Eight people. Two hours in the calendar. A whiteboard, a fresh pack of markers, and a founder who opened the session by explaining, in some detail, the idea he had already decided on.
You know how the next ninety minutes went. A few people agreed with him. One person offered something different and watched it die in the silence that followed. Two people did not speak at all. At the end there were four ideas on the board and three of them belonged to the person who called the meeting.
As everyone packed up, someone said the thing that people always say after a session like that.
“This is why I don’t really believe in brainstorming.”
Here is what makes it so convincing. The research appears to be on its side.
Brainstorming as a named method comes from Alex Osborn, an advertising executive who set out four rules in his 1953 book Applied Imagination: go for quantity, withhold criticism, welcome wild ideas, and combine and improve the ideas you have. That is the whole foundation of the thing every company now does badly in a meeting room.
Then the psychologists went looking. In 1987, Michael Diehl and Wolfgang Stroebe published a study in the Journal of Personality and Social Psychology asking why groups underperform, and the finding was blunt. Groups brainstorming together produce fewer ideas than the same number of individuals working separately. Not slightly fewer. Reliably fewer.
The main culprit they identified has a name: production blocking. Only one person can talk at a time. While you wait your turn, you forget your idea, or you rehearse it, or you quietly decide it is not as good as what you just heard. The queue eats the thinking.
So the sceptic in that room had data behind him, whether he knew it or not. If you put eight untrained people around a whiteboard, brainstorming really does produce less than those same eight people would have produced alone at their desks. The activity fails. Measurably.
Which is where I would have left it, if I had not been reading Tom Kelley.
Kelley was the general manager of IDEO, the design firm behind a long list of products you have used without thinking about who designed them. In The Art of Innovation he writes about people who dismiss the method, and he puts his finger on exactly what they are missing:
“They overlook the possibility that brainstorming can be a skill, an art, more like playing the piano than tying your shoes.”
Every one of those studies measured untrained groups. That is not a criticism of the research, it is what the research was designed to test. But look at what we did with the result. We took a room full of people who had never been taught the method, watched them perform it badly, and concluded the method was worthless.
Imagine testing whether pianos work by sitting eight people who have never played in front of one for an hour, recording the noise, and publishing a paper called “The Myth of the Piano.”
Tying your shoes is a thing you can do. Playing the piano is a thing you become able to do. Almost every company in the world has quietly filed brainstorming under the first category, and then felt betrayed when it behaved like the second.
Now, I want to be careful here, because this is the point where most articles get lazy and hand you a list of rules as if rules were the answer.
They are not, on their own. IDEO does publish its seven rules, and they are good ones: defer judgment, encourage wild ideas, build on the ideas of others, stay focused on the topic, one conversation at a time, be visual, go for quantity. Kelley’s own book adds seven secrets on top, including sharpening the focus into a pointed question before anyone walks in, numbering the ideas so the room can feel its own output, and getting physical rather than only talking.
You can print all of that on a poster and still run a terrible session. I have watched it happen. Rules describe what skilled practice looks like from the outside. They do not install the skill.
What actually changes the outcome is smaller and harder. Somebody has to prepare the question in advance, because a vague prompt produces vague output and no rule fixes that. Somebody has to hold the room, which means interrupting the person who is monologuing, including when that person is the boss. And the most senior person present has to accept, genuinely, that their idea enters the room with no more status than anyone else’s.
That last one is where I see it break most often in founder-led businesses, and I see it constantly across the corridor I work in. In a lot of African business cultures, seniority carries real weight in a room, and disagreeing upward in public costs something socially. That is not a flaw to be scolded out of people. It is a condition you design around. If you know the hierarchy will suppress dissent, you do not run an open free-for-all. You have everyone write silently first, then collect, then discuss. You have removed production blocking and the deference problem in one move, before anyone has said a word.
I learned to think this way long before I learned design, in a microbiology lab. You are not taught to have the right hypothesis. You are taught to generate a spread of them, then design experiments that kill most of them off. The generating and the judging are separate activities, deliberately, because doing both at once corrupts them both. That is precisely what “defer judgment” means, and it is why the rule exists.
I use the same discipline now when I am building for African markets. There is no single answer that survives twenty countries. Bandwidth, device age, payment behaviour, literacy, trust in institutions, all of it varies enormously between markets that outsiders treat as one place. You do not solve that with one good idea. You solve it by generating enough real options that at least a few of them fit conditions you did not personally experience. Quantity is not a creative indulgence. It is how you cover ground you cannot see from where you sit.
So here is the reframe I would offer to anyone who has written brainstorming off.
You did not test the method. You tested a room of untrained people doing an impression of the method, and it failed the way untrained things fail. The failure was real. The conclusion you drew from it was not.
Brainstorming is not a meeting you hold. It is a capability you build, and like any capability it is embarrassing at first and useful later.
And there is a simple way to tell whether yours is improving. At the end of the session, count the ideas on the board that nobody in the room walked in with. Not the ideas you liked. The ideas that did not exist ninety minutes earlier.
If that number is zero, you did not run a brainstorm. You held a meeting where somebody’s mind was already made up.