Diagnosing the Challenge

You can't fix what you won't honestly name.

That's the whole reason diagnosis comes first, and it's why it's the step everyone wants to rush. Writing a guiding policy feels like progress. Listing actions feels like progress. Naming, out loud, the real thing that's broken, the one that's usually nobody's favorite to say in a room full of the people responsible for it, feels like conflict. So teams skip it, jump to the plan, and spend the next year executing beautifully against the wrong problem.

A diagnosis isn't a list of complaints. It's a frame that takes a tangled mess and explains the nature of it. Most of the problems on your list are downstream of one or two real causes. The job is to find those one or two and say them plainly.

This is the deep version of the kernel's first step. The kernel is the five-step method from the last page, diagnose, find the crux, write a guiding policy, build coherent actions, implement and adjust. One note before we start, so two ideas don't blur together. This page is about diagnosing an organization's strategic challenge, the obstacle between the company and its vision. That's different from defining a customer's problem, which is its own discipline and lives in the Opportunity Backlog. Both matter. They're not the same step.

Use a questionnaire, not a meeting

Here's what I use. Before writing a single line of strategy, send a short questionnaire to your senior leaders and a few select team or department members and have each of them answer it on their own, in writing, before anyone compares notes.

On their own and in writing matters. Put the same people in a room and you get the loudest voice, the company line, and a lot of nodding. Ask them to write first and you get what they actually think, including the things they'd never say first in a meeting. You're not running this to reach consensus. You're running it to surface the honest reads, especially the ones that disagree.

The questions below aren't mine. They come from Richard Rumelt's work on how to diagnose a challenge, and they're built to pull the cause to the surface instead of the symptoms.

• What's been the most important change in technology, competition, and customer behavior in the last five years, and how has it hit us?
• Looking out three to five years, what changes do you expect in technology, regulation, competition, and buyer behavior? Which of those are problems for us, and which are opportunities?
• What programs or projects in the last five years are you genuinely proud of? What made them hard, and what let us get through it?
• What programs or projects in the last five years failed? What prevented success, and what would you do differently?
• What are our top priorities right now, and what's actually underway to address them?
• What are the two critical challenges facing us? Note what's blocking each one from getting resolved.
• Are any of these challenges created by our own structure or core policies? Are any of them big enough to block progress on everything else?
• Of all the challenges you named, which single one, if we overcame it, would free up the most progress, and do we have a realistic way to act on it right now?

That last question does a lot of work. It's the bridge from diagnosis to the next step, finding the crux, because it forces each leader to point at one thing and pressure-test whether it's actually addressable.

What the answers actually give you

When I've run this, the same patterns show up every time.

Not everyone answers. The people who do answer don't all say the same thing, and the disagreement is the point. Where two thoughtful leaders see the same company completely differently is usually exactly where the real challenge is hiding, because it means the thing is unresolved enough that smart people can't even agree on what it is.

And the honest answers, the ones written alone before the room could soften them into the company line, tend to name the problem the official strategy has been carefully avoiding. That's the one you're looking for. Read for the obstacle that keeps showing up underneath the surface complaints, even when people describe it in different words. The frequency is the signal.

You're not trying to address everything that comes back. You're trying to find the one pivotal, addressable obstacle that the rest of the strategy will concentrate on. The diagnosis narrows the mess. The crux picks the one.

What it looks like filled in

Take the Apple example from the worked strategy, the decision to build their own chips. Run the same questions at Apple around 2007 and the honest answers start pointing the same direction.

The biggest change of the last five years? Our products live or die on how tightly hardware and software fit together, and we keep getting better at controlling that, everywhere except the one component that matters most. A project that didn't succeed? A supplier's slipped schedule, or a chip that was never tuned to our product, gated a launch we owned. A challenge created by our own structure or core policy? We've organized around buying the processor instead of owning it, so the most important part of the product runs on someone else's calendar.

And the last question, the one whose answer would free up the most progress and that we could actually act on now? Control of the processor roadmap. Notice what that answer is not. It isn't "ship a better phone this year," that's a goal. It isn't "beat Intel," that's an opponent. It's the obstacle in the way. That's a diagnosis you can build a strategy on, and in the worked example you can see exactly the strategy it became.

Then, and only then

A diagnosis done honestly is most of the work, and it's most of what separates a real strategy from a wish. Once you can name the cause instead of the symptoms, and once you've found the one obstacle worth concentrating everything on, you're ready for the next move, turning that crux into a guiding policy, the overall approach you'll take to get past it.