Performance dropped. You assume the ad got tired and swap in a fresh one. Sometimes that fixes it. Sometimes it does nothing and you've wasted a week. Here's how to tell the difference before you act.
Why diagnosing first matters
Every fix costs money, and the wrong fix costs the most because you also lose the week finding out. A routine worth building: performance dips, you check three readings, and each points at a different cause. Skip the reading and you're guessing. Guess right half the time and you still lose half the weeks.
The diagnostic is genuinely cheap. Frequency and CTR you can pull in a minute. Landing page conversion is in your analytics. CPM is on the campaign card. Four numbers, one glance, and the swaps usually make themselves.
Fatigue has a specific signature
Creative fatigue shows up as frequency climbing while click-through rate falls, on the same audience, at the same time. That combination is the tell. The ad is being shown more to the same people, and those people are responding less each time. That's fatigue, and a new creative is the right fix.
The logic is easy to internalize if you walk through it once. Rising frequency means the pool isn't growing, or isn't growing fast enough for the budget. Falling CTR means the same viewers respond less as repeats stack up. Both together, with the spend unchanged, tell you the audience has absorbed the message and now scrolls past it. The ad hasn't stopped working. It's finished the work it was sent to do.
If frequency is sitting around 2 or 3 and performance still dropped, it's not fatigue. The audience hasn't seen it enough to get tired of it. Something else moved, usually the auction, a competitor pushing spend, seasonality, or a tracking change. Swapping creative won't help because the creative wasn't the problem.
The numbers to actually check
Pull frequency and click-through rate together over the last two weeks. If frequency is rising and CTR is declining on a downward slope, that's fatigue. If CTR is steady but conversions dropped, the problem is downstream, your landing page, your offer, your pricing, not the ad.
That second pattern trips up the most people. The ad still gets clicks, people still respond to the message, but they don't buy. That's a landing page problem or an offer problem, and a new ad only changes the front door. Same funnel, same drop-off, same disappointing week. When I audit an account, this is the most common misdiagnosis: three creative rounds with no change while the page checkout kept leaking money.
If performance swings up and down day to day, that's auction noise, not fatigue. Fatigue is a gradual, sustained decline, not a bad Tuesday. A single weak day means nothing. A week-long slide with climbing frequency means real fatigue.
Split the read by device and audience while you're in there, because two days of diagnosis beats three weeks of wrong fixes. If mobile frequency is 4 and desktop is 1.5, fatigue lives on mobile, mobile creative or mobile spend is the lever. If it's a single ad set sliding while the rest hold, fatigue lives there, that's your new creative. Whole-account dips don't point at creative at all, they point at the auction or the market.
Why this matters
Founders panic-swap creative on every dip and never learn what's actually wrong, because they changed the variable before diagnosing it. If the real issue is a landing page that stopped converting, a new ad sends the same traffic to the same broken page and the numbers stay bad. Now you've blamed the creative for a CRO problem and you'll keep repeating it.
There's a subtler version of this mistake. Fresh creative resets the fatigue numbers, so after a swap, the fatigue signal washes out for a week. Meanwhile the real problem, a slow page, a price that stopped matching the market, keeps costing you. Each unnecessary swap hides the real issue for another seven days. Diagnosis first, because a wrong call has a whole week's rent attached.
The move
Diagnose before you swap. Keep the order short and fixed:
- Check frequency and CTR for the fatigue signature, rising with falling, same audience.
- Check your landing page conversion rate for a downstream drop.
- Check whether the auction got more expensive, your CPM, for a market shift.
Only change the creative when the data says creative. That discipline is the difference between a founder who fixes problems and one who keeps swapping ads and wondering why nothing changes.
A worked read, start to finish
One full pass so the method has no gaps. Say a founder reports "ROAS dropped from 3.4 to 2.3 over ten days, should I make new ads?" First look: frequency went from 1.6 to 3.9 over those ten days while CTR fell from 1.8% to 0.9%. That's the signature, both moving the wrong way together, gradual and sustained, no single-day crash. Creative fatigue, and a new concept is the right move.
Second case, same question, different answer. Frequency steady at 1.7, CTR holding at 1.7%, but cost per purchase still climbed a third. Next check: CPM up 40% in the window and up for competitors too, so the auction got expensive. New creative won't fix an auction, bid strategy and budget pacing might. Third case: CTR fine, frequency low, but site conversion fell from 2% to 1.1% in analytics. The ads are still pulling clicks that the page is failing to turn into orders. New ads would've been the third wrong fix in a row, and each one would've cost a week.
Same complaint three times, three different causes, and only the four numbers made the calls. Guessing averages one right answer out of three. Reading the numbers gets you three out of three, every time, and it takes minutes.
I'm Hirdesh Matai, founder of HDM GRO-ads, the science of e-commerce advertising.