The Dilemma of Group Decision Making

I love it when a writer manages to combine a simple, powerful and relevant theory with a commonsense example that embodies the principle. Malcolm Gladwell does this regularly. Paul Krugman, recent Nobel prize winner in economics, does it as well.

Recently Bob Frisch has done something similar in his HBR article, “When Teams Can’t Decide.

Theory first. When a group must vote on a choice between three or more options, it is possible to get a circular, not a transitive, result. That is, choice A may result in a majority of people who would have preferred B over A; yet choice B may result in a majority who preferred C over B; and choice C could result in a majority who preferred A over C. Bye bye symbolic logic (If A>B and B>C then A>C); hello to legislatures, for one thing.

Worse yet, Frisch suggests, even a simple A vs. B decision often has a third option—do nothing.

Now for the practical application—this sounds an awful lot like corporate decision making, as well as politics. And the sub-optimal solution we have evolved has often been—the boss gets the tie-breaking vote. Which, given the circular possibility, makes “the decider” an often unpopular and unstable role.

So what’s the solution? Basically, redefine the problem. Which is often a brilliant idea—yet, equally often, unexplored. Frisch gives some specific rules, which seem “obvious” only in the theoretical rear view mirror. The truth is, they are infrequently observed.

One is to clearly articulate definitions. Don’t assume others understand an outcome to mean what you understand it to mean. (Clear definitions also ensures you’re all dealing with true outcomes, not indicators or milestones).

Another is what Frisch calls “testing fences and walls”–explicitly asking whether a given constraint is fundamental, or just presumed.

He also advocates transparency by getting instinctive opinions and biases out in the open sooner rather than later; one interesting approach is the formal use of a devil’s advocate role.

The purpose of all these techniques is to alternately expand and then contract the number of options so as to convert circular vote problem into more transitive patterns—before voting. In short, when you come to a problem—don’t vote on it until you’ve redefined it to improve the odds of majority acceptance.

Frisch positions this kind of organizational problem-solving as an alternative to “psychological” approaches like trust. Using his terms, I might redefine the “problem” of which alternative to choose until both approaches are seen to be integral parts of a differently-defined whole.

But whatever you call it, this is a nice piece of theory-meets-commonsense. It’s the kind of thing that made HBR special back in the day; I’m glad to see it’s still alive and kicking.