Prompts for Business Ideas and Testing Their Assumptions

The most useful AI prompts for business idea work push past the initial list of ideas and into the assumptions underneath each one. An assistant that helps you generate options is easy to find. One that helps you test whether a specific assumption holds is more valuable. Tools like fenbs keep the chosen direction, the rejected assumptions, and the validation tasks in one record, so the next session starts from what was actually decided rather than re-exploring ground you already covered.

The templates below cover four stages of business idea work: generating options, narrowing down, testing a specific assumption, and planning the first validation task. Each includes the context to fill in and a note on what a useful response looks like.

What a business idea prompt needs

A business idea prompt without constraints produces a generic list of options. Constraints make the list useful: they scope the ideas to your resources, your market, and the problem you are actually solving. Before you send a generation or testing prompt, include these elements.

• The problem: what specific situation the business is trying to improve, and for whom.
• Your constraints: budget range, time available to test, skills you have, and markets you know.
• What you have already considered and rejected: so the assistant does not repeat it.
• The criteria for a good idea: what success looks like for you at this stage.
• The stage you are at: generating, narrowing, testing, or planning the first step.

Prompts for idea generation

A generation prompt that specifies your constraints and the problem you are solving produces ten ideas you can actually evaluate rather than a list of anything that is technically a business. Include what you have already considered so the assistant does not spend three of its ten slots on directions you already ruled out.

Generate 10 business ideas based on the following:

Problem to solve: [describe the situation you want to improve and for whom]
My constraints: [budget range, time to first revenue, skills I have, markets I know]
Already considered and rejected: [list with brief reason for each]

For each idea:
- Name and one-sentence description
- The specific customer and their problem
- Why someone would pay for it
- The biggest assumption that must be true for this to work

Prioritize ideas that can be tested with low cost before significant investment.

Prompts for comparing and narrowing down

Once you have a list of ideas, use a comparison prompt to evaluate them against the criteria that matter most to you. Stating the criteria in the prompt prevents the assistant from picking criteria it thinks you should care about, which may not match your actual situation.

Compare these business ideas against the criteria below.

Ideas:
[list the ideas you are comparing]

Criteria (in order of importance):
1. [most important criterion, e.g. testable within 30 days]
2. [second criterion]
3. [third criterion]

For each idea, rate it on each criterion (high / medium / low) and give a one-sentence reason.
End with a recommendation of the top two ideas and why.
Note any assumption in your recommendation that I should verify.

Prompts for testing a specific assumption

Every business idea rests on assumptions. The useful question is not whether the idea is good in general but whether the most critical assumption holds for your specific customer in your specific market. This prompt helps you identify what to test first and design a test that will actually tell you something useful.

Help me test the key assumption in this business idea.

Business idea: [describe the idea]
Key assumption: [the one thing that must be true for this to work]
My resources for testing: [what I can spend in time and money]
Target customer: [who I am testing with]

For this assumption:
- What would prove it is true?
- What would prove it is false?
- What is the simplest test I could run in [timeframe] with [resource]?
- What result would cause me to stop or change direction?

Be specific about what counts as evidence.

For prompts that take a validated idea into a full plan, prompts for business plans covers the structure for a complete plan document. For broader business use of AI tools, ChatGPT for business covers task types across the business.

Prompts for planning the first validation task

The most common failure at this stage is planning a long validation project when a short one would give you the same answer. This prompt designs the smallest task that would meaningfully change your confidence in an idea, so you spend resources on the test rather than on building something based on an untested assumption.

Design the first validation task for this business idea.

Idea: [describe it]
Assumption to test: [the one thing you are trying to confirm or rule out]
Resources available: [time, money, access to potential customers]

Propose a task that:
- Takes no more than [your timeframe, e.g. two weeks]
- Requires no more than [your budget]
- Produces a clear answer about the assumption
- Can be done before building a full product

Describe the task step by step, the result that confirms the assumption, and the result that would cause you to pivot or stop.

Keeping the chosen direction and rejected paths

Business idea sessions produce a lot of output: generated lists, comparisons, test designs, and validation results. The decisions that shape the direction of the business are worth recording separately from that output. Which idea you chose and why, which assumptions you tested and what they showed, which directions you ruled out and the reason: these are the decisions that every future session should start from.

Without a shared record, the next session can relitigate a direction you already settled, run a test you already ran, or generate ideas you already rejected. Writing the key decisions into a record as you make them is what prevents that. Prompts for business plans shows how to build on those decisions once the idea is ready to become a plan.

An assumption test result is particularly worth recording in detail. What you tested, how you tested it, who you tested it with, and what the result showed are all important parts of the record. A test that showed the assumption does not hold is as valuable as one that confirmed it: both change the direction of the work. Record both kinds so the next session starts from what you actually know rather than from the untested assumptions the idea began with. A rejected assumption with a reason is a constraint that keeps future work from heading in a direction you already explored.

How to do this with fenbs

1. Create a free account and add your business idea as a project.
2. Write the chosen direction and the criteria behind the choice into Decisions and rules.
3. Add each validation task as a task on the board with its plan, and update it when you have a result.
4. Record rejected ideas and the reason they were ruled out as decisions, so you do not revisit them without new information.
5. Connect your assistant over MCP from Integrations so it reads the project state before each session.

fenbs does not evaluate the commercial viability of a business idea or connect to market research databases. It keeps the decisions, tasks, and context your idea development depends on.