What you will learn
- Generate a wide range of ideas on a single brief
- Push past obvious ideas with constraints and variations
- Combine two ideas into a stronger concept
- Sort ideas by fit, cost, and effort
- Turn a shortlist into testable next steps
- Avoid the trap of accepting the first list you get
Write a brief before you ask for ideas
Brainstorming works better with a brief, even a short one. State the problem, who the idea is for, and any constraints such as budget, time, or format. Without a brief, ChatGPT will generate generic ideas that could apply to anyone. With a brief, the ideas start to fit your situation, and the ones that do not fit are easier to discard.
Include one sentence about what success looks like. If you are naming a product, success might mean a name that is easy to spell and remember. If you are planning content, success might mean topics your audience already searches for. That sentence becomes the test every idea must pass, and it stops the session from drifting into ideas that are interesting but irrelevant.
I run a small online shop selling handmade candles. Give me 20 blog post ideas that help customers choose a candle as a gift. Each idea should be one line, aimed at a first-time buyer, and avoid anything that requires a video. After the list, pick the 5 ideas most likely to bring in search traffic and say why.
Ask for quantity first, then for variety
The first list ChatGPT produces usually contains the most obvious ideas. That is normal and not a failure. The useful work starts when you ask for more: another twenty ideas, ideas that avoid the approach used in the first list, or ideas aimed at a different audience. Quantity forces variety, because the obvious options run out and the later ideas have to be more specific or more unusual.
Try changing one variable at a time. Ask for ideas for a different audience, a lower budget, a shorter format, or a different season. Each change produces a fresh set of options, and one of those sets often contains the idea that fits best. Keep the lists separate so you can compare them rather than merging everything into one long, undifferentiated list.
Combine ideas instead of picking one
The strongest ideas often come from combining two ordinary ones. Take an idea from one list and an idea from another and ask how they could work together. A content idea combined with a format idea, or a product idea combined with a pricing idea, can produce something neither list contained on its own.
Ask ChatGPT to do this combining explicitly. Give it two ideas you like and ask for three ways to merge them, including the trade-off of each version. This turns brainstorming from a choice between options into a process of building, and it usually produces concepts that feel more original than anything in the first lists.
Tip: When a list feels stale, add a constraint instead of asking for more of the same. A limit on budget, time, format, or audience forces different thinking, and constrained ideas are often the most practical ones.
Sort the shortlist honestly
Once you have a shortlist, sort it against your brief. Score each idea roughly on fit with your audience, cost to try, and time to produce. You do not need a formal scoring system; a simple note of high, medium, or low on each factor is enough to reveal which ideas are genuinely promising and which ones only looked good in a list.
Be honest about ideas you like but cannot execute yet. An idea that needs skills, money, or time you do not have is not a bad idea, but it belongs on a later list, not on this week's plan. Parking good ideas for later is part of brainstorming well, not a failure of it.
Turn the winners into next steps
A brainstorm is finished when each surviving idea has a next step. Ask ChatGPT to turn a shortlisted idea into a small plan: the first three actions, what you need to start, and how you will know whether it worked. If an idea cannot be turned into a first action, it is probably still too vague, and it needs one more round of questions before it is ready.
Test small. Pick the idea with the lowest cost to try and run it before committing to the rest of the list. A quick test tells you more than another hour of brainstorming, and the result of the test gives you a better brief for the next session. Brainstorming and testing work as a loop, not as two separate stages.
Frequently asked questions
Is brainstorming with ChatGPT better than brainstorming alone?
It is different, not simply better. ChatGPT generates more options faster, which helps you get past obvious ideas. Your judgement is still needed to pick, combine, and test the ideas that fit your situation.
Why are ChatGPT's first ideas often generic?
The first list usually contains the most common approaches to a problem. Ask for more ideas, add constraints, or change the audience to push past the obvious options.
How many ideas should I generate?
Generate more than you need, then shortlist. Twenty to thirty raw ideas is a reasonable starting pool for most small projects, from which you might keep three to five worth testing.
Can ChatGPT help me choose between ideas?
Yes, by scoring them against criteria you set, such as audience fit, cost, and effort. You make the final call, since only you know your real constraints.
What do I do after a brainstorming session?
Turn each surviving idea into a first action and a way to judge whether it worked. Test the cheapest idea first, then use the result to brief the next session.