When everyone agrees, find out who spoke first

Five people agreeing on an investment can sound like five pieces of evidence. It matters whether all five examined the investment independently. If four heard the first person's conclusion before forming their own, the agreement may tell us more about the discussion than about the asset.

That distinction is easy to lose in a hurried meeting. The conversation moves forward, objections disappear, and a recommendation becomes the team's view. By the time someone asks why there is so much confidence, the number of people repeating the conclusion can itself become part of the answer.

Following others is often sensible. They may have expertise or information you lack. A fund manager's trade, a colleague's recommendation, or a competitor's decision can contain information worth considering. The difficulty is working out how much independent knowledge stands behind the behavior you can see.

Sushil Bikhchandani, David Hirshleifer, and Ivo Welch examined this problem in their 1992 theory of informational cascades. Their model showed how people observing earlier decisions could rationally follow those decisions while setting aside their own information. Under those conditions, conformity can spread and useful private signals can stop reaching the group. A theoretical mechanism establishes a possibility; it does not diagnose every crowded trade. Informational cascades ↗

The model offers a warning for organizations that seek diverse perspectives and then collect them in the wrong order. Hiring people with different experience will not produce independent judgment if the meeting gives everyone the preferred answer before asking what they think.

A small change could help. For a consequential decision, ask participants to record an initial view before they see colleagues' recommendations. Keep it brief enough to be practical. Then use the discussion to examine disagreements and share information. People should be free to change their minds, but the team should retain enough of the initial record to see what produced the change.

Suppose three analysts begin with reservations about a proposal. After a discussion, all support it. That might be an excellent result if a colleague supplied information that answered their concerns. It is a different result if the concerns vanished without being addressed. A final unanimous vote cannot distinguish between those histories. An initial record can.

Expertise becomes useful when people share it, so this procedure should leave plenty of room for collaboration. Teams also need to coordinate, and not every decision merits a formal record. Preserve independent information long enough for the group to examine it. After that, convergence may be exactly what the evidence warrants.

Disagreement deserves scrutiny too. A person can mistake contrarianism for analysis and oppose a sound recommendation merely because it is popular. The dissenter should explain what others have missed and what evidence supports that view. Independence is valuable because it can add information. It has little value when it becomes a pose.

The same caution applies when investors study the market. A widely held position can reflect shared fundamentals, a common mandate, or information that genuinely points in one direction. It can also reflect imitation. Popularity alone does not identify which explanation is correct. A serious analysis asks how the consensus formed and whether the apparent sources are independent.

Pressure makes this inquiry especially worth protecting. When time is short, counting endorsements is easier than tracing their origins. A team should decide in advance which decisions require that extra work, instead of trying to invent a procedure after the room has already converged.

At the next consequential meeting, collect the initial views before the discussion starts. If the eventual consensus survives an examination of how it formed, the organization will have a better reason to trust it.