Ad Fatigue on Facebook: 7 Signs and What to Do About It

Every ad wears out eventually. The people most likely to respond have seen it, the rest have learned to scroll past, and your cost per result slowly climbs. That’s ad fatigue.

The problem is that many advertisers switch ads off far too early, because they assume an ad is “old” after two or three weeks. Others leave a dead ad running for months because the account still spends. This post shows how to tell real fatigue from noise, and what to do when you find it.

How long ads really last

Start with what the market shows. In our archive of 9,839 Meta ads from 5,185 brands that each ran 60 days or more in Europe (snapshot 31 October 2026), the median ad had been running for 66 days, and many ran far longer. Brands don’t keep paying for ads that stopped working, so a winning creative clearly isn’t used up in two weeks.

Copy length doesn’t change that much. Short, medium and long post texts in the archive had nearly the same median run: 68, 67 and 66 days. What makes an ad last is the idea behind it, not the word count.

So the question isn’t “how old is this ad?” It’s “is this ad getting worse, and why?” See how long to run a Facebook ad for more on what “long-running” means.

The 7 signs

Look at these over a week or more, not day to day. A single bad day is noise.

1. Frequency keeps rising while results fall. Frequency is the average number of times each person saw the ad. A rising frequency on its own isn’t a problem; a rising frequency with a falling click-through rate is the classic pattern.

2. Click-through rate drifts down week after week. Compare the ad with its own first weeks, not with an industry average. A slow, steady decline is fatigue. A sudden drop is usually something else.

3. Cost per result rises while the cost per thousand impressions stays flat. If you pay the same to reach people but get fewer results, the ad is doing less work. If both rise together, look at the auction and the season first.

4. Fewer people watch past the first seconds. For video, the share of impressions that turn into 3-second views is a good early signal. When the hook stops stopping people, the rest of the video doesn’t matter.

5. The comments change. “I keep seeing this ad”, “stop showing me this”, or the same question asked for the tenth time are all signs that the audience has seen it plenty.

6. More of the spend goes to people who already know you. If an ad’s results lean more and more on returning visitors and existing customers, it may have run out of new people to convince.

7. The newer ads beat it in the same campaign. When Meta’s delivery starts moving budget away from an old ad towards newer ones, that’s a vote. Respect it.

One sign alone proves little. Two or three together, over a week or more, are a strong case.

What looks like fatigue but isn’t

Before you blame the creative, rule these out:

• Seasonality. Costs often rise before big shopping dates and in competitive weeks. Every ad gets more expensive at once.
• Tracking problems. A broken pixel, a consent banner change or a checkout update can make results “drop” overnight. If every ad fell on the same day, check tracking first. Our guide to setting up the Meta Pixel has a checklist.
• The landing page or offer changed. A price rise, a sold-out size or a slower page will hurt the ad’s results, but the ad isn’t the cause.
• Account changes. New ad sets that compete for the same people, budget jumps or edits that restart learning can all make an ad look worse.
• A competitor’s big push. If a competitor launches a big offer in your market, your costs can rise for a while.

A quick test: if all your ads got worse at the same time, it’s probably not fatigue. Fatigue hits one ad, or one angle, at a time.

How to refresh a fatigued ad

You don’t need a new idea every time. Most of the value of a winning ad is in its angle, and the angle usually outlives the creative. Refresh in this order, from cheapest to biggest change:

1. New hook, same body. A new first line or new opening two seconds. Often enough to win back attention. Our hook formulas give you structures to try.
2. New first frame or thumbnail. For video, a different opening shot. For statics, a different main image with the same message.
3. Same message, new format. Turn a winning video into a static with a bold claim, a carousel, or a UGC-style clip.
4. Same angle, new person. A different creator or customer telling the same story.
5. New angle. Only when the angle itself has stopped working across several creatives.

Keep the old ad running while the new versions prove themselves, unless it’s clearly losing money. Switching off a winner before you have a replacement is a common way to have a bad month.

Build a pipeline so fatigue never surprises you

The best defence is to have the next ads ready before you need them:

• Launch a small batch of new creatives every week or two, even when results are good.
• Keep a list of your current winners with the date they started and their weekly cost per result. Watch the trend, not the day.
• Keep a swipe file of long-running ads from your market, so you’re never starting a new batch from a blank page.

The free Paid Ads Workbook helps you plan how much new creative you need each month, and the adthief app shows which ads in your market have kept running for months, so you can see what lasts before you make it.