Behavioral New World
June 1, 2025
Thoughts on group decision-making
As a first approximation, we can think of decisions as either individual or group. Many of the biases I’ve written about over the past 5+ years influence our individual decisions, e.g., how confirmation bias (link) impedes the revision of our cherished beliefs.
Other decisions happen in a group setting, such as with a board of directors, a jury, or a group of friends planning a vacation. Here, I consider two factors that influence the quality of a group’s decisions—group size and the potentially pernicious effects of groupthink. (Yes, I went for the alliteration there.)
The size of the group. Is the size of the group determined, as with a jury? If not, consider the Ringlemann effect (link), which says that each member of a group puts in less effort as the number of members increases. Perhaps that’s not surprising. But more importantly, Ringlemann also found that, as group size grows, the group often becomes less efficient.
This phenomenon suggests that there is an optimal size for group decision-making. Some research (link) suggests that the optimal group size is 4.6 (Ouch! Let’s call that five). Other research[1] suggests that seven is the upper limit for good group decision-making. Beyond that point, there is a demonstrable decline in effectiveness. Unless compelled to, groups of 17 or more rarely make decisions (“Let’s table it.”)
I once served on a strategic planning committee for the business school where I taught. There were at least 40 people on the committee, probably because the dean wanted all stakeholders to be represented. I’m pretty sure I didn’t say anything and was not called on to say anything; there were others in the same boat. Perhaps subcommittees of 5-7 members focused on individual aspects of the strategic plan would have been more efficient.
Groupthink. As the name suggests, there is a tendency for the views of group members to converge. This is not necessarily a bad thing—if there is an obviously best decision, for example, convergence will ensure that it will be adopted and will minimize the time required to come to the decision. And after the fact, broad agreement helps promote unified support for the implementation of the decision.
But groupthink might also occur because members don’t want to offend the boss or coworkers. They might be reluctant to point out a weakness in a line of reasoning or an overlooked consequence of a particular decision. As a result, salient input is missed.
The reluctance to speak can arise unconsciously because of an uneasiness about standing out from the apparent majority. “The nail that sticks out gets hammered down,” as the saying goes.[2] Whoever is leading the meeting should make people comfortable speaking up with contrary views.
Groupthink is similar to herd behavior. See my newsletter about the similarities and differences. (link)
Having chosen the correct group size and banished groupthink, decision-making can be further improved by employing two related techniques.
Use of a devil’s advocate. The American Heritage Dictionary defines a devil’s advocate as “One who argues against a cause or position, not as a committed opponent but simply for the sake of argument or to determine the validity of the cause or position.”
It is important that all group members understand that the devil’s advocate doesn’t necessarily believe the argument that they are making. Sometimes hostility is directed to that person—they’re slowing up the decision, they’re muddying the waters, they’re keeping me from my tee time, etc. For this reason, people are often reluctant to serve as a devil’s advocate. An alternative is to use “red teaming,” in which there is a team of devil’s advocates.
A related tactic: the pre-mortem. As the name suggests, this tactic takes place before the final decision is accepted. The group is asked, “Let’s pretend that the decision we’re talking about has been implemented. Now imagine that some time has passed, and the decision turns out not to have been a good one. What went wrong?”
This question can identify risks that have been overlooked. It need not change the decision—all decisions have risks of some sort—but will allow the group to examine how to mitigate any newly identified risks, creating more confidence in the decision.
Of course, for these techniques to be used effectively, buy-in is needed; group members must agree to them in advance, which is itself a group decision. Is there a chicken-and-egg problem here? What are good rules for setting good rules? Readers, your thoughts? What have been your experiences, good or bad, with group decision-making?
[1] Decide and Deliver: Five Steps to Breakthrough Performance in Your Organization by Marcia Blenko, Michael C. Mankins, and Paul Rogers, Harvard Business Review Press, 2010.
[2] More detail here:
https://crossidiomas.com/the-nail-that-sticks-out-gets-hammered-down/
I hadn't heard of Brook's Law, but it make sense!
"Ringlemann effect (link), which says that each member of a group puts in less effort as the number of members increases," I didn't know about that one, but there is also Brook's law, which says that the more team members you add to a project, the longer it will take. Attributed mainly to the increasing number of communications channels.