The One-Page Business Plan You'll Actually Use

The traditional business plan is a strange ritual: weeks of writing, forty pages, financial projections to 2031 invented to two decimal places — read once by a bank officer and never opened again. Even the person who wrote it doesn't consult it.

Plans fail because they're documents. What a working business needs is a page — short enough to revisit monthly, honest enough to argue with. Here's the nine-box version, and how AI turns filling it from a weekend of dread into an afternoon of sharp thinking.

The Nine Boxes

1. Who it's for — the specific customer. "SMEs" is not an answer; "Colombo clinics with 5-30 staff drowning in appointment admin" is.
2. The pain — what problem, in their words, costs them money or sleep
3. The offer — what you sell and what changes for them after buying
4. Why you — the honest reason they'd pick you over the incumbent and over doing nothing (doing nothing is always the biggest competitor)
5. Channel — how they find out you exist, specifically
6. Money maths — price, cost to deliver, cost to acquire a customer. Three numbers. If you don't know them yet, write your guesses and date them.
7. The engine — what you must do weekly for this to work (outreach volume, content cadence, partner calls)
8. Risks — the three assumptions that kill the business if wrong
9. Now numbers — this quarter's three targets, so the page connects to Monday morning

Filling It With AI as Sparring Partner

The boxes are easy to fill badly. AI's real use here isn't writing the answers — it's attacking them.

• Draft, then demand objections. "Here's my nine boxes. Give me the five strongest reasons this fails, ranked by likelihood." The model has read ten thousand post-mortems; its pattern-matching on failure is genuinely useful.
• Pressure-test the customer definition. "Interview me as a skeptical [your customer]. Don't be polite." Ten minutes of this exposes vague value propositions faster than a month of building.
• Research the boxes you guessed. Market size, competitor pricing, channel benchmarks — a sourced research tool fills your gaps in an hour. Our competitor analysis guide covers the method.
• Kill the jargon. "Rewrite box 3 so a fifteen-year-old understands what we sell." If it can't be done, the offer isn't clear yet — and that's a product problem, not a writing problem.

The Financial Projections Question

If a bank or investor demands the long version, generate it from the page — AI drafts the narrative sections in an hour and spreadsheet tools model the scenarios. But be honest about which document is theatre and which one runs the company. Detailed five-year projections for a business that hasn't survived year one are fiction with formatting; the three numbers in box six, updated monthly with reality, are worth more than all of them.

Keeping the Page Alive

The plan only works as a standing meeting with reality:

• Monthly, 30 minutes: reread the page. What did the month prove or disprove? Update in red.
• The assumption ledger: box eight is a list of bets. Each month, mark them — confirmed, busted, still open. Busted assumptions aren't failure; unnoticed ones are.
• Quarterly, rewrite: a page that hasn't changed in six months means either a perfectly stable business or — far more likely — a page nobody's reading again.

An Afternoon, Honestly

Ninety minutes drafting the nine boxes. An hour letting AI attack them. Thirty minutes of research on the two weakest. By evening you'll have a plan that fits on one page and survives contact with a skeptic — which puts you ahead of most funded startups.

Founders who want the full toolkit — validation, research, build and go-to-market with AI at every step — will find AI for Startups built for exactly this stage. And the free AI Readiness Score will tell you which of these skills you already have.