How to audit your own Meta ad creative in 30 minutes
Open Ads Manager at ad level, last 30 days, sorted by spend. In half an hour you can establish three things: whether your live creative is fatiguing, whether you have enough distinct concepts to be testing at all, and whether anything is in production to replace what dies next month. Every check below has a published number to measure against, so the answer is a threshold rather than an opinion.
What you are actually looking for
Almost every creative problem in a paid social account is one of three things, and they need different fixes. Fatigue means the work was good and is now worn out — the answer is replacement, not reinvention. A thin concept pool means you are not testing enough distinct premises to find a winner, so the account is running on luck. A missing pipeline means both of the above will keep happening, because nothing is being made while the current set burns down.
Rising CPL is downstream of all three, which is why it is a terrible thing to diagnose from. By the time cost per lead moves, the money is spent. The checks below all sit upstream of it.
The thresholds worth knowing
These are the published numbers the audit measures against. Write them down before you open the account, so you are comparing rather than rationalising.
| Signal | Watch | Act |
|---|---|---|
| 7-day frequency (prospecting) | Above 2.0 | Above 2.5 — at ~3.5 the ad is losing to a fresh variant |
| CTR vs its own baseline | Down 10–20% | Down 20–25% |
| Repeat exposure | — | Conversion likelihood is down ~45% by the 4th exposure |
| Distinct concepts taking spend | Under 4 | Under 3 — one fatigue event moves blended CPL |
| Creative age | Approaching 28 days | About half of all creatives are retired before day 28 |
Where teams misread what they find
The most common misread is blaming delivery. Frequency climbing and first-time impressions falling is not the algorithm turning on you — it is the algorithm running out of new people to show a finite set of creative to. That is a supply problem on your side.
The second is counting variations as concepts. Twenty cuts of one premise fatigue together, because the audience is not tired of the edit, they are tired of the idea. If your audit finds twelve live ads and two premises, you have two.
The third is judging a creative on CTR alone. A hook can buy the click and lose the lead. Pair thumb-stop with what happens after — if hook rate is healthy and CPL is not, the problem is downstream of the first two seconds and no amount of new hooks will fix it.
What we see most often
We run this pass before quoting anyone, and across four years of MENA fintech accounts the findings cluster tighter than you would expect. Three patterns cover most of it.
- Two premises wearing twelve costumes. The account looks busy — a dozen live ads — and the audit finds two ideas. They fatigue in the same week, because it was always one thing.
- Nobody can name the replacement. Everyone can name what is dying. Ask what goes live when it does and the answer is a plan, not a file.
- The last winner is being defended. Budget stays on a fatiguing ad because it worked, and the account spends three more weeks proving that it used to. Frequency said so a fortnight earlier.
What to do with what you find
None of the three are strategy failures. They are supply failures wearing a strategy costume, which is why the fix is a production cadence rather than a new deck.
If fatigue is the finding, the fix is inventory: you need replacements ready before the current set dies, not commissioned after. If the concept pool is thin, the fix is upstream — more distinct premises, tested small. If nothing is in production, that is the finding, and it outranks the other two.
The arithmetic behind how many you need is its own subject, and we wrote it up separately in how many ad creatives a MENA fintech needs per month. The short version: at a roughly 5% win rate, the number of winners you end a month with is set by how many concepts you shipped.
The 30-minute creative audit
Minutes 0–5 — set the view up honestly
Ads Manager, ad level, last 30 days, sorted by amount spent. Not campaign level, not lifetime. You are auditing creative, so you need the view where creative is the unit and recent spend decides what matters.
Minutes 5–12 — check frequency on your top spenders
Add the frequency column and look at the top five ads by spend on a 7-day window. Anything above 2.5 on prospecting is in the refresh zone. Add first-time impression ratio if you have it — falling below roughly 50% means delivery has shifted from reaching new people to re-serving the same core.
Minutes 12–20 — measure hook rate, then the drop after it
Divide 3-second video views by impressions for each top spender. Compare each ad against its own first week rather than against a global benchmark. A hook rate holding while CPL climbs points downstream — offer, landing page, audience — not at the first two seconds.
Minutes 20–26 — count concepts, not ads
List the distinct premises that took meaningful spend in the period. Different hooks on the same idea are one concept. Most accounts that feel busy turn out to be running two or three, which is below the level where a test result means anything.
Minutes 26–30 — find the refresh gap
Count what is in production right now against what will fatigue in the next 30 days. If more creative is dying each month than is being made, the account is shrinking regardless of what this month's numbers say. That gap is the finding worth acting on first.
Questions we get asked
What frequency is too high on Meta?
For cold prospecting, risk rises noticeably once 7-day frequency passes roughly 2.0–2.5. Around 3.5 an ad is usually losing money against a fresh variant. Retargeting tolerates more, because the audience is small by design and already knows you — judge it on cost per result rather than on the frequency number itself.
How do I know if it is the creative or the offer?
Split the funnel at the hook. If thumb-stop and hook rate are healthy but cost per lead is climbing, the first two seconds are working and the problem is downstream — the promise, the landing page or the audience. If hook rate itself is falling against its own baseline, that is fatigue, and new creative is the fix.
Can I audit creative without a big spend?
You can run every check above at any budget, but the smaller the spend the noisier the read. Below roughly $25,000 a month there is often not enough traffic for creative tests to separate cleanly, so weight the concept-count and pipeline checks more heavily than the performance deltas — those are structural and do not need volume to be true.
How often should this be run?
Monthly is enough to catch structural problems, weekly if you are actively scaling. The one moment it is worth running immediately is before you brief an agency or a freelancer — knowing whether you have a fatigue problem or a volume problem changes what you should be buying.
Sources
Keep reading
- What to brief a video agency for performance creative →
Most briefs describe a film. Performance creative needs a brief that describes an argument, a market and a kill condition — and the teams that cannot write twelve of those are the ones that stall at four ads a month.
- Hook rate vs thumb-stop rate: which one actually tells you something →
Most teams use the two names for the same calculation, then wonder why the metric keeps looking fine while cost per lead climbs. The useful split is not between the two names — it is between stopping and staying.