The Strategy Template

Every place I've worked ran into the same gap. There was no repeatable way to share a strategy and actually discuss it. It lived in someone's head, or in a deck that got shown once and never opened again. And in that deck, not enough information was shared to make the strategy clear (let's be honest, it was likely a goal anyhow).

So I built one. A single standard format for writing a strategy down, the same shape every time, so anyone can read one and write one, and the whole team is looking at the same plan.

The goal is to make the strategy clear enough that the people who have to execute it actually can.

When you take it up to leadership, you usually won't hand over the whole document. Pull the few components that matter onto a slide. For how to do that, go see Communicating with Executives.

The order is the whole point. Diagnosis first, objective last. The template forces a crux section and forces you to write down what you will not do, because those are the two parts everyone skips and they're the two parts that make it a strategy instead of a slogan.

Here's the template, section by section. Steal it. Fill it in for something real you're working on, in this order, and don't skip ahead.

Title

Write this last. A good title names the challenge you're taking on or the focused bet you're making, not a vague aspiration. "Owning the Processor Roadmap" is a title. "Innovation Excellence 2026" is a banner.

1. Diagnosis

Start here. Always. Do not write an objective until this section is done.

Describe what's actually going on. Name the critical obstacle. Use the facts you have, including the data and the prior failed attempts. What changed in your market, your technology, or your operations that created this? What's the root, not the symptom? Most of the problems on your list are downstream of one or two real causes. Find those, and name them plainly.

The goal is a frame that shrinks the mess and explains the nature of the challenge, not a list of every complaint.

2. The Crux

This is the heart of the document. Name the one pivotal, addressable obstacle you'll concentrate everything on. If you can't fill this in, you don't have a strategy yet. You have a wish list.

The crux has to pass all three tests at once.

• Important. Solving it eases or clears a stack of other problems downstream.
• Addressable. You can act on it with the resources and authority you actually have. Not "someone, somewhere, should."
• Tractable. You have a real chance of moving it. A real shot, not a moonshot.

The crux is not the biggest problem, not the easiest problem, and not the goal. It's the obstacle in the way of the goal that's worth concentrating force on. State it in a sentence or two.

3. Guiding Policy

Outline the overall approach for getting past the crux. This is a signpost that points the direction, not a task list. It channels action without naming every move. A good one leans on a source of advantage, a strength or a leverage point where concentrated effort pays off more than spread effort. A real one sounds like "fix the data foundation before we build anything on top of it," a direction, not a task list.

Then the part almost everyone leaves out.

What you will deliberately not do right now. Strategy is choice, and choice means refusal. Name what you're choosing to ignore, defer, or stop so you can concentrate on the crux. If this part is empty, you haven't made any real choices.

4. Coherent Actions

List a small set of actions that reinforce each other and all point at the crux. Cut anything that doesn't serve it.

Before you finalize, read the actions together and ask one question. Do any two of them fight each other? Cutting price to win volume while repositioning as the premium choice is two actions at war. If they pull in opposite directions, you don't have coherence yet.

The pivot action. Name the single action that, once it's done, makes the others easier or possible at all. Sequence it first.

5. Proximate Objective

Now, and only now, state the objective. It should fall out of the diagnosis and the crux, not come before them.

Make it proximate. Close enough to be believable and achievable with the resources you have. A target nobody believes is worse than no target, so avoid the blue-sky numbers. State the near-term, feasible target that signals the crux has been overcome.

Execution support

The sections below are not the strategy. They're how you run it. Keep them subordinate to the kernel above, everything from the diagnosis through the objective.

Metrics that matter. Measure outcomes, not activity. Tool adoption, AI usage, and the number of processes you touched are activities. Cost, cycle time, error rate, and utilization are outcomes. Pick the few outcomes that prove the crux is being overcome, and set real numbers with real dates, not placeholders.

Timeline. Sequence the work so the pivot action comes first and creates leverage for the rest. Phase it.

Communication and feedback. How progress gets shared, and how feedback gets back into the strategy. This one is my own addition. Rumelt's kernel doesn't include it, and most strategy docs skip it, which is exactly why so many strategies die quietly on a shared drive. A strategy nobody hears about and nobody can push back on isn't being run. Name how you'll share it and how often, and name how the people doing the work tell you it's wrong.

A final check before you share it

Run the document against this list. If any answer is no, it isn't ready.

• Did the diagnosis come before the objective?
• Is there a clearly named crux that passes all three tests?
• Did you write down what you will not do?
• Do the coherent actions reinforce each other, with a named pivot action?
• Are the metrics outcomes rather than activities?
• Is the objective believable?

See it filled in

A blank template only gets you so far. Next, two worked examples that run the whole thing end to end. One on a simple, relatable problem, winning close games at the free-throw line. One on a heavier bet, Apple deciding to design its own chips. Same template both times.