Paid Ads & Creative
7 Signs of Creative Fatigue on Meta Ads and When to Refresh

Joshua Jackai
18 min

Published 16 November 2026 ยท Last updated and fact-checked November 2026
Short answer: Creative fatigue on Meta ads is when an ad's cost per result climbs because the same people have seen it too often. Meta flags it in Ads Manager at twice your past cost per result. Refresh on signals, not the calendar: about 24% of viewers in one study turned weary, so cap and rotate before you rebuild.
Picture a DTC founder whose best ad carried the account for six weeks. Cost per purchase crept from $40 to $68, the founder paused the ad, and the replacement batch did worse. A month later the old ad came back and beat everything. Nobody knew whether the ad had worn out, the audience had changed or the budget had simply moved. That is the creative fatigue problem in one story.
I make ad creative for DTC brands and B2B teams, and this page is the checklist I use before I tell a client an ad is tired. It is written for a DTC founder or growth lead who already runs Meta ads. Every research claim links to a peer-reviewed paper I checked through Crossref on 16 November 2026, and the platform facts link to their source. If you still need the plan for what to make, start with my Meta ads creative strategy rules. To compare new ads fairly once you refresh, use my ad creative testing steps.
Key takeaways
Meta's Delivery column shows "Creative fatigue" when cost per result is at least twice your past level and "Creative limited" when it is higher but less than double, as Jon Loomer documented (Loomer, 2025).
Schmidt and Eisend (2015) pooled repetition studies in the Journal of Advertising and found attitude toward an ad peaks at about ten exposures in experiments, and repetition effects decay over time.
Chae, Bruno and Feinberg (2019) tracked more than 12,000 users across over 400 websites: about 24% turned weary, and profiling and capping improved ad deployment by as much as 15%.
Calder and Sternthal (1980) found wearout was not stopped by tactics designed to grab more attention.
Naik, Mantrala and Sawyer (1998) showed ad quality falls during continuous running and recovers during breaks, so resting an ad can beat deleting it.
Sign | What it tells you | Key number | Source type |
|---|---|---|---|
1. Meta's Delivery label changes | Cost per result versus your own past | 2x past cost per result | Industry (Jon Loomer) |
2. Costs rise but the audience changed | Fatigue or audience drift | Result sign can flip | Peer reviewed |
3. Frequency passes the response peak | Whether repetition still helps | About 10 exposures | Peer reviewed |
4. One group turns weary first | Who to cap | About 24% weary | Peer reviewed |
5. Attention tricks stop working | Whether to swap the message | Wearout not forestalled | Peer reviewed |
6. Weak concepts tire fastest | What to build next | Creative ads wear out slower | Peer reviewed |
7. A rested ad wins again | Whether to pause instead of kill | Quality restores in breaks | Peer reviewed |
1. Meta's Delivery label changes: best for a first warning
Meta's Delivery column is the fastest fatigue check in Ads Manager. It reads "Creative fatigue" when an ad's cost per result is at least twice its past level and "Creative limited" when it is higher but under double. Jon Loomer (2025) documented both labels and noted Meta leaves out its full detection logic.
Where to look: the Delivery column at ad level in Ads Manager
Creative limited: cost per result above your past level, under 2x
Creative fatigue: cost per result at 2x your past level or more
Baseline: your own account history, not an industry benchmark
Time cost: about 2 minutes a week
The label is a cost alarm, not a diagnosis. Meta compares the ad's cost per result with your own history, so anything that pushes cost up can trip it: a new audience, a seasonal auction, a landing page change or a real case of people tiring of the ad. I treat the label as the reason to open the ad, never as the reason to kill it.
The useful habit is to write the ratio down. Divide this week's cost per result by the ad's own 28-day average. A founder whose cost per purchase moved from $40 to $68 has a ratio of 1.7, which sits in the "Creative limited" band. At $80 the ratio is 2.0 and the label turns to "Creative fatigue". Those dollar figures are an illustration, not a benchmark; the calculator below works the ratio from your own numbers.
"A high frequency, by itself, is not problematic," Jon Loomer wrote in his guide to creative fatigue on Meta ads.
Loomer's point matters for the next six signs. The label tells you cost has moved. The signs below tell you why.

