What’s wrong with what is that it’s not how

Even if you know what that means, knowing is not always enough.

My first gainful employment was with a small management-consulting firm writing marketing material and articles on quality management. As I churned out explanations of topics such as the importance of reducing variance in output, Gerry Michaelson, an éminence grise in the firm, would read my missives and offer various suggestions for improvement. He always ended with, “It’s not enough to tell people what to do. You have to tell them how to do it if you’re really going to make a difference.”

This seemed like good advice to me, and I’ve tried to apply it ever since, despite the fact that “getting to how” can be extraordinarily difficult.

Clayton Christensen is a Harvard Business School professor of no small reputation, and author of The Innovator’s Dilemma, the book that popularized Clayton’s discovery of “disruptive innovation” (whereby a product or service enters at the bottom of a market before eventually moving up to displace competitors). Clayton has built on that concept an edifice of research and consulting that has spawned significant new growth at hundreds of companies.

Part of the reason for disruption theory’s success, I had come to believe, was that it bridged the “what-how” chasm. For example, Intel used disruption theory to respond to low-cost threats from competitors with its Celeron line of low-cost microprocessors. According to Clayton, then-CEO Andy Grove credits disruption theory with this success, not because it prescribed what the company should do—rather, it suggested how to think about the problem, and so the company was able to discover for itself what the appropriate response should be.

Some years after the success of Celeron, in Clayton’s re-telling of events, Andy asked Clayton, “How do I set up these new businesses so that they’ll be successful?” Clayton summarized what has become the orthodoxy of disruption: “You need to set up new business units independent of the core. They need to have clear strategic charters that set them on a path to disruption,” and so on. Andy’s response was, “You haven’t told me how to do it, you’ve only told me what to do. I knew that!”

“It was then,” Clayton told me, “that I realized I didn’t even know the difference between what and how.”

If Clayton Christensen can fall into a what/how trap talking about disruption theory with a client who understands it and has used it successfully, what chance do the rest of us have?

To make any headway on this problem, perhaps we need to define what it means to explain to someone how to do something.

Here’s my best guess: When you have explained to people how to do something, they are able to do it. People must actually be capable of taking the right actions in the right sequence under the right circumstances to achieve the right result consistently. So, a cooking show that tells you how to make a soufflé will leave you able to make a soufflé. A how-to book on golf will leave you able to hit the ball down the middle of the fairway, within the limits of your physical abilities. With those criteria, I’ve never seen a cooking show or read a golf book that taught me how to do anything.

I think the same can be said of advice-giving in just about every field, including management. Jack Welch, Larry Bossidy, Lou Gerstner—they’ve all tried to give us some insight into how to achieve the kinds of results they did. Yet with so much knowledge out there, why does the “knowing-doing gap” persist?

I increasingly think that what or how advice depends less on the advice than the advice’s recipient. If Tiger Woods’s coach were to say, “Don’t hesitate at the top of the backswing,” Woods would be able to act on it. Me? Not a chance. The reason is that Woods’s abilities allow him to translate direction into action in ways that I cannot. In general terms, how devolves into what when it comes up against limits of one’s personal experience.

In determining how to explain how, we may be able to take a page from Toyota’s approach to root-cause analysis. When something goes wrong, typically the company quickly identifies a proximate explanation. Then by asking, “So why did that happen?” four more times, Toyota finds that it often gets to the root cause of a problem.

I’ve started using the “five whys” to test the usefulness of any advice I give or receive—whether a new task can be specified to whatever additional level of detail is required to connect with what people already know how to do. Until recipients can find an equivalence between what they already know how to do and whatever new thing they are being asked to take on, the advice will remain in the ultimately sterile land of what.

So I think I’d tell Clayton not to feel bad. None of us knows the difference between what and how, if only because whether a particular recommendation counts as one or the other is a function of whom you’re speaking to.