How to Generate Ten Business Ideas in One Afternoon
Having IdeasOctober 2, 2026
The Charlie Page Letter
26 years of experience and common sense delivered to your inbox.
To generate ten business ideas in an afternoon, stop trying to invent and start filling slots. Four prompts, one hour, one page per idea, and a rule that nothing gets judged until the tenth one is written. Volume first, judgment later, in that order.
The reason this works is that quality and quantity are different skills, and mixing them ruins both. When you generate and critique at the same time, the critique wins every round, because criticism is faster than creation. You end the afternoon with one idea you love and no way to tell whether it's any good.
I run this exercise with clients who arrive certain they have no ideas, and every one of them leaves with at least eight. The exercise isn't magic. It just removes the two things that actually stop people: the blank page and the inner editor.
What's the setup for the exercise?
Three hours, a timer, and a page per idea. Each page has five lines: the person, the friction, what they do today, what you'd sell, and how you'd reach the first ten of them. If a line stays empty, the idea doesn't get a pass, it gets parked.
No judging while you fill pages. An idea that sounds silly gets its five lines too. Silly ideas are often a real idea wearing a costume, and you can't tell which until it's on paper where you can look at it.
Keep the timer visible. Deadlines do something that good intentions don't: they make you stop polishing the first two and move on. Most people's tenth idea is dramatically better than their first, and they never get there because they spent the afternoon on number one.
Which prompts fill the pages fastest?
Prompt one: three workarounds you've seen this month. Every spreadsheet-as-database, every group chat doing a service's job, every person paying a human to do something software should do. That's usually four ideas before you've warmed up.
Prompt two: three things you or someone near you pays for and grumbles about. The grumble is the product. People who pay while complaining are waiting for a better version and will switch the moment a credible one shows up.
Prompt three: three people whose job you understand well enough to describe badly. Then ask what part of their week they'd pay to never do again. Prompt four: three industries you have an unfair view into, because you grew up in one, dated one, or worked next to one.
How do I stop the ideas from being vague?
Force the person line to be a sentence with no plural nouns in it. "Freelancers" fails. "A freelance video editor with four repeat clients and no invoicing system" passes, because you can picture them and you know where they sit online.
Force the today line to name a real behaviour, not a feeling. "They're disorganised" is a feeling. "They keep a phone note and re-send the same three files weekly" is a behaviour you could go watch, and behaviours are what you build against.
Force the reach line to name one place. A subreddit, a trade Slack, a list you already have, a door you can knock on. Ideas without a reach line aren't bad, they're just not ready, and parking them is a better use of the afternoon than pretending.
How do I pick from ten ideas?
Score them on two things only: how well you know the person, and how clearly you can see the money. Two lines, one to five each. The idea you're most excited about usually scores badly, and that's information rather than a tragedy.
Then cut to two. Not one, two, because a single idea has no context and you'll fall in love with it. Two ideas get compared, and comparison is the cheapest form of criticism available to you.
Finally, put both through something that isn't you. A friend will tell you it's great. A pressure test won't: it asks the investor questions, scores the answers, and tells you which one falls apart before you spend a month on it.
What if the ideas are all variations of one thing?
Good. That means you found a lane you actually understand, which is rarer than variety. Five ideas about the same trade is not a failure of imagination, it's a market you can see from the inside.
Now go deeper instead of wider. Ask which of the variations shares the same customer and which shares the same mechanism, and whether any two of them should be one offer. Overlap is usually a hint that you're circling something real.
The risk is anchoring on the first version you thought of. If your ten ideas all serve the same person in slightly different ways, that's a cue to test the sharpest one and put the rest aside as future offers, not to build all of them.
What do I do the day after?
Send the two survivors through the same four questions: who exactly, what's in it for them, what they'd pay, and how you reach the first hundred. Answer them in writing, because vague answers feel fine in your head and read badly on paper.
Then find five people who match the person line and ask what they did the last time the friction showed up. Not whether they'd use it. What they did. Five conversations in a week will teach you more than another afternoon of generation.
Keep the ten pages. In six months, when you know more about the market, two of the parked ones will look obvious, and you'll be glad you wrote down how you'd reach them while you still remembered.
The afternoon isn't about finding the idea. It's about ending up with a short list you can argue about instead of a single hope you have to defend.
Start from the friction you already noticed here, then let the free pressure test tell you which survivor is weakest, and track how the answer changes in My Tests as you fix it.
Frequently asked questions
The Charlie Page Letter
26 years of experience and common sense delivered to your inbox.