2. Costs rise but the audience changed: best for ruling out a false alarm
Rising costs can come from a changed audience, not a tired ad. Braun and Schwartz (2025) showed in the Journal of Marketing that Meta-style platforms deliver each ad to a different, algorithm-picked mix of people, which can confound and even flip what a comparison tells you about the creative.
What to check: age, placement and new versus returning split this week against the ad's first two weeks
Red flag: budget moved, a new ad set launched or a placement opened
Test: compare cost per result on the same placement only
Verdict: if the audience moved, fix the targeting before the creative
Time cost: about 10 minutes
Most fatigue calls I review are audience calls in disguise. An ad that started on returning visitors and warm lookalikes gets pushed toward colder people as spend rises. Cost per result climbs, the label changes, and the ad gets blamed. The creative did not change. The people seeing it did.
Braun and Schwartz explain why the platform cannot tell you this on its own. Delivery is tuned per ad, so two ads, or the same ad in two weeks, reach different mixes of users. Your before and after numbers compare creative plus audience, not creative alone. The practical check is to hold one thing still: look at the same placement and the same age band in week one and this week. If cost per result rose inside that slice too, the ad is wearing out. If it held steady inside the slice and only the mix shifted, you have an audience problem, and new creative will inherit it.
"The estimated comparisons confound the effect of the ad content with the effect of algorithmic targeting," Braun and Schwartz wrote in the Journal of Marketing.

3. Frequency passes the response peak: best for knowing when repetition stops helping
Repetition helps until it stops. Schmidt and Eisend (2015) pooled advertising repetition studies and found attitude toward an ad peaks at about ten exposures in experiments, while recall keeps rising to at least the eighth. Spaced exposures strengthened the attitude effect, and all effects decayed over time.
Metric: frequency per person over 7 and 28 days at ad level
Lab peak: about 10 exposures for attitude (Schmidt and Eisend, 2015)
Spacing: exposures spread over time helped attitude more than bursts
Caveat: lab exposures are not feed impressions; use it as a ceiling
Time cost: about 5 minutes a week
Frequency is the most quoted fatigue number and the most misread. Ads Manager reports average frequency, which hides a spread: some people saw the ad once and a small group saw it twenty times. The average can look calm while the heavy group is already tired.
The meta-analysis gives a sensible ceiling, not a target. In controlled experiments, people liked an ad more with each viewing up to about ten, then liking flattened or fell. Recall behaved differently and was still climbing at eight exposures. For a DTC brand, that means a high frequency on a retargeting audience can still be doing memory work even after liking stops growing. It also means pacing matters. Schmidt and Eisend found spaced exposures helped attitude more than massed ones, which supports even daily budgets over sudden spend spikes.
My working rule: when 28-day frequency on a prospecting ad passes about 6 and click-through is falling at the same time, start the replacement brief. I set the margin below 10 because feed impressions are glanced at, not watched like a lab ad, and that rule is my judgement, not a study finding.
"In an experimental setting maximum attitude is reached at approximately ten exposures," Schmidt and Eisend wrote in the Journal of Advertising, from Technical University Dortmund and European University Viadrina.

4. One group turns weary first: best for deciding who to cap
Fatigue rarely hits a whole audience at once. Chae, Bruno and Feinberg (2019) studied more than 12,000 users across over 400 websites and found about 24% reached weariness, where extra exposures had a negative effect. A profiling and capping strategy improved ad deployment by as much as 15%.
Who tires first: heavy users who see the ad most often
Share weary in the study: about 24% of users
Fix: cap or exclude that group, keep the ad for everyone else
Gain in the study: up to 15% better deployment
Time cost: about 15 minutes to set up
This is the finding I use most with clients. Weariness is not the same as wearout. Wearout means each extra view adds less. Weariness means each extra view does harm, and in Chae and colleagues' data only one class of users reached it. The other 76% were still responding.
That changes the fix. Killing an ad because a quarter of its audience is weary throws away the three quarters who are not. On Meta, the closest tools are audience exclusions and campaign structure: exclude recent engagers or recent purchasers from prospecting, move heavy viewers into a retargeting ad set with different creative, and keep the original ad running for new people. Reach and frequency buying offers a direct cap where your account has it.
The model in the paper also found large differences between publishers, with the best nine times more effective than one ranked 26 places lower. On Meta, the nearest parallel is placement. Before you blame the creative, check whether one placement is carrying most of the rising cost.
"One subgroup, accounting for about 24% of the sample, evincing weariness," Chae, Bruno and Feinberg reported in the Journal of Marketing Research.

