There is a good idea going around that says your offer doesn't have one buyer. It has about seven.
Everybody writes to the same one. The person in enough pain that they just want it to stop.
Fair enough. That buyer is real, they're ready, and they convert quickest. Which is exactly why every competitor you have is standing on that same square, bidding against you, driving your CPMs up while you all shout the identical sentence at the identical person.
Meanwhile there are six other people looking at the same problem for completely different reasons.
The one who wants to become somebody. The one who's scared of what happens if they do nothing. The one for whom this is an identity question, about who they are and who they refuse to be. The one who's already fine and wants an edge. The one who cares how it looks to everyone watching. The one who has tried five things and needs this to be different.
Same offer. Same mechanism. Different reason to care.
I watched a media buyer post a version of this last week with a screenshot of the ad set: three image variants, body copy written to all seven buyer types, twenty-one ads live in one ad set. Ten years of writing to the pain buyer, thrown out.
The move is right. I'd defend it.
Then you actually run it
We built one out. A hundred and twenty statements across six buyer types for a relationship-coaching offer. Eighty went live across four of them, two held back.
Here's the part I keep chewing on.
The buyer type we'd ranked #1 in the planning doc came second, at 2.33% CTR. A different one won at 2.57%. The two we'd ranked below both of those landed around 2.0% and were close enough to each other to be the same result.
Good news, in a way. That's the whole argument for the exercise. If you'd only written to the planned winner you'd have shipped the second-best angle and never known, because there'd have been nothing to compare it against.
BUT!
The strategic hooks are the ones that died
This is the part nobody posts a screenshot of.
The lines we were most confident about, the ones with the best reasoning behind them, lost hardest.
- The category's proven promise for the planned #1 avatar, the line you'd put on the homepage, the one everyone in the industry already agrees works, finished worst of all eighty.
- A line pulled verbatim off a client call, where the client said the thing so well we wrote it down as "a finished hook, as spoken," came second-to-last in its group.
- A self-blame framing that our own copy rules explicitly blessed for the winning buyer still produced two of the worst statements in the test.
Every one of those had a paragraph of good argument behind it.
The argument was the problem. A well-reasoned hook feels finished, so it never gets held to the same standard as a hook you're unsure about.
Ogilvy: "Never stop testing, and your advertising will never stop improving."
He's usually read as a productivity note. It's closer to a warning about confidence.
Desire beat pain everywhere
The other result, and the one that quietly argues with the premise.
Across every buyer type we tested, desire-framed statements beat pain-framed statements. Not narrowly. Every single one.
Which is a strange finding if pain-avoidance is supposed to be the ripest fruit on the tree.
My read: pain-avoidance is the most crowded segment, not the most responsive one. Those are different claims, and the first one has been quietly standing in for the second for years.
The trap sitting underneath all of it
Here's where twenty-one ads in an ad set can go wrong.
You run them, you sort the report by CTR, you take the top row, you scale it.
A hook that wins on curiosity and a hook that wins on recognition look identical in a CTR column. They behave nothing alike at spend.
I had this exact thing on another account. Top statement at 6.49% CTR, nearly double anything else, pure clickbait. Excluded from the brief entirely. The angle actually scaling on that account sat at 3.25%, roughly half.
The difference is who can click. "Thinking about a Roth conversion?" can only be clicked by somebody weighing one, so every click arrives pre-qualified. A story hook can be clicked by anybody who wants to know what happened next, and most of them will never buy anything.
So the test to run on any winner: does this name a decision the reader is currently facing?
If yes, it survives spend. If it just opens a loop, CTR flatters you and cost per lead collects the debt later.
Which means the angle that scales usually isn't the top row. Sorting by the biggest number in the sheet is how you end up scaling the worst ad in the test with total confidence.
Where we've landed on it
Seven buyer types, written properly, one variable moving at a time.
The planning doc ranks them, and that ranking is treated as a guess. Not a finding. Nothing goes in a brief because it was well argued on a call.
Results get read by construction before they get read by number. Curiosity hooks go in one pile, recognition hooks in another, and we say out loud which pile the top performers came from before anybody writes a brief off them.
Re-rank after every round. The avatar that won this time leads next time, and the lines that lost get properly buried instead of quietly reappearing because someone still likes them.
We hold two avatars back on round one on purpose. Eighty live statements is already more signal than most accounts generate in a quarter, and the held ones give round two somewhere to go.
The seven-avatar idea buys you a much bigger tree.
It does not tell you which fruit is any good. 🤷
If your account is running one angle at four different aspect ratios and calling it a test, that's the first thing we'd change. Book a strategy call and we'll map what your seven actually are.