I’ll Be Brief

Dubner’s post on perfectionism reminds me of a parallel phenomenon: People who say that they’re going to be brief often aren’t.

Indeed, the very time taken to say that you are going to be brief works to negate the claim. You couldn’t honestly begin by saying, “I’m only going to say the words that are absolutely necessary.”

Wouldn’t it better to forgo this temporal throat clearing entirely? Or if you have to say it, why not say it at the end: “I have a lot more to say, but I’m stopping now because I want to be brief.”

Of course, one of the strongest reasons for starting with this statement of intent is as a soft form of commitment. I’m giving the audience permission to cast aspersions toward me if I speak for too long.

As a co-founder of stickK.com (naked self-promotion), I’m a big fan of commitment devices. But “I’m going to be brief” is just cheap talk. The consequences of speaking too long are so wimpy that they don’t seem to constrain.

One of the problems is that people underestimate how long they speak.

It would be useful at Quaker meetings (or other places where people share the floor) to have speakers immediately, upon finishing an oration, estimate how long they spoke. I bet initially many people would report that they speak fewer minutes than they actually do.

Or (like asking people if they think they are better or worse than the average driver), you could ask them if they think they generally speak longer or shorter than the average meeting speaker. I’m guessing most people believe they are more concise than their average neighbor.

Better yet, groups might develop a norm to have speakers publicly estimate how many minutes they intend to speak before they begin: Instead of saying, “I’m going to be brief,” it would be a stronger commitment to say to the moderator, “Please interrupt me if I speak more than X minutes.”

This is a great example of what Thaler and Sunstein call libertarian paternalism in their great book, Nudge. The individual speakers retain the freedom to speak as long as they want, and this new verbal convention would give all of us a credible commitment device. The problem of speakers droning on at conferences and meetings isn’t one of the biggest problems in the world — but it is an example where cognitive error leads to a persistent dysfunction.

An analogous dysfunction concerns follow-up questions. Here’s a simple test:

How often do you find it useful to hear an initial questioner ask a “follow-up” question after hearing the initial response? For me, it’s less than 10 percent of the time. That’s the percentage of the time when I find that the follow-up is more useful than the new question that it squeezes out.

Follow ups aren’t usually helpful because the restated question normally doesn’t pull any more information out of the target. But people think that their follow-ups are more successful than the low-value follow-ups of their colleagues. Indeed, one of the reasons I rarely (but I probably underestimate my rate) follow-up is that I trust other people in the audience to pick up my point if it is really important (and if it was inadequately answered).

A simple solution to the follow-up problem is to push for a norm that only allows follow-ups if there are no new question waiting to be asked. If you’re not crowding out a new question, follow up to your heart’s content.

And here’s a final “why not” idea for better time-management at meetings. Let the last speaker on the panel (instead of the moderator) enforce the time allocation of the previous speakers. Too many moderators are wimps and let the initial speakers speak too long so that the last speaker is cheated. The last speaker has a lot better incentive to keep the train running on time.