5. Attention tricks stop working: best for knowing when to swap the message
When a tired ad gets a new colour, a louder hook or a flashing sticker and still declines, the message is what wore out. Calder and Sternthal (1980) found in the Journal of Marketing Research that wearout was not forestalled by strategies designed to grab more attention.
Symptom: surface edits lift the first day, then cost per result keeps rising
Cause: viewers have already processed the argument
Fix: a new angle, offer or proof, not a new thumbnail
Keep: the format and length if they worked
Time cost: one brief, about 30 minutes
Calder and Sternthal ran repeated television commercials and watched viewers' views of both the ads and the products fall. Their information processing explanation is simple: once people have thought through what an ad says, more exposures give them nothing new to think about, so they start arguing with it. Making the ad harder to ignore does not help, because attention was never the problem.
This is where most refreshes go wrong. A team sees the fatigue label, changes the hook frame and the colour grade, and calls it a new ad. Meta may even treat it as new for a few days. The buyer does not. They recognise the same claim and the same proof, and the decline resumes.
A real refresh changes what the ad says. Swap the angle (from price to time saved), the proof (from a review to a demo), the speaker (from founder to customer) or the offer. Keep the parts that earned attention in the first place. For the question of whether a new speaker should be a creator or an AI avatar, see my comparison of AI UGC vs real UGC.
"Wearout was not forestalled by strategies designed to enhance attention," Calder and Sternthal wrote in the Journal of Marketing Research, both at Northwestern University.

6. Weak concepts tire fastest: best for building ads that last
Some ads wear out slower than others. Lehnert, Till and Carlson (2013) found in the International Journal of Advertising that highly creative ads were more liked, worked sooner and were less susceptible to wearout, although repeated exposure shrank their recall advantage.
What lasts longer: ads rated original and meaningful, not just polished
What tires fast: generic product shots and stock claims
Trade-off: the recall edge narrows with repetition
Brief test: could a competitor run this ad by swapping the logo?
Time cost: built into the brief
The cheapest fatigue fix is upstream. If every ad in the account is a variant of the same product-on-white shot with a discount line, the whole library tires at once, and refreshes only reshuffle the same idea.
Lehnert and colleagues embedded creative and control commercials in a television programme and followed viewers up a week later. The creative ads earned higher recall, wore in faster and resisted wearout better. That fits what I see in DTC accounts: a concept with a real point of view (a founder story, a strong before and after, an honest objection answered) keeps working weeks after the polished but generic version flatlines.
The brief test I use is the logo swap. If a competitor could run the ad by changing the logo, it will tire fast because there is nothing in it for the viewer to keep. My Meta ads creative strategy piece covers how to build concepts with distinct angles so a refresh has real alternatives to draw on. Plan three to five distinct concepts per batch so you are never refreshing from an empty shelf.
"Creative ads are more liked, demonstrate wear in effects more quickly and are less susceptible to wearout," Lehnert, Till and Carlson wrote in the International Journal of Advertising, from Grand Valley State, Xavier and Saint Louis universities.

