What is creative fatigue in Facebook ads?

It is when the people in your audience have seen the same ad so often that fewer of them click or buy. Meta describes it as a cause of a higher cost per result.

At what frequency does a Facebook ad fatigue?

There is no fixed number. It depends on the audience size, the budget and the creative. Watch for frequency rising while CTR falls on the same ad.

How often should I refresh my ad creatives?

When the data says so, not on a fixed calendar. Review your ads every week and refresh the ones whose CTR has clearly fallen while frequency and cost per purchase rose.

What is a good hook rate?

There is no universal benchmark, because it depends on the audience, the placement and the product. Compare hook rates between your own videos for the same audience, and watch each video over time.

Creative fatigue in Facebook ads: when to replace an ad

Q: How does ROASaurus detect creative fatigue?

It flags a creative when its click-through rate over the last 7 days is 25% or more below the 7 days before, as long as both weeks have at least 1,000 impressions.

Key takeaways

Creative fatigue is when the same people have seen an ad so often that fewer of them respond, which raises the cost per result.
The signs usually come together on one ad: frequency rises, click-through rate falls, and the cost per purchase rises.
Measure fatigue per ad by comparing the last 7 days with the 7 days before, and only on enough impressions to rule out chance.
For video, a falling hook rate means the opening has worn out; a falling hold rate means the rest of the video has.
ROASaurus flags a creative when its CTR over the last 7 days is 25% or more below the 7 days before, with at least 1,000 impressions in each week.

Creative fatigue in Facebook ads happens when your audience has seen the same ad so many times that fewer people stop, click or buy. Meta describes it the same way: people may be less likely to engage with the ad, which can lead to a higher cost per result. The answer is to refresh or replace the creative before the cost rises too far.

This guide covers the signs of fatigue, hook rate and hold rate for video, how to measure fatigue per ad with a worked example, what looks like fatigue but is not, a refresh routine, and what ROASaurus shows.

What is creative fatigue?

Every ad has a limited pool of people who will respond to it. Ad platforms aim to show an ad to the people most likely to act on it. As the budget keeps running, the ad reaches the same people again, and then people who care less. Each extra impression earns fewer clicks and fewer sales.

Meta shows two delivery statuses for this. Creative limited means the cost per result is higher than for ads you ran in the past, but less than twice as high. Creative fatigue means it is twice as high or more. Meta counts all recent exposures of the image or video, including those from other campaigns on your Page.

Meta only shows these statuses for ad sets with a single creative, and not for every campaign type. Most accounts run several ads per ad set, so you need your own check per ad as well.

What are the signs of creative fatigue?

Fatigue rarely shows in one number. It shows as several numbers moving together on the same ad, week after week.

Example: one ad over four weeks

CTR (%)Frequency

Day 1Day 7Day 14Day 21Day 28

Illustrative example, not benchmark data. As frequency climbs, the click-through rate falls. Both lines share one scale: CTR in percent, frequency in average impressions per person.

Signal	What it measures	What fatigue looks like
Frequency	Impressions ÷ reach: how often the average person saw the ad	Rises week after week while reach grows slowly
CTR	Link clicks ÷ impressions	Falls compared with the same ad in earlier weeks
CPM	Cost per 1,000 impressions	May rise, but it also moves with the season and the competition
Cost per purchase	Spend ÷ purchases	Rises. This is the part that costs you money.

One signal alone proves little. Fatigue is when several of them move together on the same ad.

Meta's own help on frequency says the same: if performance starts to drop as frequency rises, the audience may be experiencing ad fatigue, and it may be time to change the creative or the targeting.

There is no frequency number at which every ad wears out. A broad audience with a large budget behaves differently from a small retargeting audience. Compare each ad with its own earlier weeks, not with a general rule.

Hook rate and hold rate for video ads

For video ads, two attention metrics show which part of a creative has worn out. Neither is a standard column in Ads Manager. They are ratios you calculate from Meta's video metrics. ROASaurus uses these definitions:

Hook rate=3-second video plays ÷ video impressions

3-second video plays: Plays of at least 3 seconds, or of 97% of the video if it is shorter than 3 seconds (Meta). On TikTok: 2-second video views.
Video impressions: Impressions of video ads only, so images have no hook rate.

An ad with 20,000 impressions and 5,600 plays of at least 3 seconds has a hook rate of 5,600 ÷ 20,000 = 28%.

Hold rate=ThruPlays ÷ video impressions

ThruPlays: Plays of at least 15 seconds, or of at least 97% of the video if it is shorter than 15 seconds (Meta). On TikTok: 6-second video views.

With 1,800 ThruPlays on the same 20,000 impressions, the hold rate is 1,800 ÷ 20,000 = 9%.

Some teams divide ThruPlays by 3-second plays instead, to see how many of the people who stopped kept watching. ROASaurus divides both rates by impressions, so they share one base and you can read them side by side.

How to read them together

Hook rate falls, hold rate holds up. The opening has worn out, but people who still stop keep watching. Test a new first 3 seconds on the same video.
Both fall. The whole video has worn out for this audience. Make a new concept.
Hook rate is steady, CTR falls. People still watch, but fewer click. Check the offer, the call to action and the landing page.

How to measure fatigue per ad

Compare each ad with itself, over two periods of equal length, on the same metric. Week over week works well. Seven days smooth out the difference between weekdays and weekends, and two weeks is short enough to react in time.

