Groupthink: Why Smart Teams Make Bad Decisions (and How to Stop It)

Groupthink is what happens when a team’s desire to agree quietly overrides its ability to think — people stop voicing doubts, alternatives go unexamined, and the group talks itself into a decision nobody would have made alone. The way out isn’t less teamwork; it’s building dissent back into how the group decides: let the leader stay quiet until last, assign someone to argue the other side, gather views independently before the room converges, and make it genuinely safe to disagree. Good group decisions come from structured disagreement, not smooth consensus.

The psychologist Irving Janis coined the term in 1972, defining it as a mode of thinking where a cohesive group’s striving for unanimity overrides any realistic appraisal of the alternatives. He traced it through fiascos like the Bay of Pigs invasion and, later, the Challenger launch — where engineers reportedly knew of faulty parts but the decision to fly went ahead anyway. None of those failures came from stupidity; they came from smart people not wanting to break the consensus. Here are seven ways to keep that from happening in your team.

Seven ways to prevent groupthink

Each of these targets a different mechanism Janis identified — premature agreement, suppressed doubt, or insulation from outside views.

1. Have the leader speak last

When the most senior person states their preference early, everyone else quietly anchors to it, and the discussion becomes a search for agreement rather than truth. Janis’s first recommendation was for leaders to withhold their own opinion at the start and explicitly invite criticism. If you run the meeting, ask the questions, hear the room out, and save your view for the end — what you gain in honest input is worth far more than the comfort of leading with your answer.

2. Assign a devil’s advocate

Give one or two people the explicit job of arguing against the emerging decision — not to be difficult, but to stress-test it for blind spots and downsides. Making it a formal, rotating role matters: it frees the dissenter from looking disloyal, because they’re “just playing the part.” This was one of Janis’s core fixes, and it works precisely because it normalizes the objection that self-censorship would otherwise bury.

3. Gather views independently before discussing

A lot of groupthink is premature convergence — the first confident opinion sets the tone and everyone files in behind it. Counter it by collecting people’s honest takes before the group talks: a quick written round, a silent vote, or one-on-one input. When views are formed independently first, you surface the real spread of opinion instead of a roomful of people agreeing with whoever spoke first.

4. Bring in an outsider’s eyes

Groups that only talk to themselves drift, because they share the same assumptions and blind spots. Janis recommended inviting outside experts or fresh observers into the discussion to challenge the group’s thinking. Someone with no stake in the in-group’s harmony will ask the obvious question everyone inside has learned not to. Even one outsider in the room changes what feels sayable.

5. Split into subgroups and compare

For a big decision, have two smaller groups work the same problem separately, then bring their conclusions together. Janis suggested exactly this, because parallel groups develop different framings and catch each other’s gaps — and where they disagree is usually where the real risk lives. It’s a simple structural way to manufacture the diversity of thought that a single cohesive group suppresses on its own.

6. Make it genuinely safe to dissent

Techniques only work if people believe speaking up won’t cost them. Self-censorship — Janis’s term for swallowing your doubts to keep the peace — disappears only when disagreement is visibly welcomed and never punished. Thank the person who raises the awkward point, ask the quiet members directly what they think, and watch your own reaction to bad news. If you want a read on how readily you speak up when the room is leaning the other way, that’s a habit worth measuring honestly.

7. Build in a second-chance meeting

Before you lock in a major decision, hold one more session whose only purpose is to surface remaining doubts. Janis called for giving members a chance to reconsider once the first flush of agreement has cooled. A night’s distance often loosens the illusion of unanimity enough for someone to say the thing they didn’t say the first time — and a decision that survives a deliberate second look is far stronger than one rushed through on momentum.

The skills underneath better group decisions

Notice how little of this was about the group and how much was about what individuals do inside it. Preventing groupthink comes down to a few underlying, learnable skills.

Decision-Making is the most direct, and the framework’s advice on deciding with others reads almost like an anti-groupthink checklist: recruit differing opinions, gather input independently first, play devil’s advocate, call a time-out if you need one, ask the quiet members what they think, and split into subgroups for divergent thinking. Good group judgment is a process you build, not a mood you hope for.

Teamwork is what makes that process survivable, because dissent without trust just becomes conflict. The framework treats disagreement as a normal, even necessary part of working together — engage the hard conversations, stay on the topic without making it personal, and once the group genuinely decides, disagree and commit. That combination is what lets a team argue well and still pull together afterward.

Building Confidence is what it takes to be the one who breaks the false consensus. Speaking up in a meeting when everyone seems to agree is a real act of nerve, and the framework treats that confidence as something built by doing — acting despite the discomfort, focusing on the next contribution rather than the fear of looking difficult. The lone sensible objection often depends entirely on one person being willing to voice it.

Those three are part of a wider set of twelve work skills the framework treats as learnable rather than fixed, and the test maps where each of yours stands — useful, because whether you prevent groupthink or get swept up in it usually traces to where your real gaps are across exactly these.

What this means for you

You might already do some of this without naming it — asking the quiet person what they think, holding your view so others speak first, being the one who says “what are we missing?” That’s worth building on, because steering a group toward honest decisions is a learnable habit, not a personality trait, and you can grow it while staying entirely yourself. And it matters more as you rise: the more decisions you influence, the more a single unchallenged consensus can cost. By learning to spot groupthink, you’re already doing what most people in the room won’t.

See where you stand in a room full of agreement

You know the fixes now; the only thing left is an honest read on the skills that let you use them under real pressure. The free Work Skills Test is a short self-assessment that shows where you stand across all twelve work skills — including the decision-making, teamwork, and confidence habits that good group decisions depend on — and points you to the one worth strengthening first.

Take the test

Free, and it takes about 7 minutes.