Crowd Wisdom

I wanted to write a piece about the wisdom of crowds for a long time, but I needed a clip from “Monty Python and the Holy Grail” to make it work. After a lot of procrastinating, I went to YouTube, did the hard work of reviewing all of the Python clips and found the right one. There is no limit to the effort I will make to write a piece — this research was physically demanding and my sides still ache.

If you are a Python fan, you might want to review the clip for sheer fun, and even if wacky British humor is not your cup of tea, watching the clip will enhance your understanding of what I am going to say. So, go ahead, take a look. I will wait. Really, it’s OK — there’s a recession on and I have the time. It’s a little over four minutes.

OK, welcome back.

If you cheated and just forged ahead without the benefit of Michael Palin and the gang, the clip is a medieval mob scene in which the mob, convinced that a woman is a witch, asks permission from the local knight to burn the witch at the stake. The knight is skeptical of the claim. The mob has dressed the poor woman in a witch’s costume and given her a conical hat and fake carrot nose.

The whole scene is contrived, and at first you expect that through his tortured reasoning, the knight will convince the crowd that the woman is not a witch. I will leave it to you to watch the clip to find out how the woman is nonetheless condemned and the gods of comedy are served.

How Smart Is Your Crowd?

The point I’ve been making for a long time, which this clip illustrates, is that the wisdom of crowds is only as good as the crowd. That’s because you can’t get new information by asking what people already know. Those people — even the knight — believed in witches, so it was easy to proceed from that error to a burning at the stake. New information — and, hopefully, truth — comes from research, including setting up an experiment, collecting data, and then undertaking careful and dispassionate analysis of the data.

Monty Python’s mob simply illustrates that a crowd can pool its existing knowledge base to provide accurate feedback of group-think. It also cleverly illustrates Mark Twain’s brilliant observation that “It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.” At least by Twain’s time, we weren’t burning witches.

My point is that in our rush to embrace all things social, we run the risk of acting — with all good intentions — like that mob, and that’s especially unhelpful because no one has been given the knight’s job of playing devil’s advocate. I think we can take too much as fact just because it comes from a community of customers, especially if that community has no controls.

To be sure, there are valid and useful things to be gleaned from customer feedback, but feedback is only the first step. Ideally, feedback should give us some testable hypotheses from which we can extract true knowledge. All the testing in the world would not have helped the woman in the movie, but that’s the essence of comedy.

Feedback vs. Discovery

The film clip and the wisdom of crowds show the difference between feedback and discovery. We can run our companies pretty well using feedback most of the time, but unless there is a little discovery mixed in, and unless the lessons of the discovery process are put to use, we can find ourselves in trouble.

Last week, we saw an example. Chrysler got ready for a bankruptcy filing, and GM is not far behind. In their histories, both companies spent prodigious sums gathering customer feedback. It showed how much customers liked big cars (maybe that’s all they knew?) and the car companies used the information to justify building even bigger cars and trucks. I wonder if these companies bothered to ask what people felt about climate change or the high price of fuel. Since public perception and opinion can change much faster than a company can design and develop cars with higher fuel efficiency, it might have been good to know those things.

Customers can tell us an awful lot, but we have to ask the right questions, and we have to step out of our contemporary biases. Technology can do a lot to automate and speed up the asking process, but the human element is still important. We still have to ask the right questions and not flinch when we get the answers.