How to Run a Brainstorming Session That Actually Works

In 1953, advertising executive Alex Osborn promised that a group brainstorming together could think up twice as many ideas as the same people working alone. Five years later, Yale ran the first controlled test of that claim and found the opposite: people working on their own produced roughly twice as many good ideas as people brainstorming together. Here is how to run a brainstorming session that respects that result — because the meeting is still worth having, just not the way most teams run it.

The experiment that ruined brainstorming

In 1958, Taylor, Berry and Block at Yale ran 12 real groups of four through three problems using Osborn's official rules: no criticism, freewheeling, quantity first, build on each other's ideas. They also had 48 other people work alone, then randomly recombined those individuals into "nominal" groups of four and scored them as if they had worked together. The nominal group is exactly what the real group would have produced if talking to each other were neutral.

It wasn't neutral. The real groups were, in the authors' words, "markedly inferior" on total ideas, unique ideas, and three quality-weighted measures. The people who never spoke to each other produced roughly twice as many ideas.

The part that stings is the precision of the reversal. Osborn had written that "the average person can think up twice as many ideas when working with a group than when working alone." The first proper test found the effect going the other direction by about the same factor. One honest caveat: both conditions used Osborn's rules, so the study tested group versus individual under brainstorming rules — not brainstorming versus no method at all. The enemy was the group, not the rules.

Why talking first loses: production blocking

Three candidates were tested: social loafing (people coasting), evaluation apprehension (fear of sounding stupid), and production blocking — only one person can talk at a time, so everyone else waits, and while they wait their ideas decay. You forget the idea, or you spend your mental energy tracking the speaker instead of thinking.

Diehl and Stroebe (1987, Journal of Personality and Social Psychology) manipulated all three mechanisms directly and found that blocking accounted for "most of the productivity loss" of real brainstorming groups. A 1991 follow-up with 400 students confirmed it, and pinned down something sharper: the damage comes from cognitive interference during the delay between generating an idea and getting to say it, not from limited speaking time itself. The waiting is the poison.

Then the meta-analysis made it undeniable. Mullen, Johnson and Salas (1991) pooled 20 studies from 1958 to 1990: nominal groups produce substantially more nonredundant ideas and substantially more high-quality ideas than interactive groups — a large effect (r around .56–.57 on both counts). And the deficit grows with group size. Your eight-person war room is worse than two four-person ones.

Why it feels like it works anyway

This is the objection every manager raises, and it deserves an honest answer: if brainstorming is so bad, why does every company still do it?

Part of the answer is a documented self-deception. Stroebe, Diehl and Abakoumkin (1992) showed that group members rate their own brainstorming as more effective than it was — the "illusion of group effectivity." Group brainstorms are enjoyable, social, and loud. They feel creative. So we conclude they were.

The deeper answer comes from Sutton and Hargadon (1996), who spent a field study inside IDEO, then the largest product-design firm in the US. Judged purely on ideas per person, IDEO's brainstorms were no better than the lab said. But they served at least six functions the labs never measured: organizational memory, designer skill-building, an "attitude of wisdom" (acting with knowledge while doubting it), status competition, impressing clients who watched and joined — and, as they put it, fun. "Nearly everyone at IDEO likes them, and they feel their jobs are more interesting because of them."

So the ritual does real social work. The correct response is not to cancel the meeting. It is to keep the meeting and fix the mechanics.

What the evidence says actually works

The fix has a name: the hybrid approach. Write alone first, talk second. Girotra, Terwiesch and Ulrich (2010, Management Science) ran 4-person groups in both orders: the hybrid process generated about three times as many ideas, with average quality roughly 30 percentile points higher — "the difference between the 1st and the 30th idea in a pool of 100." And then the knife twist: Osborn's beloved "build on each other's ideas" rule was measured as counterproductive. Teams that did more building produced neither more ideas nor better ones.

