In a long article on X, Meta ads consultant Adam Taylor argues that rising acquisition cost is rarely a spend or platform problem. His diagnosis for stuck accounts: one ad carrying a third of the budget, frequency at 5 or higher, CPMs creeping up. You are paying more to show the same message to an exhausted pool.
The strategy
The core exercise is a grid. Personas down the side. Across the top, five awareness levels: unaware, problem aware, solution aware, product aware, most aware. Place every active ad in its cell. He says the audit almost always looks the same: 15 to 20 cells, two or three occupied, nearly everything crammed into the existing buyer persona at product aware with a discount. More budget does not make a full cell bigger. The empty cells hold the next customers.
To fill them, he collects angles the market has already validated rather than brainstorming: export a few hundred reviews and have an AI assistant like Claude pull out the customers' exact language, search Reddit for the pain point sorted by upvotes, note competitor ads that have run 90 or more days, and sort your own account by spend. Keep the customer's words verbatim. Then make genuinely different concepts for each new cell, not the same UGC style with a new hook. Scale winners by recreating them with new creators, formats and personas, and watch CTR, CPM and frequency weekly so angles are retired before they die.
Why creative is the targeting on Meta in 2026
On Meta today the creative effectively is the targeting. Each distinct angle opens a new pocket of buyers; near duplicates bid against each other.
How to copy it
Build the grid this week. Brief one ad for each empty cell using language lifted from reviews.
Credit: @adamtaylorl on X