For most of software history, the phrase "one-person company" described a compromise. One person could write the product or talk to customers or do the books, pick one, neglect two. The standard advice was to raise money precisely because the workload of a real company was structurally impossible for an individual. That advice is now quietly out of date.
What actually changed
Three cost curves crossed in the last few years, and their intersection is the story.
The marginal cost of competence collapsed
AI assistance did not make experts obsolete; it made adequacy nearly free. A strong backend engineer can now produce a passable marketing page, a serviceable privacy policy, competent ad copy, and a working mobile layout, none award-winning, all shippable. The one-person company was never blocked by its founder's specialty. It was blocked by the eleven other jobs. Those jobs did not disappear, but their floor cost dropped from "hire someone" to "supervise a machine for an hour."
Infrastructure stopped charging admission
A credible product in 2026 runs on rented or open-source everything: payments as an API, a global CDN for pennies, open-source software that ten years ago would have been a licensed enterprise product, and hosting that scales from zero. The capital requirement for a software business has fallen so far that the remaining costs are mostly your own attention. Attention, notably, is the one input AI genuinely multiplies.
Distribution unbundled from institutions
You no longer need a sales force or a press relationship to reach a niche. You need to be findable at the moment someone searches for their problem, which is a content and patience game, not a headcount game. A single well-written article answering a question your customers actually ask will outperform a mediocre sales team in a niche market, and it works night shifts forever, free.
The new shape of the solo company
The viable one-person company is not a smaller version of a startup. It has a different shape:
- Niche by design. You are not capturing a market; you are capturing a corner nobody with a payroll can afford to serve. A $30k/year problem shared by four thousand businesses is invisible to a venture-backed firm and life-changing for an individual.
- Boring technology, ruthlessly standard. Every exotic choice in the stack is a future 3am page with your name on it, and there is no one else's name. Solo companies run on the most battle-tested software available, and their creativity budget is spent entirely on the product.
- Automation as headcount. The mental model that works: you are not a freelancer with tools, you are a manager whose staff happens to be software. Support triage, monitoring, content drafts, invoice chasing, each automated process is an employee you never recruit, pay, or lose. Your calendar is the org chart.
- Margins as the moat. A solo company at 90% gross margin can survive prices that would kill a funded competitor. Being structurally cheaper is a strategy, not an accident.
Where the ceiling still is
Honesty requires the other half of the ledger, because the failure mode of this article's genre is unbounded optimism.
Trust-heavy sales don't compress. If your product needs a procurement process, security questionnaires, or a golf game, one person cannot substitute for an enterprise sales motion. Solo companies sell things people can buy with a credit card and adopt without a committee.
The bus factor is one. Sickness, burnout, a family emergency, the company's uptime is coupled to yours. Mitigations exist (radical automation, documented runbooks, generous margins that buy slack), but the coupling never fully breaks.
AI adequacy is a floor, not a ceiling. When every solo founder can generate adequate everything, adequate becomes worthless. The differentiator moves up a level: taste, positioning, and the accumulated judgment of actually knowing your customers. The machine drafts; whether you can tell a good draft from a plausible one is suddenly the whole game.
The practical takeaway
If you have been waiting for permission, funding, a co-founder, a hiring budget, the honest news is that the excuse inventory has run low. The viable path in 2026 looks like this: find a specific, recurring, budgeted problem (ideally by reading the complaints already written about it); build the smallest tool that deletes the problem; charge real money from day one; automate every job that repeats; and put your irreplaceable hours only where judgment lives.
The one-person software company used to be a lifestyle consolation prize. It is now simply a company, small on the org chart, normal-sized in the bank account. That is a genuinely new fact about the economy, and most people have not noticed it yet. The window where that is true is the opportunity.
Related: the boring stack, choosing technology that never pages you at 2am.
Discussion
Sign in with Google or just a name. No email link, no password to remember.