Positioning is what you refuse

The standard positioning engagement produces a statement, a set of pillars, and a messaging framework. It is presented, approved, and circulated. Six months later the company is pursuing exactly the same opportunities it was before, describing them slightly differently.

This is not a failure of the work. It is what happens when a positioning exercise is treated as a communications project rather than a resource-allocation one. Words are cheap to change and cost nothing to hold. A position is expensive by construction — it works precisely because it removes options, and removing options is the part nobody signs up for.

The diagnostic question we ask before starting is simple and slightly rude: what would you have to stop selling for this to be true? If the answer is nothing, we are being asked for copy, and it is better to say so than to deliver a document that will be quietly ignored.

You cannot assess a position by reading it. You assess it by looking at what the business declined.

Ask what work was turned away in the last two quarters, and why. A company with a real position can answer immediately and specifically — a named client type, a project shape, a budget band. A company with a described position gives a general answer about fit, which usually means nothing has ever been refused.

The second question is harder: what did you take that you should not have? Every business has these, and the honest ones can name them. What matters is whether the exception was discussed as an exception or absorbed without comment, because the second is how a position dissolves without anyone deciding to abandon it.

Nobody ever announces that the positioning is being dropped. What happens is a slow quarter, an enquiry that is adjacent rather than aligned, and a reasonable argument that this one is different. It usually is different. The problem is that the next one is also different.

The asymmetry is what makes this so hard. The cost of holding a position is immediate, specific and attributable — a named deal, a known value, a person who has to make the call. The benefit is diffuse, delayed and impossible to attribute: work that fits better, references that compound, a reputation that eventually generates inbound. One of those shows up in a forecast and the other does not.

The practical defence is to decide the exception rule before you need it. Who can approve off-position work, what has to be true, and how many per year. A position with a stated exception budget survives contact with a bad quarter; one that depends on discipline in the moment does not.

The most common objection is that a narrow position shrinks the addressable market, and arithmetically that is true. What it misses is that addressable market is not the constraint for most companies — attention is. A firm that could serve anyone competes for every deal against everyone, at the cost of being memorable to nobody.

There is a real trade here and it deserves naming rather than dismissing. Narrowing genuinely does reduce the number of opportunities you are eligible for, and if you are already capacity-constrained on a broad base, narrowing may cost you more than it returns. The case for it is strongest when you are winning too few of the deals you enter, not when you are entering too few.

Where it consistently pays is referral behaviour. People cannot refer a generalist, because nothing triggers the recommendation. A specific position turns your existing clients into a distribution channel, and that effect compounds in a way paid acquisition does not.

We quote from a written brief rather than a negotiation, which removes us from any process that depends on haggling a number down. We turn down roughly one in four enquiries, most often because the work would be execution against a plan we think is wrong, and we would rather not spend a year being blamed for a strategy we disagreed with.

The most expensive one: we decline engagements where the client will not give us access to revenue data. It sounds procedural and it is actually the whole position — we sell measurement-led work, and without the outcome data we would be selling activity. That refusal has cost us real money in quarters where we could have used it.

None of that is offered as virtue. It is offered as evidence, because the argument of this piece is that a position is only demonstrated by what it costs you, and it would be inconsistent to make that case without stating our own bill.