This is the frame I use. It is empty on purpose. Fill it in your head as you read, and pay attention to where you stall.
Most companies can do the left half in a minute. It is the same half that goes on the careers page, and there is a test for it at the bottom of this page that almost nobody passes.
The one page
Swipe the frame sideways
Watch what happens in the third column.
One year priorities. Three of them. Not eleven. If your list has eleven, you do not have priorities, you have a backlog with a nicer name.
Northstar KPIs. One leading, one lagging. This is where most companies stop entirely. Not because they lack measurement. Because naming one number as the one means telling several people their number is not it.
The blank cells are not gaps in the document. They are the argument you have not had.
The fourth column has the same rows as the third, filled in for a year further out. That is deliberate and it is the cheapest diagnostic on the page.
If the two columns are identical apart from bigger numbers, you do not have a strategy. You have a plan to keep going.
And the same object, one resolution down
Swipe the frame sideways
That second line in every cell is the only part of this frame I will not let a company drop.
A format with nowhere to put the customer's words will produce a feature factory.
And it will do it without anybody deciding to. The constraint has to live in the layout, because a principle in a document does not survive a busy quarter.
It is high level because that is the only honest resolution at the start. And what I am really testing is not whether you can write the plan. It is whether you can change it when the research says you should.
Those two are the top of a ladder. Underneath them are four more, and each rung exists to be traced back to the one above it.
Six rungs, each one answerable to the one above
Swipe sideways
Rung six is the test, and it is the cheapest one on this page.
Take any item on your current roadmap. Write the sentence a customer would say to ask for it. First person, their words, no product language.
If the sentence comes out easily, the item is probably real. If you find yourself writing a sentence no human has ever said, you have found something.
Every rung is answerable to the one above it. That is the whole method.
I said at the top that the easy half has a test. Here it is, and it takes ten seconds.
Get the whole strategy down to four words. Three of them will be things any competitor in your category could also say. That is fine. That is what the first three are for.
The shape it has to end up in
The other thing that matters here is the order they arrive in.
Say the first three out loud to a room and nobody argues, because nobody can. By the time the fourth one lands, they have already agreed to everything holding it up.
You do not get to tell a company what its strategy is.
You can build the four lines so that they say the last one themselves.
The document is the easy half. The reason it takes six weeks is that the third column cannot be written until somebody has been told their number is not the one.
If you can already fill both of these in, and the people around you would fill them in the same way, do not hire me.
Send them to four people who run different parts of the business, separately, and compare. That costs you nothing and it tells you most of what I would tell you.
If the four versions disagree, the disagreement is the work, and it is worth having somebody from outside hold it.