The Messaging Test That Takes an Hour

Messaging is testable before anyone books a workshop, on material you already own: one recorded sales call, one month of Search Console, and the last three things you published. Four checks are below. Each comes out of work on a real account, and each one fails visibly.

What is the hour actually made of?

Three sources and four questions. The call needs to be recent enough that the pitch in it is the current one, and the Search Console export only needs queries where you already rank.

No customer interviews, no new brief. Every check runs on evidence that already exists, which is what makes it hard to argue with when it fails.

The four checks, and what a fail looks like:

Check	What you read	Fails when	Where we hit it
First question on the call	One recorded sales call	The buyer opens on how pricing works	Construction procurement SaaS, per-deal pricing replaced
The swap test	Your positioning line	It stays true with a rival name in it	Three-brand operator, three voices and ICPs
Click-through on ranked pages	One month of Search Console	Positions hold and clicks do not arrive	US reputation-management SaaS, 0.11% average CTR
Source of the sentences	Your last three published pieces	No line traces to something a customer said	Global consulting firm, six thin survey responses

What does the first question on a sales call tell you?

On one account pricing had been structured per deal. Prospects spent the call working out what they would owe rather than whether they wanted the product. It was replaced with flat tiers.

That is the check. Play a recorded call and write down the first three things the buyer asks about. If they are about how the model works, the buyer is decoding in minutes you needed for something else.

I can report the change and not the result. The tiers replaced the per-deal model, and what happened to close rates after that was never measured cleanly enough to quote.

Would your positioning survive a swap?

Put a competitor name into your positioning line and read it back. If it stays true, no decision is recorded in it, and the line is doing description.

The sharper version runs inside one company. We work with an owner who holds three separate products. Each brand needed its own voice and its own ICP, and messaging built for one was inert on the next.

Can messaging fail while the rankings look fine?

Yes, and Search Console is where it shows. One account averaged 0.11% organic click-through across a month with positions that looked healthy. Nobody had a ranking problem. People were seeing the words and choosing not to click.

Sort queries by impressions, take the ones where you sit on the first page, and read the title and description that appear against them. Those two lines are messaging in its most testable form, in front of a large sample of your buyer, for free.

Positions looked healthy and organic click-through averaged 0.11% for the month. People were reading the words and not clicking.

Where should the sentences have come from?

A ghostwriting programme at a global consulting firm needed the executives’ own thinking. A survey went out and six responses came back, all too thin to write from. The material existed, in recordings of conversations those people were already having.

So the fourth check. Open the last three things you published and mark every sentence that traces to something a customer or a salesperson said. If nothing is marked, the messaging was written from inside the building, where interchangeable category language comes from.

Two of these checks have their own post: when your pricing model needs explaining and why executives will not fill in your form.