Here's one of the most unsettling findings in decision science:

A team of smart people will routinely make a decision dumber than the smartest person in the room would have made alone.

You've probably watched it happen. A group of capable, well-meaning people walks into a room, and somehow walks out with a choice that, in hindsight, nobody actually thought was the best one. How?

The four ways groups sabotage themselves

Unstructured group decisions fail in predictable patterns:

• ✕Groupthink — the pull toward agreement overpowers the search for the right answer. Dissent feels disloyal, so it stays quiet.
• ✕The loudest-voice problem — the most confident or talkative person dominates, and confidence gets mistaken for correctness.
• ✕The HiPPO effect — the Highest Paid Person's Opinion anchors the room. Once the boss leans, everyone "discovers" they agree.
• ✕Conformity cascades — the first opinion voiced shapes the second, which shapes the third, until "consensus" is really just an echo of whoever spoke first.

Notice the common thread: in every case, the best argument loses to something irrelevant — volume, status, timing, or the sheer discomfort of disagreement.

But groups aren't the problem

Here's the twist that changes everything.

Groups aren't inherently bad at deciding. Unstructured groups are.

Under the right conditions, a group will reliably outperform even its smartest member. It's called the wisdom of crowds — and it's not a new idea. Back in 1785, the mathematician Condorcet proved it formally: as long as each person is more likely right than wrong, a group's majority gets more accurate as it grows.

But — and this is the entire game — only when three conditions hold:

• ✓Independence — people form a view before being anchored by the loudest or most senior voice.
• ✓Diversity — genuinely different perspectives are in the mix, not five versions of the same person.
• ✓Structure — there's a method for surfacing and weighing the arguments, instead of just talking until someone gives up.

Take those away and the crowd gets dumber than an individual. Add them and it gets smarter than anyone. Same people. Opposite outcome. The only variable is structure.

The one thing that reverses it

So what's the single lever?

Make the arguments visible, and judge them on merit instead of on who said them.

That's it. When the actual reasons — for and against — are laid out where everyone can see and weigh them, the irrelevant stuff loses its grip. Status stops standing in for correctness. The quiet person's strong point can't be drowned out. The first opinion doesn't get to anchor the rest. And six months later, you can see why the call was made.

It's the difference between "we talked about it and went with Dave's gut" and "we put the arguments side by side and the strongest one won."

This is exactly why structured, collaborative decision-making beats a room full of smart people winging it. The Challenger engineers were right — they just had no structure that could make "right" win over "convenient."

The lesson isn't "trust groups" or "distrust groups." It's sharper than that: a group is only as smart as the structure you give it. Give it none, and brilliance cancels out into mush. Give it the right structure, and ordinary people make extraordinary decisions together.

So: think of the best group decision you've ever been part of. What was different about how it was made — not who was in the room, but the process?