Read hook and hold before CTR

"Every campaign that produced more than ten orders had a 45% hook rate and 8–12% hold."

That sentence holds the entire answer. And she said it without noticing, while describing the problem.

A cosmetics account in the UAE, targeting 25 orders a day.

The numbers she presented: CPM reaching $3 USD5 to $45 USD, USD sometimes dropping to $8 to $1 USD0 USD when budget rises. CTR between 1% and 3%. Hook rate between 30% and 45%. Hold rate between 4% and 13%. One campaign producing 15 clicks and no orders, another producing an order from one or two clicks at a dollar.

We stopped at that sentence above.

It isn't a question. It's a very clear pattern.

Ads that sell share two traits together: they catch the eye in the first three seconds (45%) and they keep people watching (8% to 12%). Ads that don't sell catch 30% and lose them at 4%.

So it isn't the audience, the pixel or the budget. Half the creatives are taking spend while failing in the first three seconds.

And what about the $45 USD CPM? The UAE is genuinely expensive, that's true. But a CPM that drops to $8 USD when budget rises means the auction finds cheaper people once you give it room. The final cost still spikes because orders don't rise proportionally. Which brings us back to the creative, not the CPM.

The plan she was given:

Build a sheet with every ad and put hook, hold and orders beside it. Anything below 40% hook closes immediately, regardless of CTR. Take the first three seconds from the 45% ads and attach them to the bodies of the ads with high hold. That's assembly, not production from scratch. Fix the budget for five days on winners only, because daily up-and-down movement is what damages cost. And 15 clicks with no order means the page isn't finishing the job, so the ATC funnel she plans is right, but only after the creative is sorted, not before.

It isn't the audience, the pixel or the budget. Half the creatives are taking spend while failing in the first three seconds.