Validation & Idea Testing
A founder's method for brainstorming business ideas: gather raw material, run five techniques solo or in a small group, then score a shortlist of three.

One focused session can give you 30 or more raw business ideas and a shortlist of three worth testing. To get there, you brainstorm ideas from material you collected beforehand, use techniques that force variety, and score the results before you fall for any one of them.
This guide covers each part: what to gather first, five techniques, how to run the session alone or with a few people, and a scoring sheet you can copy.
Staring at an empty page and waiting for inspiration rarely works. A session goes faster when you walk in with two lists: problems you have noticed and assets you already have.
For seven days, write down every moment something wastes your time, money or patience. Note who had the problem, what they did about it and what it cost them. Include problems you hear from coworkers, clients and friends, not only your own.
Essayist Paul Graham opens his essay on startup ideas with the advice behind this habit: "The way to get startup ideas is not to try to think of startup ideas." He suggests looking for problems instead, "preferably problems you have yourself."
Expect anywhere from a handful to a few dozen entries in a week. Some will be trivial. A few will repeat, and the repeats are your best starting points. If you want a structured way to dig into what people need, the methods for how to identify customer needs pair well with this journal.
Next, list what you could put to work tomorrow. Use four headings:
If the skills list comes out thin, borrow vocabulary. O*NET, the occupational database sponsored by the U.S. Department of Labor, groups transferable skills into folders such as Resource Management Skills and Social Skills. Reading through them often reminds you of abilities you take for granted.
Each technique below pulls ideas from a different direction. Use at least three in one session so your list does not come out as 30 versions of the same thought.
Constraints narrow the field, which makes it easier to think. Pick a problem from your journal and run it through prompts like these:
Write at least one idea per prompt, even a weak one. Weak ideas often turn into good ones two steps later.
SCAMPER is a checklist for changing something that already exists. Take a product or service you know well and ask one question per letter:
| Letter | Question to ask |
|---|---|
| Substitute | What part could you swap for something cheaper or faster? |
| Combine | What two services do customers buy separately today? |
| Adapt | What works in another industry that could work here? |
| Modify | What if you made it bigger, smaller or more frequent? |
| Put to other use | Who else could use this, for a different job? |
| Eliminate | What step could you remove entirely? |
| Reverse | What if the customer did part of the work, or you did more? |
A dog walker who applies "Combine" might pair walks with basic grooming. "Reverse" might turn a cooking class into a service where the chef cooks in the client's own kitchen.
Write one customer type in the center of a page, such as "new parents" or "independent contractors." Branch out to the tasks they do each week, then to the frustrations inside each task, then to possible offers.
Starting from a person keeps your ideas tied to someone who could pay. Three levels deep, most branches end in a specific problem you can name.
Ask the opposite question: how could you make this problem worse? List every answer, such as "hide the prices," "require three phone calls," or "make people wait two weeks."
Then flip each answer. "Hide the prices" becomes "publish clear, flat prices." Each flipped answer is a feature, and sometimes a whole business, that the current options lack.
A general AI assistant is most useful after you have your own list. Paste in your ideas and ask it to find the weakest assumption in each one or suggest a customer you have not considered. You can also ask it to combine two of your ideas into one.
Treat its suggestions as more raw material, then score them like everything else. Using AI tools to generate ideas from scratch is a different job, covered in the guide to the business idea generator and the prompts that work with one.
The classic rules come from Alex Osborn, whose influential book (cited here in its 1957 edition) proposed brainstorming as a method of group problem solving. A 1987 paper summarized them as four rules: aim for as many ideas as possible, welcome wild ones, build on or combine ideas already suggested, and do not criticize.
Those rules hold up. The group format is the weak part. In research on brainstorming groups, Michael Diehl and Wolfgang Stroebe reviewed 22 earlier experiments. Eighteen found that people working alone, with their ideas pooled afterward and duplicates removed, outperformed the same number of people brainstorming together on the number of ideas.
Their own experiments pointed to production blocking as the main cause. In a group, only one person speaks at a time, and the others may forget ideas while they wait their turn. The fix is simple: generate alone first, then share.
This format works for one founder or a group of up to five.
Solo founders can run the same steps. Swap step 3 for a sparring round with a friend or an AI assistant, and take a short break before scoring so you judge with fresh eyes.
Hold all judgment until generation is over, then judge every idea the same way. Give each idea a score from 1 to 5 on four criteria:
| Criterion | Score 5 when... | Score 1 when... |
|---|---|---|
| Pain | The problem repeats often and costs people money or time | It is a minor annoyance |
| Reach | You can message 20 likely buyers this week | You have no route to buyers |
| Fit | Your skills cover most of the work | You would need to learn or hire for most of it |
| Test cost | You can test it this month for under $200 | A test needs months of building |
Add the four scores for a total out of 20. Ideas that score 14 or more go on the shortlist. If more than three qualify, keep the three highest.
Here is a hypothetical example. A bookkeeper finishes a session with 34 ideas and scores her top clusters:
| Idea | Pain | Reach | Fit | Test cost | Total |
|---|---|---|---|---|---|
| Monthly bookkeeping for food trucks | 4 | 4 | 5 | 4 | 17 |
| Workshop on organizing receipts for contractors | 3 | 4 | 4 | 5 | 16 |
| Receipt-scanning app for contractors | 4 | 3 | 2 | 1 | 10 |
The app has real pain behind it, but she cannot build software and a test would take months. The two service ideas go on her shortlist, and the app goes into a "later" file. Nothing is thrown away: a later session might find a cheaper way to test it.
A shortlist of three is the end of brainstorming and the start of evidence. The next step is to test the idea before you build, with interviews and small experiments that show whether anyone will pay. If your session came up thin and you want examples to react to, the list of small business ideas under $5K is a useful source of raw material for the next round.
Keep your problem journal running after the session. The second session is faster than the first, because you start with a backlog instead of a blank page.
EntraPath, EntraWorld's optional 7-phase founder roadmap, starts with Discover and Validate, so your shortlist has a place to go next. Join EntraWorld free and take your top three ideas into the Validate phase.
Start free. The first 5,000 Premium memberships include a full year of every tool.
Join EntraWorld free →