Prompts for a Business Plan You Can Check and Revise

The most useful prompts for a business plan are not the ones that write it for you. They are the ones that expose what you have assumed and have not yet proven. A business plan that reads well but rests on unexamined assumptions is a document, not a plan. Prompts that push back on those assumptions make the plan more reliable than any amount of polishing. fenbs keeps the approved assumptions and open research tasks as the plan changes, so an AI assistant reading the plan context knows what has been confirmed and what is still an open question.

This page provides prompts organized by the area they test: assumptions, evidence, market, competition, and financials. Use them directly or adapt the language to your situation.

How to use these prompts

Feed each prompt a section of your plan, not the full document at once. A prompt that receives a 40-page plan and asks for assumption analysis will produce a shallow result. A prompt that receives the problem statement section will find the assumptions in that section. Work section by section and record what you learn.

When a prompt reveals an assumption, you have three options: find evidence that confirms it, revise the plan to acknowledge the uncertainty, or note it as an open question for further research. All three are useful. The ones to watch are assumptions that are left neither confirmed nor flagged: they become the surprises later.

Prompts to surface assumptions

Every business plan section rests on things that are treated as true without being stated. Pricing assumes willingness to pay. Distribution assumes a channel will accept the product. Growth assumes customers will come back. These prompts ask the AI to name what the section is taking for granted.

• "Read the following section of my business plan. List every assumption it makes: things treated as true that have not been proven. For each one, say whether it is a customer behavior assumption, a market assumption, an operational assumption, or a financial assumption."
• "Which assumptions in this section would change the viability of the plan most if they turned out to be wrong? Rank the top three by impact."
• "This section assumes [restate a specific assumption]. What evidence would confirm or contradict it, and how would I find that evidence?"
• "What would a skeptical investor ask about this section? List the five questions they are most likely to challenge."

Prompts to find evidence gaps

An evidence gap is a claim in the plan that needs a source or a verification step but does not have one. These prompts help you find them before a reader does. Note: any data point the AI provides about market sizes, competitor revenues, or growth rates should be checked against a primary source. Use these prompts to identify the gaps, then fill them with research you do yourself.

• "List every factual claim in this section that would need a source to be credible in a pitch deck. Mark each one as: confirmed with a source, plausible but unsupported, or unverifiable."
• "I wrote that [specific claim]. What kind of evidence would make this convincing to someone who has not already accepted it, and where would I look for that evidence?"
• "Which claims in this section rest on my own observations or experience rather than external data? For each one, is that sufficient, or does it need corroboration?"

Prompts to test the market and competition

Market and competition sections are where optimism most often displaces accuracy. A total addressable market stated without a methodology, a competitor analysis that omits the strongest players, or a differentiation claim that the competitor also makes: these are common and easy to miss when you are close to the idea.

• "My market size estimate is [amount]. Walk me through the assumptions that would need to be true for this to be right. What methodology would produce this figure, and what is the most likely way it is wrong?"
• "Here is my competitor analysis. Which types of competitors am I most likely to have left out? For each type you name, explain why they might be a more significant threat than the ones I listed."
• "I claim our main differentiator is [differentiator]. If a competitor read this, how would they respond? Which of our differentiators could they match in six months?"
• "What would the market look like if our assumptions about customer behavior are wrong? Describe the plan's performance under three scenarios: best case, expected case, and the case that would cause us to reconsider the idea."

Prompts to check the financials

Financial projections carry assumptions in every line: customer acquisition cost, churn rate, average order value, and headcount growth. A prompt cannot audit your model, but it can identify which line items are load-bearing and which assumptions underneath them are the most uncertain.

• "Here are my revenue projections for years one through three. What are the three or four assumptions that these numbers depend on most? For each one, what happens to year-three revenue if the assumption is off by 50 percent?"
• "My customer acquisition cost estimate is [amount]. What would need to be true about our marketing channels and conversion rates for this to hold? What is the most common way this number is underestimated?"
• "I have projected [growth rate] monthly growth. What milestones or conditions would need to be true for this rate to be achievable, and what would it look like if growth came in at half that rate?"
• "What costs am I most likely to have underestimated? Look at the expense section and name the line items that typically grow faster than founders expect in year two."

Revising and keeping the plan current

A business plan is not a document you write once. It changes as you learn. The useful discipline is to keep track of which assumptions have been confirmed, which have been revised, and which are still open. An AI assistant can help you update the plan, but it needs to know the current state of those decisions to give useful advice.

Use a prompt like: "Here is the current version of the plan. These assumptions have been confirmed: [list]. These have been revised: [list]. These are still open: [list]. Given this, which sections need to be updated and what should change?"

How to do this with fenbs

1. Create a free account and add a project for the business plan.
2. List the key assumptions your plan rests on as tasks in fenbs. Each assumption is a task: open until it is confirmed or revised, then completed with a note on what you found.
3. Record confirmed assumptions as Decisions in fenbs: the assumption, the evidence, and who confirmed it. These become the starting context for the next AI session.
4. Connect your AI assistant to fenbs over MCP from Integrations. Ask it to read the project context, including decisions and open tasks, before advising on the plan.
5. When the plan changes after new information, update the relevant Decisions in fenbs and move the completed research tasks to done. The AI sees the current state of the plan on the next session.