout has often eroded confidence in very specific ways — not the confident performance of competence that remains intact longest, but the deeper private belief that you're actually capable of the things you've been doing. Leaving the structure that provided external proof of your capability is frightening when the internal certainty has already been depleted.

The first six months: what actually happened

I gave three months' notice — longer than required, because I felt I owed it and because the longer runway meant I could have proper conversations about potential work without it feeling like poaching. By the time my last day arrived I had one confirmed engagement, two conversations at an advanced stage, and one vague warm lead. That's a better starting position than most people who make this move, and it was the result of spending months five and six of my notice period doing the business development work that felt uncomfortable and that I'd been avoiding.

The first thing that happened when I went freelance was not relief. It was a kind of vertiginous emptiness — the sudden absence of the structure that had been organising my days, even when that structure was the thing I was depleted by. I'd expected to feel free. Instead I felt, for about three weeks, like I'd jumped off something and wasn't sure yet what I'd land on. The burnout recovery and the transition into freelancing happened simultaneously, which is not ideal. In retrospect I would have taken six weeks off between the two, but the financial logic of going straight from employment to a confirmed engagement was too clear to ignore.

The first confirmed engagement was eight days of work over five weeks for a startup I had a relationship with. I billed at a rate that I thought was reasonable and that turned out to be substantially below market — a classic first-engagement mistake that I made despite having read that I would make it. Knowing you're going to underprice yourself and not doing it anyway requires a confidence in your own market value that burnout has specifically damaged. That's worth factoring into your expectations.

What freelancing fixed — and what it didn't

What it fixed was the specific form of depletion that came from performing engagement I didn't feel, for a company I'd stopped being interested in, inside an organisational context I couldn't change. That particular exhaustion — the cost of the daily performance — stopped almost immediately, and I had not understood how much of my energy it was consuming until it was gone.

What it didn't fix: the deeper fatigue took considerably longer. The sharpness I'd lost over eighteen months of grinding didn't come back in the first three months of freedom. I noticed it returning gradually — a moment of genuine interest in a problem, an insight that felt real rather than performed — but the recovery of the thing that burnout takes last (the genuine intellectual engagement, the enthusiasm that isn't manufactured) ran on a timeline of months, not weeks.

What freelancing also didn't fix: the isolation. This is consistently underestimated by people who are burnt out in team environments and associate the team with the depletion. The team wasn't the problem, even if it was part of the environment that had become unsustainable. Working alone, answerable only to project deliverables and your own schedule, is isolating in ways that can accelerate the low mood that burnout produces rather than relieving it. I managed this badly in the first months and needed to deliberately rebuild social structure into my working week in ways that felt slightly artificial at first and became genuinely sustaining.

The financial picture, eighteen months in

Eighteen months into freelancing, I earn more than I did as an employee. This is not the typical first-year experience — the typical first year involves income that averages out to less than the salary, though sometimes significantly less and sometimes not, depending on how aggressively you price and how warm your initial network is. My second year crossed the salary threshold in month eight of that year. The income volatility has not fully resolved — there are good months and there are quiet months — but my relationship with the volatility has changed. I budget to the quiet-month floor and treat the surplus as savings and runway, rather than spending to the good-month ceiling.

The things I hadn't budgeted for properly: the accountant (necessary and worth the cost — do not try to navigate self-employment tax alone in any jurisdiction), the professional insurance that two of my first three clients required before signing contracts, the home office costs that add up incrementally over a year, and the pension contribution that nobody makes for you automatically any more.

The question people ask most often when I describe this is whether I'd recommend it. My honest answer is that I'd recommend it for a specific subset of people in a specific set of circumstances: people whose burnout was driven by the loss of autonomy and the cost of performing inside a culture that didn't fit, who have marketable skills and at least a nascent network, who have sufficient runway to weather a slow first two months, and who have thought seriously about what the isolation of working independently actually feels like for them specifically.

For people whose burnout was driven by the work itself rather than the conditions around it, going freelance produces more of the work with fewer of the structures that at least provided distraction from the depletion. That's not the right move. Knowing which kind of burnout you have is the most important question to answer before deciding what to do about it.

The five-question framework is a useful place to start if you're trying to get clarity on that. And the PM to consultant account has the detailed financial numbers on the first 90 days of independent work that I wish I'd had before I made the move.