# Bootstrapping a startup: what it means and what it costs Published 2026-09-19 by Panche Isajeski · Tallinn Source: https://bootstrappedmakers.com/blog/what-bootstrapping-actually-means --- In 30 seconds- Bootstrapping means no investors. Grants count too, and nobody tells you. - The guides skip the hardest part: a finished product nobody is buying. - That is a distribution problem, and those move in weeks. - The fix: ask people by name to pay. Every week. Bootstrapping means funding a company from its own revenue and your own money, instead of selling equity to investors. That is the definition, it is correct, and it will not help you at all. It describes a company by something it did not do. You could say the same about a company that failed in month three. I have built two companies this way. One reached a peak valuation of about $26M and I exited it. The other sat at €0 for four months with a finished product before it made a single euro. Neither outcome had much to do with the definition. **$2M**in grants for Kromatika. No investors. **$26M**peak valuation, November 2021 **4 mo**at €0 with AgentaOS, product finished **€3,600**a month, 30 days after I started asking ## Bootstrapped does not mean no money This is the first thing the guides get wrong. They present two options — raise venture capital, or fund it from savings — and most people reading assume they need savings they do not have. There is a third source almost nobody writes about: **grants**. Kromatika, which I started in 2021 under the name Aristokrates, ran on **$2M in grants**. No round, no deck, no investor, no equity given up. It reached 1,000 users and a peak valuation of around $26M in November 2021, and I exited in 2023. That is still bootstrapping by any sensible reading: nobody owned a piece of it but me. But it is not the garage-and-savings story, and pretending otherwise is why the word puts people off. Ecosystem funds, innovation agencies, EU programmes, university and regional schemes — this money exists, nobody takes a share of the company for it, and applying is a skill rather than a lottery. So the real question is not "do I have savings." It is: **what can fund this that does not cost equity?** Revenue, your own money, grants, or the consulting work you already do. Most bootstrapped companies run on a mix. 01**Revenue** Customers pay for the next month. The cleanest money there is. 02**Your money** Savings or client work. What most guides assume is the only option. 03**Grants** The one nobody writes about. Kromatika ran on $2M of them. ## The part the stage-by-stage guides skip Search for how to bootstrap and you will find the same shape every time: Stage 1 validate, Stage 2 build, Stage 3 grow, Stage 4 scale, with an MRR band attached to each. It reads like a staircase. Do the thing, reach the number, climb. It is a tidy model and it hides the only part that is actually hard. In April 2026 I split from the consulting company I had co-founded and continued alone. The product became AgentaOS, a merchant of record. It took four months to find product-market fit — not adding a feature, changing the model entirely. Then the product was finished. Good, live, priced, with a working checkout. **And for four months it made €0.** Not €200. Zero. I was not stuck at a stage. I had passed the stage. I talked to founders, wrote audits nobody asked for, rewrote the pricing page more times than I will admit, and seriously considered that I had misread the whole thing. On 10 August I launched the version those four months had taught me to build, and started writing to specific people, by name, asking whether they would pay. Thirty days later it was at **€3,600 MRR**. By 14 September, **€6,800**. No ads. No sales team. No new feature. The staircase has no step for that. It has no step for four months of a live product earning nothing while you work out what it should have been, and no step for the fortnight after you start asking. The product did not change. The month did. AgentaOS monthly recurring revenue, 2026. Finished, live and priced the whole way along. _Chart: Monthly recurring revenue: zero from April through early August 2026, then 3,600 euro by 10 September and 6,800 euro by 14 September._ _Four months at €0 with a working product, then €0 to €3,600 MRR in the thirty days from 10 August, and €6,800 by 14 September. No ads, no sales team, no new feature._ ## What actually decides it Both times I thought the product was the work. Build it well enough and people come. Both times the product sat there, finished and good, while nobody came — because nobody had heard of it. What moved the number was distribution: who I talked to, where they already were, how often I showed up, and whether I asked them to pay. Kromatika grew because I put it in front of the right people every day. AgentaOS sat at €0 for four months precisely because I stopped doing that in order to build, and it moved the month I started again. This is the asymmetry that makes bootstrapping hard in a way funded companies do not experience. A funded company can buy distribution while it works out the product. You cannot. You have to build the thing *and* carry it to people, and the carrying is the part nobody writes up. The build gets written up. The raise gets written up. The part where you have a good product and no customers and no idea whose inbox to walk into is the part everyone skips — and it is where most products quietly die. **Built a SaaS product, and nobody has paid yet?**Three free playbooks. The first is in your inbox a minute from now.[Steal my playbook →](https://bootstrappedmakers.com/#hwl) ## What it costs, honestly Not money. The costs that actually bite: - **Revenue is your runway.** Customers funded Kromatika, which sounds noble and mostly meant that if nobody paid this month, there was no month. - **Nobody hands the work off.** There is no team to pass it to, and no one waiting to tell you it is good enough. - **You stay present when everyone else is on holiday.** A solo company does not have a quiet August. - **Nobody notices if you stop.** No board meeting, no investor update. Nothing external forces the week to have a shape. Those are real and they are the reason people quit, far more than running out of money. ## When bootstrapping is the wrong choice It usually is not, but it genuinely can be. Raising makes more sense when the product cannot exist in a small version — heavy infrastructure, regulatory licences, hardware. When the market is a winner-takes-all race and being second is worthless. When the thing needs a sales team from day one because the buyer is an enterprise with a nine-month procurement cycle. If none of those describe you — and for most software they do not — the choice is not between bootstrapping and raising. It is between bootstrapping and not starting. ## How to tell if you are actually bootstrapped One test: **if nobody pays you this month, what happens?** If the answer is "the month does not happen," you are bootstrapped, whatever is in the bank. That is the constraint that shapes every decision — what you build, what you charge, who you talk to, how fast you find out you are wrong. It is also, once you stop fighting it, the useful part. A funded company can spend two years finding out nobody wanted it. You find out in a fortnight, because the money either arrives or it does not. ## What I would tell myself in 2021 The product is not the work. It is the ticket that lets you start the work. If the number on that page is the thing you are stuck on, that is its own problem: [what 230 companies actually charge in 2026](https://bootstrappedmakers.com/blog/saas-pricing-models-2026), and why investor preference points the opposite way to what works without investors. The work is carrying it to specific people, by name, and asking them to pay — every week, whether or not you feel ready, and long before it feels finished. Do that from day one and the four months at €0 do not happen. I have now learned this twice, expensively. That is the whole reason I run a community around it rather than writing another guide: distribution is a habit, and habits need witnesses. Figures here are my own: Kromatika, $2M in grants, ~$26M peak valuation November 2021, exited 2023. AgentaOS, €0 to €3,600 MRR in the thirty days from 10 August 2026, €6,800 on 14 September 2026.