Performance Creative Strategy: The Andromeda-Era Playbook

Most "strategy" advice is a list of tactics. A real strategy is a system that produces winners repeatably, because the winning ad is unpredictable and no human reliably picks it in advance. Here is the framework we run, built for how the platforms actually deliver ads in 2026.

The thesis: creative is the targeting

Media-buying arbitrage is dead. The algorithm reads your creative to decide who sees it, so the creative is the targeting. That single shift reorders everything: your job is to produce high-volume, high-diversity, hyper-specific concepts at a solid quality bar, and let the platform find the winner. We unpack the mechanics in why creative is the new targeting.

The unit of work: the concept

Stop thinking in assets, start thinking in concepts. A concept is a specific persona × angle × offer × format. Change any one variable and you have a new concept that can reach a new pool of people. Tracking which variable changed, and why, is what turns "make more ads" into a system you can learn from.

Personas have to be hyper-specific. "Busy professional" is dead; "the consultant who flies twice a week and needs a carry-on that works for boardrooms and bars" is alive. If a persona could describe more than one type of customer, it is too broad to brief against.

Test in the right order

Not all four variables have equal leverage. Test them in this order:

1. Angle first. highest leverage, change it when things are not working.
2. Offer next. high leverage but limited surface area.
3. Persona once angle and offer are exhausted.
4. Format last. lowest leverage; do not lead with it.

Most teams do this backwards, swapping formats and editing styles while leaving a losing angle untouched.

The production mix: 60-20-20

Allocate output deliberately, weighted to proven ideas: 60 percent replicate (restage a winner with a new avatar or scene), 20 percent iterate (hook and format swaps on a working concept), 20 percent net-new (genuinely new concepts). Net-new feels like progress but is the lowest-leverage work, so it gets the smallest slice. Full breakdown in how many ad creatives you need per month.

Hook strategy and the awareness gradient

The hook is the full combination of what the viewer reads, hears and sees in frame zero, plus the package vibe, not just the first line. Because hook changes count as new creatives, rinsing one winning body into many hooks is the cheapest scale lever there is.

The body should move the viewer along the awareness gradient (unaware, problem-aware, solution-aware, product-aware, most-aware) and introduce the product as late as it credibly can. The counterintuitive truth: viewers who reach minute 1:30 of an educational asset convert far better than viewers who hear the brand name at second 5. A hook that jumps straight to the product feels like an ad and gets skipped.

Lead with pain

Best-performing accounts run roughly 60 to 70 percent pain-led creative; legacy campaign-shoot brands over-index on desire. Problem agitation beats persona call-out almost every time. See the split in performance vs brand creative.

Make it a loop

Research the market, brief concepts, produce at volume, test, read the leading indicators, rinse winners, kill losers, feed the result into the next batch. Every result sharpens the next one. That loop, run consistently, is the strategy. If you want it run for you, book a free creative audit.