The oldest working version of this is brainwriting 6-3-5, published by Bernd Rohrbach in 1968. Six people, each writes 3 ideas every 5 minutes, worksheets rotate through 6 rounds: 108 ideas in 30 minutes. Silent simultaneous writing kills production blocking and verbal dominance in one move — and everyone sees everyone else's written ideas without paying the turn-taking tax.

One more data point: when groups contribute simultaneously by typing instead of speaking, the verdict flips. Electronic brainstorming groups have been measured beating nominal groups — simultaneous typing removes blocking entirely and the synergy researchers kept hoping for finally shows up. (A caveat from the video era: camera-on video groups underperform face-to-face ones on fluency, which is why "write before you talk" is the design principle.)

And it is the best idea that counts, not the pile. Kornish and Ulrich (2014, Journal of Marketing Research), using real sales data from crowd-designed products, found that raw ideas one standard deviation better translated to roughly 50% higher sales. Organizations don't need more ideas. They need the ceiling raised.

The 60-minute agenda that actually works

Here is the session, end to end, with the research baked in. No special tools required.

0:00 — State the question as one sentence. "How might we improve onboarding?" is too broad. "How might we cut the time from signup to first value from five days to one?" is a brainstormable question. Write it where everyone can see it, and don't start until it's sharp.

0:05 — Silent writing, 15 minutes. Everyone writes alone. One idea per sticky note or card, no talking, no laptops, no peeking. This is the entire game — the science says this phase is where the ideas come from. Push for quantity; ten per person is a good floor.

0:20 — Round-robin read-out, 15 minutes. Each person reads their ideas out, one at a time, with no commentary or defense. A scribe clusters them into themes as they emerge. The rule is simple: evaluate nothing, capture everything.

0:35 — Silent dot-vote, 15 minutes. Everyone gets three votes (anonymous if the room has hierarchy) across two criteria: impact and feasibility. Voting creates a shortlist, not a verdict — but five survivors is about right.

0:50 — Convergence, 10 minutes. For each surviving idea, name an owner, one deliverable, and one deadline. An idea with no owner is entertainment, not work. And once the meeting ends, the notes still matter — write them for the person who wasn't there, or the decisions you just made won't survive the week.

Three rules that hold the whole thing together: no devices during the session, the same single question for everyone, and the meeting ends with owners or it failed.

Next week, change one thing about your next brainstorm: fifteen silent minutes before anyone speaks. The research has been sitting there since 1958. The only teams that benefit from it are the ones that rearrange the chairs.

References

• Taylor, D. W., Berry, P. C., & Block, C. H. (1958). Does group participation when using brainstorming facilitate or inhibit creative thinking? Administrative Science Quarterly, 3(1), 23–47.
• Diehl, M., & Stroebe, W. (1987). Productivity loss in brainstorming groups: Toward the solution of a riddle. Journal of Personality and Social Psychology, 53(3), 497–509.
• Diehl, M., & Stroebe, W. (1991). Productivity loss in idea-generating groups: Tracking down the blocking effect. Journal of Personality and Social Psychology, 61(3), 392–403.
• Mullen, B., Johnson, C., & Salas, E. (1991). Productivity loss in brainstorming groups: A meta-analytic integration. Basic and Applied Social Psychology, 12(1), 3–24.
• Girotra, K., Terwiesch, C., & Ulrich, K. T. (2010). Idea generation and the quality of the best idea. Management Science, 56(4), 591–605.
• Sutton, R. I., & Hargadon, A. (1996). Brainstorming groups in context: Effectiveness in a product design firm. Administrative Science Quarterly, 41(4), 685–718.
• Rohrbach, B. (1968). Kreativ nach Regeln — Methode 635, eine neue Technik zum Lösen von Problemen. Absatzwirtschaft, 12, 73–75.
• Kornish, L. J., & Ulrich, K. T. (2014). The importance of the raw idea in innovation: Testing the sow's ear hypothesis. Journal of Marketing Research, 51(1).
• What is Brainstorming? Techniques, Tools, and Frameworks — Storyflow (2026 guide, framing section)
• The Simple Secret to Running a Productive Brainstorm — The Heretic (silent-ideation opener)