The First 90 Minutes

This is a working session, not a reading exercise. Ninety minutes, one person or two, one company. At the end you have a plan you would defend, and enough written context that agents can produce useful work against it rather than generically correct work.

It assumes nothing about tooling. It works faster with a system that already holds shared context, and it works on paper.

Before you start

Have open: your website, your last four sales calls or their notes, your pricing page or the reason you do not have one, and your CRM's list of deals lost in the last quarter.

Have closed: your competitors' websites. You will get to them, and looking first will bend everything that follows toward their framing.

Minutes 0 to 15: what you actually sell

Write one paragraph answering: what does someone stop doing once they buy this?

Not what your product does. What it replaces. Almost every useful positioning decision falls out of this one answer, and it is surprisingly hard to write. If your answer is "nothing, this is new", write instead what budget it comes out of, because that is the thing it is really competing with.

Then write the one sentence a customer said on a call that you keep repeating internally. If there isn't one, that is the most important finding of the session and the rest of it should be spent getting one.

Minutes 15 to 45: who it is for, said with exclusions

Write your customer description. Then do the work: cross out every clause that does not exclude anyone.

"Growing companies" excludes nobody. "Companies with between three and eight sites and no dedicated IT" excludes almost everyone. The second one is a target; the first is a mood.

Then answer four questions in writing. These are the four things any agent will need and cannot infer.

1. Who signs, and who blocks? They are rarely the same person and the blocker is rarely in your CRM.
2. What are they measured on? Not their goals. The number their own review is based on.
3. What do they use today? Including "a spreadsheet and one person's memory", which is the most common and most under-rated competitor in every market.
4. What is the strongest argument against buying? The real one. Write it as they would say it.

If question four is hard, take it from a lost deal rather than inventing it.

Minutes 45 to 75: the market, deliberately late

Now open the competitors. Three, not ten.

For each, write two lines. What they are genuinely better at, and who should honestly buy them instead of you. Doing this before you write anything about your own advantages is what stops the whole exercise from becoming a list of things you would like to be true.

Then write your position as a single sentence with this shape: for [tightly described buyer], we replace [the specific thing they do today] so that [the outcome they are measured on], unlike [the alternative] which [honest trade-off].

The trade-off clause is compulsory. A position with no trade-off is a slogan, and it will produce slogans downstream forever.

Minutes 75 to 90: what happens Monday

Pick three pieces of work. Not ten. For each, write the one line explaining why now rather than in three months.

Then, and this is the part that matters most, write one thing you are deliberately not doing this quarter and why. A plan without an explicit exclusion gets quietly expanded until it is everything, at which point it stops functioning as a plan.

Finally, write the constraints list: claims you cannot make, numbers you do not have, competitors you will not name. This takes four minutes and it is what stands between you and a confidently fabricated statistic in a published post.

What you should have

• One paragraph on what you replace
• A customer description that excludes people
• Four answers: signs, blocks, measured on, uses today, plus the strongest objection
• Three competitors with an honest concession each
• One position sentence containing a trade-off
• Three pieces of work with a why-now
• One deliberate exclusion
• A constraints list

That is the entire input an agent needs to produce work that could not have been written for anyone else. It is also, not coincidentally, the input a new senior hire needs in their first week.

How to tell whether it worked

Give the output to somebody who knows the market but not your company, and ask them to name one thing they disagree with. If they cannot find one, the session produced consensus rather than a position, and consensus does not survive contact with a competitor.

Run it again in a quarter. The parts that changed are the parts that were guesses; note which, because those are the questions worth putting real research behind next time.