Brainstorming Doesn't Work. These Techniques Do

A conference room, a flipchart, three colors of marker. Someone writes "How do we boost sales?" in the middle and tells the team to throw out ideas, no bad ones allowed. Five minutes in, the same two people are doing all the talking while everyone else nods and checks their phone.

An hour later the sheet is full and the room feels productive. Most of what's on it is unusable. Here's the part that stings: if those six people had sat in separate rooms and written ideas alone, together they would have produced more of them, and better ones. That runs against everything you've heard about teamwork, and it's exactly what the research has shown for more than thirty years.

Brainstorming is the most widely used creative technique on the planet, and one of the worst performing. It's worth knowing why it fails, and what to run instead, without giving up the benefits of working together.

An ad agency invented it, not a lab

The word brainstorming didn't come from psychologists. It came from Alex Osborn, a co-founder of the New York advertising agency BBDO. That final "O" in the initials is him. Through the 1930s and 1940s he was hunting for ways to squeeze more campaign ideas out of a room full of copywriters, and over time he wrote down a set of rules for group idea generation.

He published them in Applied Imagination in 1953. You already know Osborn's rules even if you've never opened the book: withhold criticism, go for quantity, welcome wild ideas, build on what others say. The book became a bestseller and brainstorming spread into companies around the world.

Osborn promised a lot. He claimed the average person generates roughly twice as many ideas in a group as working alone. It was a seductive pitch: put five people in a room and get five times, maybe ten times, the creative firepower. The catch is that this was a marketing claim from an adman, and for a long time nobody actually measured it.

You'll invent more alone than in the pack

When researchers finally got hold of brainstorming, they tested it against so-called nominal groups. A nominal group is a group in name only: people work separately and independently, then their ideas are pooled and duplicates removed. No discussion, no inventing together at a table.

The result is consistent and uncomfortable. Nominal groups beat real brainstorming groups almost every time, on both the number of ideas and their quality. In 1987 the German researchers Michael Diehl and Wolfgang Stroebe summed this up in a study whose title framed it as a puzzle to be solved: why groups lose productivity in the first place.

Four years later came the meta-analysis. Brian Mullen, Craig Johnson, and Eduardo Salas (1991) gathered twenty existing studies and computed a combined result: brainstorming groups produce fewer ideas than nominal groups, and worse ones. The bigger the group, the steeper the drop. A team of six loses far more than a team of three.

Picture it concretely. Two people working apart barely interfere with each other, so the loss is trivial. At four it starts to show; at eight it's a chasm. A group behaves like a funnel with one narrow neck, where every person you add joins the queue for a chance to speak. Eight brains in a room don't give you eight times the ideas, more like eight people talking over each other for a single turn.

Restate that counterintuitive finding, because the whole piece rests on it. The problem isn't that group idea generation works less than perfectly. It actively lowers output compared to the same people working separately. So the question isn't "how do we brainstorm better," it's "why brainstorm out loud in one room at all."

Why a group throttles itself

Three mechanisms sit behind the drop. Diehl and Stroebe showed that the first one does most of the damage.

Only one person can talk (production blocking)

In any discussion, one person speaks at a time and the rest wait. During the wait, one of two things happens: you forget the idea you were holding, or you drop it because the conversation has moved somewhere else. And when you listen closely to others, you track your own thought less well. The authors named this production blocking as the chief culprit, responsible for most of the lost output.

Fear of judgment (evaluation apprehension)

Out loud, in front of colleagues and a boss, you only say what you feel safe saying. The half-formed, strange, slightly embarrassing ideas stay in your head. Yet those odd ones are often the seed of something original. Osborn's "no criticism" rule is meant to suppress this fear, but you can't switch off private self-censorship by banning it. One raised eyebrow from a colleague and you swallow the next idea.

Coasting on the group (social loafing)

The more people involved, the easier it is to hide. When the output is shared and your slice of it is invisible, the brain quietly eases off the gas. It isn't deliberate laziness, it happens on its own. Social psychology has known the effect since the old rope-pulling experiments, where individuals in a group pulled measurably weaker than when they pulled alone.

If it doesn't work, why do we keep doing it?

Here's an honest question for you: when did you last walk out of a brainstorm feeling it was a waste of time? Probably rarely. That's the whole trap.