Pick a metric that moves early. CTR is a good choice. Clicks are far more frequent than purchases, so CTR shows a change sooner, and it does not depend on how many orders were tracked.
Compare the last 7 days with the 7 days before.
Set a minimum volume. A CTR based on a few hundred impressions jumps around by chance.
Set a threshold for the drop, and confirm it with frequency and cost per purchase before you act.

A worked example

Ad	Impressions, 7 days before	CTR, 7 days before	Impressions, last 7 days	CTR, last 7 days	Change	Flag
A: customer review	42,000	1.50%	38,000	1.05%	−30%	Fatigue
B: product demo	25,000	1.20%	27,000	1.08%	−10%	No
C: new hook	800	2.00%	4,500	1.60%	Not judged	Too little data

Example data. Ad A fell 30%, past a 25% threshold. Ad C had only 800 impressions in the first week, so its change is not judged.

For ad A, 630 link clicks on 42,000 impressions gave a CTR of 1.50%. The next week, 399 clicks on 38,000 impressions gave 1.05%. The change is (1.05 − 1.50) ÷ 1.50 = −0.30, a drop of 30%. Ad B lost 10%, which is under the threshold. Ad C looks like it dropped 20%, but 800 impressions are too few to tell.

Ad A: CTR change, last 7 days against the 7 days before−30%

Fatigue

Down

-5025

A drop of 25% or more is flagged as fatigue. Ad A is at −30%.

What looks like fatigue but is not

Not every rising cost means the creative is tired. Rule these out first:

Season and competition. In busy periods, such as the weeks before major holidays, CPM often rises for every ad. If all your ads get more expensive at the same time, the cause is the auction, not one creative.
A change on your side. A sale ended, a price went up or a page broke. CTR may hold while the cost per purchase rises.
The learning phase. After a significant edit, the delivery system learns again and results swing for a while.
Tracking problems. If purchases go missing in the platform, the cost per purchase rises on paper only. Compare with the orders in your shop, as explained in why Ads Manager and Shopify sales differ.
Small numbers. One bad day on a few hundred impressions is chance, not a trend.

A creative refresh routine

Every creative wears out sooner or later, so plan for it. A weekly routine keeps you ahead of it:

Review every week
On the same day each week, sort ads by spend. For the ads that spend the most, check the CTR change, the frequency and the cost per purchase.
Confirm the signal
Act when CTR has clearly fallen and frequency or cost per purchase rose with it. First rule out a sale that ended, a site problem or a busy season.
Refresh the part that wore out
A falling hook rate needs a new opening. A falling hold rate needs a new middle or a shorter cut. When both fall, make a new concept. Meta suggests a new image or video that is materially different from the original.
Add before you remove
Launch the new version next to the old one. Meta notes that keeping the original ad active instead of pausing it may maximize results. Turn the old ad off when its results keep getting worse.
Keep a pipeline
Have the next creatives ready before the current ones fade. Record hook, creator and format for each ad, so you learn which kinds of creative last longest.

What ROASaurus shows

The Creatives page in creative analytics reads ad-level spend, impressions, clicks and video plays from Meta and TikTok, and joins them with the orders ROASaurus tracked on your store. For each creative it shows:

Spend, revenue, ROAS and CPA. Revenue comes from orders credited by last click to the ad in utm_content, matched on the ad name or the ad id.
CTR. Link clicks ÷ impressions on Meta, clicks ÷ impressions on TikTok.
Hook rate. 3-second video plays ÷ video impressions on Meta, 2-second views ÷ video impressions on TikTok.
Hold rate. ThruPlays ÷ video impressions on Meta, 6-second views ÷ video impressions on TikTok.
Trend. The change in CTR over the last 7 days against the 7 days before.

The fatigue rule is fixed. A creative is flagged when its CTR over the last 7 days is 25% or more below the 7 days before, and both weeks have at least 1,000 impressions. A flagged creative gets a fatigue label with the size of the drop in the Trend column, and a note at the top of the page names up to three creatives to consider refreshing. The check always uses the last two weeks, whatever period you pick for the report. Because it needs impressions in both weeks, a brand-new ad has nothing to compare with yet.

Fatigue is calculated for whatever you group by. Group by hook, creator or format, and the check runs on all the ads with that hook together. A trend chart compares up to three creatives over time on spend, ROAS, CTR, hook rate or impressions. Frequency and CPM per channel are on the Pulse page, against the previous period. The creative analytics guide explains the setup.

Two limits to know: only Meta and TikTok ad stats are synced, and creative revenue is last click. For other models, use attribution at campaign level.

Sources

In ROASaurus

Keep reading

Creatives, email and AIKlaviyo abandoned cart flows: recognize more shoppersKlaviyo abandoned cart and browse flows only reach shoppers Klaviyo can identify. Learn how it identifies them and how to recognize more at checkout.Creatives, email and AIShopify MCP: ask your store data in Claude and other AI appsA Shopify MCP server lets Claude and other AI apps query your store data. See how the ROASaurus read-only tools work and how to connect them.TrackingAd blockers and tracking: what your store losesAd blockers stop tracking pixels, so ad platforms miss purchases. See which events are lost, why server-side events are not, and how to measure your gap.