Competence, confidence, and being willing to admit what you don't know

As mentors to entrepreneurs and startups, we have our own sets of hard-won battle scars, and have experienced a lot of sympathetic pain as entrepreneurs we’ve advised have earned their own. From that experience, two ironclad rules have become clear:

1. Know not only what you’re building, but why you’re building it, and whom you’re building it for, before you build anything.
2. You can never know your customers well enough.

The most common ‘failure mode’ we see are inventors who, in the absence of evidence other than their own belief, persist in building prototype products and are baffled when the world doesn’t beat a path to their door. Many, when delivered this news, go back into the lab or workshop to tinker some more, in the hope of turning their product’s beloved features up to 11. As Stephen Colbert said in his White House Correspondents Dinner monologue – these people are steady; they believe on Wednesday what they believed on Monday, regardless of what happened on Tuesday.

It is with that in mind that this article from Cornell’s David Dunning rang so true – success comes from the ability to differentiate between what you know and what you believe. This is not a trait that comes naturally; we are wired in our subconscious to ascribe purpose to objects and events; and we are born with a bias to overconfidence which allows us to get in the car each morning safe in the feeling that we are part of the 93% that believe they have above-average driving skill.

Fortunately, we see more and more budding entrepreneurs embracing the principles of ‘customer development’. We were tremendously gratified to hear from a computer science student who reported that he described his business concept to his fellow students and was deeply disappointed that nobody asked him why he was building his product. He was justifiably proud of the work he did to gather evidence of what his customers wanted him to build – and is hoping that his classmates can make sure that they, too, build solutions that customers will demand of them, rather than products that need to be sold.

Dunning’s work suggests a few other key lessons for counteracting overconfidence in decision-making:

a. In a group, appoint a devil’s advocate to challenge arguments and differentiate between fact and opinion. By making this role assigned and explicit, it gives the person an excuse to behave in a way that in other circumstances might feel conflict-loving or annoying.
b. In absence of a group, be your own devil’s advocate: using what Keats called ‘negative capability’, consider alternate realities and test them for their likelihood; Dunning suggests imagining a situation where your projections fail, and guess as to what was most likely to be the cause to identify areas deserving of further analysis.
c. Talk openly and often about not just what you believe, but why you believe it – enlist friends and advisors to test and challenge your conclusions based on facts.

I would add two more:

d. If, pretty early in your interactions with an advisor or mentor, they don’t admit at least once that they don’t know something, find a different mentor or advisor.
e. In a team environment, recognize the implicit social rules that get created by what gets celebrated – and what doesn’t. Don’t admire the people who always have an answer, and don’t denigrate those that admit ignorance. The big heroes that rarely get the adulation that they deserve are the people who say, “I don’t know” – and then go and find out.

If you’re interested in further reading, here’s the paper that started it all – what has become known as the Dunning-Kruger effect, that people with the lowest levels of competence self-identified as highly confident, and that people with the highest levels of competence self-identified as mediocre. (A great journal article, almost as good as this seasonal classic, Waldfogel’s The Deadweight Loss of Christmas.)