The best decision in the room usually had someone arguing against it

Arguing with your boss might be the best thing that happens to you this week. Even if you are wrong. Especially if you are wrong.

Charlan Nemeth at Berkeley spent forty years proving a version of that. A single voice of disagreement makes the rest of the group think harder, look wider, and find answers they would otherwise have walked straight past. The dissent does not have to be right. It just has to be real.

Which is awkward, because most leaders quietly build teams engineered to make sure it never happens.

Why disagreement works even when it is wrong

Agreement feels like progress. The meeting moves, the decision lands, everyone leaves aligned. It is efficient in roughly the way that skipping the brake check gets you out of the garage faster.

Nemeth's experiments show what that smoothness costs. Groups exposed to a dissenting view look at a problem from more angles and find more correct solutions. Groups marching towards consensus look at it from one angle, the majority's, and stop early. The dissenter is rarely valuable for the answer they bring. They are valuable for the work they force everyone else to do.

The instinct that gets worse with power

There is a reason leaders, specifically, struggle with this. Power hands you confidence whether or not you have earned the accuracy to go with it.

Kelly See and colleagues ran four studies on power and advice. The more powerful people felt, the more confident they became, the less they took others' input, and in the end the less accurate their judgements were. Leigh Tost, Francesca Gino and Richard Larrick found the same reflex extended even to expert advice. Power does not just make you less likely to listen. It makes you less able to tell whose advice was worth taking.

So the person with the most authority in the room is often the most certain and the least accurate, and the people around them have usually worked that out and gone quiet.

Different perspectives, not automatic agreement

Demographic and cognitive diversity aren't interchangeable; neither guarantees the other. Building a team means paying attention to both.

Alison Reynolds and David Lewis ran a strategy exercise more than a hundred times and scored teams on cognitive diversity, the range of how people process information and approach problems. The cognitively diverse teams solved it faster. They found no such link with gender, age or ethnicity in the teams they measured. It is a small study, six teams, so treat it as suggestive rather than settled. But it names the right distinction.

Ask how people reached their answer. Different experiences can still lead to agreement; the useful test is whether the team can examine its assumptions.

Why it does not happen on its own

None of this is the default. The default is the opposite.

Homophily, the best-evidenced pattern in social networks, means we hire and promote people who resemble us. McPherson, Smith-Lovin and Cook documented it across every kind of relationship. A leader building a team on instinct will tend to build an agreeable one, then mistake the comfort for alignment. Nobody sets out to build an echo chamber. We just hire the people we click with, and clicking is mostly recognition.

Engineering disagreement, and the part most people get wrong

So you have to build it in deliberately. This is where good intentions usually go wrong.

The common move is to appoint a devil's advocate. The evidence says it mostly backfires. Nemeth, with Connson Brown and John Rogers, found that a role-played objection makes people defend their original position harder, not reconsider it. Everyone knows the devil's advocate does not mean it. It is theatre, and people can smell theatre from across a boardroom. Authentic dissent, from someone who actually holds the view, does the real work.

What does help is structure. Charles Schwenk pooled the experiments and found that formal methods like dialectical inquiry produced better decisions than consensus, though, tellingly, people enjoyed them less and felt less like a team afterwards. Gary Klein's pre-mortem is the cheapest version. Before committing, you assume the decision has already failed a year from now, and ask everyone to explain why. It gives the doubts a sanctioned way out.

The deeper fix is who you hire. People who will genuinely disagree with you, and the protection that lets them keep doing it.

The uncomfortable bit

The discomfort is the point, and it is also why most leaders quietly avoid it.

Structured disagreement improves decisions and feels worse while it is happening. The smooth meeting where everyone nods is the one that should worry you, not the one where someone keeps asking the question nobody wants to answer.

Most leaders say they want to be challenged. Far fewer build the conditions that would let it actually happen, and fewer still keep the person who does it.

So here is the question worth sitting with. When did someone last change your mind in a meeting? If it has been a while, is that because you are usually right, or because the room has learned not to try?

One of those is far more likely than the other.

I help boards and founders build leadership teams that can disagree well. Start a conversation

FAQ

Does disagreement really improve decisions?

Charlan Nemeth's decades of experiments show a dissenting voice makes groups consider more options and find more correct solutions, even when the dissenter is wrong, because it forces deeper thinking.

Is appointing a devil's advocate a good idea?

Often not. Nemeth, Brown and Rogers (2001) found role-played dissent can make people defend their original view harder. Authentic disagreement, from someone who genuinely holds the view, works better.

What is cognitive diversity?

The range of ways people process information and approach problems, which is distinct from demographic diversity. Reynolds and Lewis (HBR, 2017) found cognitive diversity predicted faster problem-solving, though it was a small sample.

Why do senior leaders struggle with this most?

See and colleagues (2011, four studies) found power raises confidence and lowers both advice-taking and accuracy, so the most senior person is often the most certain and the least challenged.

Sources

• Nemeth, C. (2018). In Defense of Troublemakers: The Power of Dissent in Life and Business. Basic Books.
• See, K.E., Morrison, E.W., Rothman, N.B., & Soll, J.B. (2011). The detrimental effects of power on confidence, advice taking, and accuracy. OBHDP, 116(2).
• Tost, L.P., Gino, F., & Larrick, R.P. (2012). Power, competitiveness, and advice taking. OBHDP, 117(1).
• Reynolds, A., & Lewis, D. (2017). Teams Solve Problems Faster When They're More Cognitively Diverse. Harvard Business Review. (six teams)
• Nemeth, C., Brown, K., & Rogers, J. (2001). Devil's advocate versus authentic dissent. European Journal of Social Psychology, 31.
• Schwenk, C.R. (1990). Effects of devil's advocacy and dialectical inquiry on decision making. OBHDP, 47(1).
• Klein, G. (2007). Performing a Project Premortem. Harvard Business Review.
• McPherson, M., Smith-Lovin, L., & Cook, J.M. (2001). Birds of a Feather: Homophily in Social Networks. Annual Review of Sociology, 27.