7. A rested ad wins again: best for deciding whether to pause or kill
A fatigued ad is often a resting candidate, not a dead one. Naik, Mantrala and Sawyer (1998) showed in Marketing Science that ad quality falls during continuous running and restores during breaks, which can make pulsing beat steady spend. Later work found annoyance fades faster than brand memory.
Rest period: pause the ad, keep it in the library with its numbers
Restore signal: cost per result near its old level when it returns
Why it works: annoyance fades, brand memory stays
Rotation: run 2 or 3 strong concepts in turns, not one forever
Time cost: about 10 minutes a month
Naik and colleagues set out to explain why advertisers pulse when the older models said steady spending was best. Their answer was ad quality. During continuous running, wearout drags quality down. During a break, people forget, and quality recovers. Stopping when quality is low and returning once it has restored can generate more total awareness than running flat out.
Kronrod and Huber (2019) found the same direction from the viewer's side in the International Journal of Research in Marketing. People were more annoyed by a heavily advertised brand at the time, yet preferred it weeks later, but only when the product category mattered to them by then.
For a DTC account this means two habits. Never delete a past winner: pause it, label it with the date and its last cost per result, and bring it back after a rest. And rotate two or three proven concepts so each gets time off. The founder in the opening story stumbled into this. A planned rotation makes it repeatable.
"Initial annoyance with frequent ad repetition is highly susceptible to decay over time, whereas brand memory remains relatively stable," Kronrod and Huber wrote in the International Journal of Research in Marketing.
"Pause the ad that tired, keep its numbers, and bring it back after a rest," Joshua Jackai, founder of Demand Designer, wrote in this article.

