Iterate on winners. Diversify on concepts. Do both on a schedule.
The system underneath: analyze why top performers win, iterate on those hypotheses, run fresh concepts in parallel for diversification, and watch creative fatigue trends weekly. This is how an ad account learns and improves on itself.
Book a strategy callGood creative is hard. Good creative consistently refreshed is even harder.
Most teams aren't launching ads weekly, which is the ideal frequency. It's hard to produce that much content. But even when you are, it's often just output, without the analysis and input needed to iterate on top performers.
- Regardless of the tests being done, the same 2 ads win. Nothing else compounds.
- Top performers start saturating and CPLs / CPAs rise beyond comfort.
- You've paid for a lot of content that no one is using.
- Your team isn't measuring how well your ads are grabbing attention, at least not on a trendline.
- Your team isn't able to easily show how test hooks or test concepts perform across Ad Sets.
- You don't know how long it takes new ads to "pick up spend."
- "Iteration" is whatever your editor created, not data-backed hypotheses.
The creative engine approach to creative strategy.
Beyond "more ads", winning creative strategies are made up of iteration + diversification. Our Creative Engine is the system that makes ad campaigns profitable.
The engine
- 01 Performance Report
- 02 WHY hypotheses formed
- 03 Iteration briefs
- 04 Diversification briefs
- 05 Assets
- 06 Editors · Designers
- 07 Approval · Edits
- 08 Media team launch
- 09 Test Period
The operator
Even if you get the system right, you also need the performance creative operator structuring the ads to drive ROI, observing and learning from top competitor ads, and organizing the split between iteration and diversification.
Algorithm spend is never random.
A lot of people get frustrated by which ads the algorithm concentrates spend on, the hero ads as we call them. But algorithm spending is synonymous with data plus attention. It's never random.
Get attention with the right people, strategically build data on new ads, and you'll see spend shift.
Every winning ad does three things.
Here's one in action. Watch the same ad do all three jobs in sequence: stop the scroll, hold the interest, earn the click.
How we build your creative engine.
This is how we get your ad account from always needing more creatives, to not only having enough, but also building on insights of what's actually working.
Analyze why winners win.
Every top performer gets a written hypothesis: what about this ad earned attention. Hook style, pacing, music, narrator, message frame, claim, format. Not just "this one converted" or "video works best," but "this one earned a 28% hook rate because the first 2 seconds did X." Without the why, the next 10 ads are guesses.
Iterate on hypotheses, not new ads.
The next ads should be variations testing what we believe about the winners. Different hooks on the same message frame. Different narrators on the same hook style. Different proof points on the same creator. Each iteration is an experiment, not a fresh idea.
Diversify with fresh concepts in parallel.
Iteration alone narrows. The second prong of a strong creative engine is net-new concepts running alongside iteration. This is especially important under Andromeda. When a fresh concept earns its spend, it graduates into the iteration lane.
Run it on a schedule, not ad hoc.
The single biggest difference between creative systems that compound and creative systems that don't is cadence. Editors briefed weekly. Concepts shot or sourced in a continuous pipeline. Launches on a fixed day. Iteration retros on another fixed day.
Keep it simple.
The biggest sign a creative engine is too complicated is that it's not getting executed on. We're always removing friction from the processes until they're being executed on with ease, while still hitting the goals of volume and performance improvements.
Two ways to engage.
Some brands want us to run the creative engine end-to-end. Others have talented teams in place who need the system and the framework. Both work.
Option A
We run the creative engine end-to-end.
We own the system: sourcing, briefing, iteration cadence, fatigue tracking, refresh schedule. This can be paired with or without us managing the ads and testing.
- Sourcing pipeline build-out (the assets)
- Sourcing data build-out (ad performance)
- Weekly editor briefs with iteration hypotheses
- Creative Fatigue Dashboard
- Launch cadence + refresh schedule managed
Option B
We bring the system to your team.
You have a creative team. They're good. They just don't have the framework or the cadence. We install both, train your team, and can stay on as strategic oversight when needed.
- Sourcing pipeline setup, your team runs it
- Sourcing data build-out (ad performance)
- Iteration brief template + hypothesis framework
- Creative Fatigue Dashboard built for your team
- Cadence calendar installed, your team executes
- Weekly strategic review as needed
The creative engine, in action.
A glimpse of what the assembly line produces, week over week.
Lifestyle UGC
Social-native iteration
Practitioner UGC
Dashboards that watch your creative for you.
Every Creative Strategy engagement comes with two dashboards built around your account. One watches for creative fatigue before it shows up in CPL. The other tracks creative tests so you know what's actually working before it's earning revenue.
Creative Fatigue Dashboard
Saturation scores, spend share by ad, top-spending ads change tracker, and trendlines on hook rate, CTR, and CPC. You see the warning signs of saturation before performance breaks.
- Hook rate trend per ad, week over week
- Spend share concentration tracker
- Launch frequency + test ads spend pickup trendlines
- Iteration queue, sorted by what's earned the next slot
Creative Testing Dashboard
Tracks every test hook and concept across Ad Sets so you can see what's actually moving the needle. Surfaces how quickly new ads pick up spend, normalizes performance across Ad Sets, and gives you a single read on what's earning algorithmic attention before it's earning revenue.
- Test hook performance, normalized across Ad Sets
- Spend pickup velocity for net-new ads
- Cross-Ad-Set concept comparison
- Iteration vs. fresh concept ROI tracking
The creative system, in numbers.
Lifted monthly evergreen launches roughly 3x, with hook rate climbing from 15.5% to a 28.1% peak.
Inherited a Top-of-Funnel Meta account whose creative had stalled. The first attempt at fresh inventory was generic UGC creators. Quality was sub-par. The pivot was practitioner storytelling, not creator content: real practitioners telling their own stories, edited on a continuous assembly line, with iteration briefs running alongside fresh concepts.
Excluding BFCM promotional creative, evergreen launches went from a ~24/mo agency-period average to ~76/mo across the engagement. Hook rate climbed above the prior baseline by month three and peaked at 28.1% in month four. AOV moved up 12%. Lower lead cost, higher value per customer, on a creative engine that produced winners on schedule.
Read the full case study →Want creative that compounds?
One 30-minute call. Tell us what's going on with your ad creative and we'll offer free, no-strings advice on what you need and whether we're a fit.
Book a strategy call