Prompt Generator
GeminiTemplate, 5 blanks to fill

You are a skeptical board member reviewing a business before a big decision. Your job is to find what breaks it. The business: [BUSINESS] The decision we are about to make: [DECISION] Numbers and facts we are working from: [KEY NUMBERS] What we are most worried about: [WORRIES] Do this: 1. List the six assumptions the business depends on most, each in one sentence, and mark how much evidence we have for each: strong, some or none. 2. Describe three scenarios over the next [TIME HORIZON]: a best case, a base case and a worst case. For each, say what happens to customers, revenue, costs and cash, using only our numbers or clearly labelled guesses. 3. Name what would trigger the worst case and the earliest warning sign we could watch for. 4. Group the risks into market, money, operations and strategy. Rank the top eight by likelihood and impact (high, medium, low for each) in a table. 5. Say which two inputs would change the outcome most if they moved by a small amount. 6. For the top five risks, give one action to prevent it and one fallback if it happens anyway. 7. Finish with the decision points: the dates or signals at which we should stop, change course or continue. Label every guess as a guess. Do not soften the worst case.

The words in brackets are the parts you replace with your own details.

What each part of the prompt does

Board member framing
A skeptical reviewer looks for what breaks. That framing pushes Gemini past the polite risk lists that only mention competition and the economy.
Evidence rating
Marking each assumption as strong, some or none shows where you are guessing. An assumption with no evidence that the whole plan depends on is usually the first thing to test.
Warning signs
Step 3 asks for the earliest signal of the worst case. A risk you can see coming three months ahead, such as slower repeat orders, is much cheaper to handle than one you notice in the bank balance.
Decision points
The closing list of stop, change and continue signals turns the analysis into dates on a calendar. Without it, a scenario exercise tends to end up in a folder and never gets looked at again.

Variations to try

Same idea, one thing changed.

  • Pre-mortem version

    You are a skeptical board member reviewing a business before a big decision. The business: [BUSINESS] The decision we are about to make: [DECISION] Numbers and facts we are working from: [KEY NUMBERS] What we are most worried about: [WORRIES] Imagine it is the end of [TIME HORIZON] and the decision has clearly failed. Write the story of how it failed in eight short steps. Then list the five causes in that story we could still prevent today, with one action and one warning sign each. Label every guess as a guess.

  • Cash runway focus

    You are a skeptical board member reviewing a business before a big decision. The business: [BUSINESS] The decision we are about to make: [DECISION] Numbers and facts we are working from: [KEY NUMBERS] What we are most worried about: [WORRIES] Model cash month by month over [TIME HORIZON] in a best, base and worst case, using only our numbers or labelled guesses. Show the month cash would run out in each case, the three costs we could cut fastest, and the revenue level we must hold to stay safe. Do not soften the worst case.

If the result isn't right

  • Give real figures in [KEY NUMBERS], even rough ones. Without them the scenarios turn into adjectives and you cannot compare the three cases.
  • If the worst case looks too mild, reply "make the worst case what happens if our two weakest assumptions both fail".
  • Share the risk table with someone outside the business and ask which risk they would rank higher. Their answer is often the one you were avoiding.
  • Put the decision points in your calendar on the day you run this, not later.