Build a copy-paste AI prompt pack for an entrepreneur running [BUSINESS DESCRIPTION] at [STAGE]. The next 90-day objective is [PRIMARY GOAL]. The company has these constraints, evidence, and open questions: [CONTEXT AND EVIDENCE]. The intended customer is [TARGET CUSTOMER]. Create exactly six prompts for the most valuable day-to-day work implied by the information I provide. Choose from customer research synthesis, positioning, offer design, weekly priorities, sales-call preparation, pricing analysis, hiring scorecards, investor or advisor updates, experiment design, and decision memos. Do not include a task simply because it is common; select it because it fits the stage and goal. Each prompt must be 120-180 words, self-contained, and written as the entrepreneur speaking to ChatGPT, Claude, or Gemini. Give every prompt a clear title, square-bracket inputs, an explicit output format, and a practical check the model must run. It must request analysis from supplied evidence, not pretend the model has internal company knowledge. Where a task needs primary research, legal, tax, financial, security, employment, or medical advice, say what a qualified professional should validate. Make the prompts operational. For example, a priority prompt should force tradeoffs against a goal and capacity; an offer prompt should distinguish evidence from assumptions; a sales-call prompt should identify discovery questions rather than fabricate pain points. Avoid generic frameworks unless the prompt says exactly how to use them. Output a numbered prompt pack. After each prompt, add one sentence explaining when to use it and one sentence describing the evidence to paste in. Before answering, check that each prompt serves a different decision or workflow and that no prompt asks the model to invent customer quotes, market data, traction, or legal conclusions. Ask up to three clarifying questions only if a required input is missing.
Fill in
| Placeholder | What to enter | Example |
|---|---|---|
| [BUSINESS DESCRIPTION] | Describe what the business sells, to whom, and how it currently makes money. | A two-person SaaS company selling appointment reminder software to independent dental practices on monthly subscriptions. |
| [STAGE] | Enter the current stage, such as pre-launch, early revenue, or growing team. | Early revenue, 18 paying practices and no employees |
| [PRIMARY GOAL] | State one measurable 90-day goal. | Reach 40 paying practices while keeping monthly churn below 4% |
| [CONTEXT AND EVIDENCE] | Paste relevant metrics, customer notes, constraints, current initiatives, and unanswered questions. | Most leads come from referrals. Three prospects said setup felt unclear. Churned users cited staff turnover. Founders have 12 hours per week for sales and marketing. |
| [TARGET CUSTOMER] | Describe the specific customer segment the business is trying to serve. | Owner-managers of independent dental practices with 3-12 staff members |
How to use
- Describe a current business problem and include real notes, metrics, constraints, and decisions rather than only a broad company description.
- Pick the two prompts most tied to your 90-day goal, then use them with fresh evidence each week.
- Treat outputs as working analysis; verify market claims, pricing comparisons, and customer conclusions against source evidence.
- Follow up with: “Adapt prompt 4 for a founder with only five hours this week and one upcoming customer interview.”
Variations
Weekly founder plan
Use this to choose a small set of priorities from competing founder work.
I run [BUSINESS] and have [HOURS AVAILABLE] this week. My 90-day goal is [GOAL]. Here are current metrics, commitments, risks, and possible tasks: [WEEKLY CONTEXT]. Produce a weekly plan with three must-do outcomes, the smallest next action for each, time budgets, a stop-doing list, and one metric or observable result per outcome. Explain tradeoffs against my goal. Do not assume extra staff, budget, or customer demand. Flag work that should wait for evidence.
Customer interview guide
Use this before discovery interviews or churn conversations.
Create a 30-minute customer interview guide for [CUSTOMER TYPE] about [DECISION OR PROBLEM]. Our current assumptions are [ASSUMPTIONS], and our evidence is [EXISTING EVIDENCE]. Give a short opening, 8-10 open questions in a logical order, neutral follow-ups, and questions to avoid because they lead the witness. Include a note-taking table for exact quotes, context, workflow, alternatives, and willingness to pay. Do not write fictional answers or validate our assumptions as facts.
Offer test
Use this to pressure-test a proposed offer before spending on a launch.
Evaluate this proposed offer: [OFFER]. The customer is [TARGET CUSTOMER], and these are our interview notes, metrics, competitors, and constraints: [EVIDENCE]. Produce a one-page offer brief: job to be done, target segment, promised outcome, mechanism, proof available, assumptions, objections, pricing hypotheses, and three low-cost tests. Separate evidence from inference. Do not claim product-market fit or recommend a price as fact; identify what data would change the decision.
Tips
- Give the model raw interview excerpts and call notes, not only your summary; it can distinguish recurring language from your interpretation.
- Use AI to prepare a decision, not to outsource accountability: specify the goal, the available capacity, and the irreversible risks before asking for a plan.
- When discussing pricing, separate what customers said, what they paid, and what competitors list publicly; these are different kinds of evidence.
- A useful founder prompt forces a choice among finite options, not a long list of ideas that all sound worthwhile.
FAQ
Which entrepreneur task should I prompt first?
Start with the decision most connected to your near-term constraint, such as finding a repeatable lead source or understanding churn. A general business plan is less useful when the immediate bottleneck is known.
Can AI do customer research for me?
It can organize and analyze interviews you provide, draft interview guides, and surface patterns. It cannot replace speaking to customers or prove that a pattern exists without evidence.
Should I use AI for legal or financial decisions?
Use it to prepare questions and organize information, then have a qualified lawyer, accountant, tax adviser, or other appropriate professional review consequential decisions.