Harry Potter books sold well because lots of people bought them.
If that bit of analysis makes sense to you, you’ve been seduced by the siren of circular reasoning, says Duncan Watts, a Columbia University professor and principal scientist at Yahoo! Research. And that has caused more than one business to go bust.
Circular reasoning is the handmaiden of our belief in the ability of common sense and hindsight to explain the workings of the world. But common sense and hindsight have been derailing the best-laid plans of mice, men and businesses since time immemorial. While they give us a clear view of the past, they’re blind to the unknowns of the future. And those are the things that sneak up and bite you.
In his new book, “Everything Is Obvious: Once You Know the Answer” (Crown Business, 2011), Watts shows why common sense breaks down and why the simplest and most plausible answer usually isn’t the best.
Or as H.L. Mencken, a 20th-century writer and acerbic critic, once observed: “For every complex problem there is an answer that is clear, simple and wrong.”
“It’s easy to tell yourself a story,” Watts, whose book was published this month, told BusinessNewsDaily. “You have a story that sounds right and there’s nothing to contradict it.”
We — and our decision making — are prisoners of a narrative approach. “We’ve been trained that there’s going to be one answer and it’s going to be a complete answer,” Watts said. “We’re very uncomfortable with randomness as an explanation.”
We also want to believe in the powers of our intuition. But, Watts said, you should resist. “You should suspect your intuition,” he said.
The scientific approach, on the other hand, is a way to understand the world without having any intuition about it. And it begins with a kind of story — a hypothesis.
“The difference is, we test the stories and modify them when they don’t work,” he said. “Storytelling is a useful starting point. The real question is what we do next.”
Watts said the problem with common sense is that it works extremely well in concrete, immediate circumstances but doesn’t scale well.
“It’s extremely successful and reliable,” Watts said. “It works so well we are tempted to use it in all circumstances. Common-sense reasoning works very well when the same situation repeats itself over and over.”
But no two problems are the same, he said. And that’s where common-sense reasoning gets us into trouble.
Common sense often confuses correlation with causation, Watts said. Just because A happened and was followed by B doesn’t mean A caused B.
What we need, he said, is an uncommon sense that explores alternative outcomes and doesn’t believe there has to be only one correct strategy.
This kind of sense would be based on strategies and processes that don’t depend on your intuition being right, Watts said. You need to have a portfolio of strategies and be flexible in applying them. You need to test assumptions and express assumptions you didn’t know you were making.
The scientific method, with its emphasis on theory, observation and experimentation, may provide an answer. “Measure-and-react is one way to go about it,” he said. “Success depends on reacting fast , not on predicting the future. Try lots and lots of things all the time.”
Watts cites the success of Zara, a Spanish clothing retailer. Rather than trying to anticipate what shoppers will buy next season, the company observes what people are already wearing, creates a portfolio of styles, fabrics and colors, and tests them in small batches to see what sells and what doesn’t.
Using a very flexible manufacturing and distribution operation that can react quickly to information coming directly from consumers, Zara is able to design, produce, ship and sell a new garment anywhere in the world in just over two weeks.
Measure-and-react is particularly apparent — and relevant — in the online world, said Watts, where it’s known as “bucket testing.” Online designers use it to optimize ad placement, content selection, search results, recommendations, pricing and page layout on a site. When they want to change a design element, they’ll alter it for the version that a small number of users — the “bucket” —will see and measure their feedback and actual behavior. It’s a quick way to learn what works and what doesn’t.
Yahoo used bucket testing before it rolled out its new home page in 2009 and the tactic is now routinely used by major web companies such as Google and Microsoft.
Watts said plans fail because planners rely on their own common sense to reason about the behavior of people who are different from them.
“Winning in the marketplace is not successfully predicting what’s there,” he said. “There’s no such thing as latent demand. The market builds upon itself. Luck plays a strong role in success.”
- Historyâs Not Bunk, Itâs a Business
Reach BusinessNewsDaily senior writer Ned Smith at email@example.com. Follow him on Twitter @nedbsmith.