The Solo Founder's 90-Day Launch Plan: Validation to Revenue

The problem

You quit the job. You have ninety days of savings, a domain name, a working laptop, and a Notion doc with seventeen feature ideas. By week six most solo founders are in one of three failure modes: they have built too much (no users), too little (no shippable thing), or the wrong thing entirely (no problem worth solving). The advice you read on Twitter — “just launch,” “talk to customers,” “find product-market fit” — is correct and useless without a structure that fits into a ninety-day window.

A ninety-day launch is not enough time to build something perfect. It is enough time to validate a problem, ship a thin MVP, get ten paying users, and discover whether the thing you have is worth six more months. The book is structured around that ninety-day clock: each chapter is a specific fourteen-to-fifteen-day phase with concrete deliverables, AI-prompt workflows for the parts a team would normally do, and the explicit “this is where most solos fail” warnings for each phase.

What most people get wrong

They skip validation because building is more fun. The first two weeks of the plan are problem interviews, competitor mapping, and a landing-page test. They produce nothing visible to a portfolio. Most solo founders compress validation into one afternoon, decide their idea is fine, and spend forty days building something nobody wants. The hard part is that validation is not optional. The cost of skipping it is the entire ninety days — you ship to silence in week eight and you have no information about why. The book walks the validation sprint with the exact AI prompts for prospect identification, problem-interview question design, and the landing-page test that takes one day to set up.

They build to specification when they should build to learn. The build sprint is fifteen days. Most solo founders treat that fifteen days as “build all the features I imagined.” What works is shipping the thinnest version that solves one specific problem for one specific user, and letting beta users tell you what else to add. The architecture chapter and build sprint both lean hard on the “what is the absolute minimum that ships” framing — and the AI prompts in those chapters generate the deliberately-incomplete spec that ships in time, not the perfect spec that never does.

A working approach

The ninety days break into six phases of about fifteen days each, plus a fifteen-day “tail” for revenue work:

Days 1-14:   Market Validation
              Problem interviews, competitor map, positioning, landing-page test

Days 15-30:  MVP Architecture
              Stack choice, data model, API design, dev environment

Days 31-45:  Build Sprint
              Core infrastructure, feature implementation, polish, launch prep

Days 46-60:  Beta Launch
              Recruitment, feedback loop, iteration sprint, retrospective

Days 61-75:  Growth Experiments
              Content marketing test, community test, paid acquisition test

Days 76-90:  Revenue & Scale
              Pricing, conversion, operations automation, scaling prep

Each phase has a deliverable. End of day 14: validation memo with a go/no-go decision. End of day 30: a written technical spec and a working dev environment. End of day 45: a working MVP deployed to a real domain. End of day 60: ten beta users who have actually used the product. End of day 75: one channel with positive unit economics or three failed channel tests. End of day 90: paying users or a clear answer that this is not the thing.

The chapter most solo founders read out of order is chapter 8, The AI Toolkit for Solo Founders. It is structured as a stack rather than a list, because the question is not “which tool” but “what role does this tool fill that a team would have a person for?” The roles a solo founder needs filled:

• Coding and development (Claude, Cursor, GitHub Copilot) — the senior engineer you do not have
• Copywriting and marketing (Claude, ChatGPT) — the marketing manager
• Legal and compliance (Claude with verified templates, then a lawyer for actual filings) — the in-house counsel filter, not the lawyer
• Customer support (AI-drafted responses you review) — the support team
• Analytics and decision-making (AI for query writing, dashboards) — the data analyst
• Financial planning (AI for spreadsheet structure, then your judgment) — the CFO filter

The point is not to replace humans with AI. The point is to have AI fill the structural roles that prevent you from doing the work only you can do: building the product, talking to customers, making the bet.

The growth-experiments chapter is the chapter you wish you read first. It is structured as three explicit tests — content marketing, community-led growth, paid acquisition — each with a defined budget, a defined success threshold, and an explicit fail-fast condition. The tests are not “try TikTok and see what happens.” The tests are “post twice a week to the three relevant subreddits for ten days; if no comment chain exceeds five replies, that channel is not your channel.” Most solo founders run zero experiments or twelve at once. The book runs three, in sequence.

Where this scales

The article walked through the six phases and the AI toolkit. The book covers each phase in detail — the problem-interview script with the exact questions to ask, the competitor-mapping prompt that builds the spreadsheet in twenty minutes, the data-model prompt that generates your schema with the relationships drawn out, the deployment-pipeline checklist for day 33, the beta-recruitment outreach template, the three channel tests run sequentially in days 61-75.

The handling-disruptions section in chapter 8 is the chapter most solo founders need on day 47 when the build is behind, the beta users are silent, and a freelance gig is paying for the rent. The recovery move is to compress: cut features, cut tests, ship what is there, accept that the ninety days will produce a smaller answer than you hoped, and that the smaller answer is still infinitely more than month-six paralysis would produce.

Included with the book

• Solo Launch Checklist (markdown and PDF) — every deliverable from the 90-day plan as a single checkable reference. Print it. Cross items off. Hang it where you see it.

Get the full picture

Readers of this also chose

Questions readers ask

Is this only for SaaS or also for other product types?

The 90-day structure works for SaaS, info-products, marketplaces, and physical-good D2C. The MVP architecture chapter is most SaaS-specific. The validation, growth, and revenue chapters apply across product types.

What if I need a refund?

Checkout runs on Lemon Squeezy. The standard refund window applies. You keep the PDF either way.

I have a day job. Does this work part-time?

Yes, but stretch the timeline. The book has a part-time variant that extends 90 days into 150 days, with explicit guidance on what to compress and what to preserve.

What stack does this assume?

It is stack-agnostic. The architecture chapter walks through the choice — typescript + postgres, python + sqlite, ruby + rails, whatever fits your skill set. The AI prompts work on any stack.