Group brainstorming is fun. You sit with people, you laugh, ideas fly around the room, and at the end the flipchart is full and everyone feels the energy. That feeling is real. It just has little to do with how many usable ideas actually got made.

Researchers call this the illusion of group productivity. After a brainstorm, people repeatedly overestimate how much they came up with, and they claim credit for other people's ideas too. The group also feels like a team, shares responsibility, and nobody carries the risk alone. Those are genuine goods. Just don't confuse them with creative output.

So brainstorming isn't useless. It's good for building cohesion and aligning people around a topic. It's poor as the main engine for generating ideas. And that main-engine role is exactly where companies deploy it most.

Techniques that produce more

Most of the better methods rest on one trick: let people invent alone first, then combine. That single move removes both production blocking and the fear of judgment.

Notice what they share. They pull apart the two things that ordinary brainstorming glues together. First each person produces solo, and only then do the ideas meet. You get the volume of independent minds plus the cross-pollination, just in the reverse order we're used to.

Brainwriting and the 6-3-5 method

Instead of talking, people write. The German Bernd Rohrbach described the best-known version in 1968 as method 6-3-5: six people, each writes three ideas in five minutes, then passes the sheet to a neighbor who builds on what's already there. After six rounds you have well over a hundred ideas on paper, and nobody silenced anyone.

Take an ordinary product team that switched from the classic shout-at-the-board format to brainwriting. The reason was simple. Two of the quietest people on the team, Priya and Tom, consistently had the strongest ideas on paper, yet in a loud brainstorm they never said them aloud. Writing gave them the room that a talking group took away.

Electronic brainstorming for larger groups

The same idea, over a keyboard. People type ideas at the same time into a shared tool, anonymously and without waiting for a turn. In small groups the effect is weak, but the more people you add, the bigger the advantage, because it removes precisely what slows large groups down. Research on group support systems in the 1990s repeatedly found that electronic groups above roughly eight people outperform both talking groups and nominal ones.

Alternating solo and group phases

The most practical approach for a normal meeting. First a few minutes of silence while everyone writes alone. Only then do you share and build on each other's ideas as a group. The solo phase supplies volume and variety; the group phase filters and develops. That rhythm is the same one the ideation stage of design thinking uses, where private invention and shared synthesis take regular turns.

Brainstorm questions, not answers

A counterintuitive switch: instead of "what are the solutions?", set the group the task of inventing as many questions about the problem as it can. No answers, only questions. Hal Gregersen of MIT popularized the method under the name "question burst." Nobody defends or rates a question, so the fear of judgment falls away, and a well-aimed question often opens a direction that a direct hunt for solutions would have missed.

Let constraints do the work

A blank flipchart and the brief "come up with anything" freezes most people. A tight brief starts the engine instead. Try "solve it for zero dollars" or "what do we do if we're banned from using email." An artificial limit narrows the space enough that the brain stops wandering and starts combining. Well-judged constraints tend to yield more original ideas than total freedom does.

Technique	How it works	Group size	Watch out for
Brainwriting / 6-3-5	People write instead of talk; sheets circulate around the table	4-6 at one table	Needs quiet and someone to keep the rounds on time
Electronic brainstorming	Simultaneous anonymous typing into a shared tool	8 and up	The advantage barely shows in a small group
Hybrid (solo + group)	Silence and writing first, then sorting together	Practically any number	Nobody may skip the solo phase
Question burst	Five minutes of questions only, no solutions	3-8	The pull back toward answers

Which method suits you depends partly on how you think. Introverts and people who need quiet to warm up get the most out of the written techniques. If you want to know how you personally generate ideas, our creative thinking test can point to the stages of invention where you're strong and where collaboration will serve you better.

How to run ideation that doesn't waste time

Next time you run a meeting meant to produce something new, try this sequence instead of the usual flipchart:

Send the brief ahead of time so nobody arrives with an empty head.
The first five to ten minutes belong to silence, with everyone writing alone.
Collect the ideas anonymously, on sticky notes or in a shared doc with no names.
Discussion begins only now: the group sorts, combines, and develops the ideas.
Generation and evaluation must never blur into a single phase.
Don't invite more than six people into the room; productivity falls off above that.

Leave that flipchart from the opening on the wall if you like. Just don't pull it out until the group has something to sort, which is after everyone has spent a while inventing on their own. The order of the steps decides whether the meeting manufactures ideas or simply burns an hour.