How often to refresh ad creative on Meta
Refresh on signals, not on a fixed calendar. Start a replacement brief when the Delivery label reads "Creative limited", frequency climbs while click-through falls, and the rise holds inside one placement. Schmidt and Eisend (2015) found repetition effects decay with time, so pacing and rest matter as much as new ads.
A calendar rule fails both ways. A small retargeting audience can tire in days, while a broad prospecting ad on a large audience can hold for months. What sets the pace is how fast each person's frequency climbs, and that depends on budget, audience size and how much of the audience Meta actually reaches. The calculator below turns those three inputs into an estimate of the weeks until the average reached person hits a frequency you choose.
Comparison matrix: which sign points to which fix
Use this matrix to match what you see in Ads Manager to the fix that fits. Each row names a symptom, the sign it belongs to, what to change and the source behind it. On the live page you can filter by symptom type; the table holds every row either way.
Symptom in your account | Sign | What to change | Source |
|---|---|---|---|
Delivery column reads Creative limited | 1. Delivery label | Check the cost per result ratio | Jon Loomer, 2025 |
Cost up after a budget or audience change | 2. Audience drift | Fix targeting before creative | Braun and Schwartz, 2025 |
Frequency climbing, click-through falling | 3. Response peak | Start the replacement brief | Schmidt and Eisend, 2015 |
Heavy viewers driving the cost rise | 4. Weary group | Cap or exclude that group | Chae, Bruno and Feinberg, 2019 |
New hook or colour lifted only a day | 5. Message worn out | New angle, proof, speaker or offer | Calder and Sternthal, 1980 |
Whole library tires at the same time | 6. Weak concepts | Brief distinct concepts | Lehnert, Till and Carlson, 2013 |
Old winner paused for weeks | 7. Rested ad | Bring it back and compare | Naik, Mantrala and Sawyer, 1998 |
Creative refresh calculator: weeks until your ad needs a rest
The calculator estimates weeks until the average reached person sees your ad a set number of times, and reads your cost per result ratio against Meta's labels. With a $3,000 weekly budget, $12 CPM, a 400,000 audience and 40% reached, frequency hits 6 in about 3.8 weeks, below the peak Schmidt and Eisend (2015) found.
How to use it with your own numbers:
Weekly budget and CPM: from the ad set; $3,000 and $12 are placeholders
Audience size: the estimated audience shown for the ad set
Share reached: the ad set's reach divided by audience size; 40% is a placeholder
Frequency to plan for: 6 is my working rule, under the lab peak of about 10
Cost per result now and past: this week against the ad's own 28-day average
Worked example: $3,000 at a $12 CPM buys 250,000 impressions a week. Reaching 40% of 400,000 people is 160,000 people, so frequency grows by about 1.56 a week and passes 6 in about 3.8 weeks. A cost per result of $68 against a $40 past level is a ratio of 1.7, inside the Creative limited band.
Creative fatigue on Facebook ads versus Instagram placements
Creative fatigue on Facebook ads and on Instagram comes from the same Ads Manager system and the same label, but placements tire at different speeds. Chae, Bruno and Feinberg (2019) found the best publisher nine times more effective than one 26 places lower, so check cost per result by placement before you call fatigue.
Placement | What usually tires first | First check | Typical fix |
|---|---|---|---|
Facebook Feed | The opening line and headline | Click-through by week | New angle in the first line |
Instagram Feed | The first frame | Hold rate and click-through | New opening frame or speaker |
Stories and Reels | The first 3 seconds | 3-second hold rate | New hook concept, same offer |
Retargeting ad sets | The whole ad, fast, on a small audience | Frequency over 7 days | Rotate 2 or 3 concepts |
Break the report down by placement in Ads Manager and read cost per result for each row. If one placement carries most of the rise, cut or rebalance it before you brief new creative.
Frequently asked questions
What is creative fatigue on Meta ads?
Creative fatigue on Meta ads is the rise in an ad's cost per result after the same people have seen it too often. Meta's Delivery column labels it "Creative fatigue" when cost per result reaches twice the ad's past level, and "Creative limited" below that.
How often should you refresh ad creative?
Refresh when the signals say so, not on a set calendar. Start a new brief when the Creative limited label appears, frequency climbs while click-through falls, and the rise holds inside one placement. Small retargeting audiences tire in days; broad prospecting audiences can hold for months.
What frequency is too high for Meta ads?
No single frequency is too high, as Jon Loomer notes. A meta-analysis by Schmidt and Eisend (2015) found attitude toward an ad peaks at about ten exposures in experiments. I start a replacement brief near 6 when click-through is also falling at that point.
Should you pause or delete a fatigued ad?
Pause it. Naik, Mantrala and Sawyer (1998) showed ad quality restores during breaks, and Kronrod and Huber (2019) found annoyance fades faster than brand memory. Keep the paused ad with its date and last cost per result, then test it again after a rest.
Does a new thumbnail fix creative fatigue?
Rarely for long. Calder and Sternthal (1980) found wearout was not stopped by tactics designed to grab more attention. A new thumbnail may lift results for a day or two. A real refresh changes the angle, proof, speaker or offer, and keeps the format that worked.
How I handle creative fatigue for DTC brands
I run fatigue checks as part of a monthly creative system: a Monday read of the Delivery labels, cost per result by placement, frequency on each live ad and a refresh brief ready before an ad tires. Pricing is fixed: Conversion Creatives runs $999 or $2,449 a month.
Each batch arrives with three to five distinct concepts, so a refresh swaps the message rather than the colour. Past winners stay in a rested library with their dates and numbers, and they come back in rotation. New ads go through the ad creative testing steps before they carry the budget.
If you want a quick weekly way to decide which ads to keep, rest or kill, start with my free Kill Sheet: five signals, three verdicts and ten minutes every Monday.
Sources
Peer-reviewed research
Braun, M., and Schwartz, E. M. (2025). Where A/B testing goes wrong: How divergent delivery affects what online experiments cannot (and can) tell you about how customers respond to advertising. Journal of Marketing, 89(2), 71 to 95.
Calder, B. J., and Sternthal, B. (1980). Television commercial wearout: An information processing view. Journal of Marketing Research, 17(2), 173 to 186.
Chae, I., Bruno, H. A., and Feinberg, F. M. (2019). Wearout or weariness? Measuring potential negative consequences of online ad volume and placement on website visits. Journal of Marketing Research, 56(1), 57 to 75.
Kronrod, A., and Huber, J. (2019). Ad wearout wearout: How time can reverse the negative effect of frequent advertising repetition on brand preference. International Journal of Research in Marketing, 36(2), 306 to 324.
Lehnert, K., Till, B. D., and Carlson, B. D. (2013). Advertising creativity and repetition. International Journal of Advertising, 32(2), 211 to 231.
Naik, P. A., Mantrala, M. K., and Sawyer, A. G. (1998). Planning media schedules in the presence of dynamic advertising quality. Marketing Science, 17(3), 214 to 235.
Schmidt, S., and Eisend, M. (2015). Advertising repetition: A meta-analysis on effective frequency in advertising. Journal of Advertising, 44(4), 415 to 428.
Industry and vendor sources (not peer reviewed)
Loomer, J. Creative fatigue: What it is and how to prevent it. Published 15 September 2025. Read 16 November 2026.
Free workbook
The Kill Sheet
Know when to kill an ad and when to let it run, with thresholds for the five signals that matter.
Five signals, three verdicts, ten minutes every Monday. Free.







