Ad fatigue on Meta happens when the same people see the same creative too many times, so clicks and conversions drop and cost per result rises. Meta’s own research found response falls with every repeat view. The fix is not a new audience or a bigger budget. It is new, genuinely different creative added before performance collapses.
- Meta found the average ad impression goes to someone who already saw that creative 4.2 times in the past 30 days (Analytics at Meta, 2023).
- By about the fourth repeat exposure, the likelihood of a conversion drops by roughly 45%, according to the same study.
- Frequency is an average and hides fatigue. Watch CTR, CPM and cost per result trends together, ad by ad.
- Refresh on a schedule tied to spend, and make new ads look different, not just reworded.
What is ad fatigue on Meta?
Ad fatigue, which Meta calls creative fatigue, is what happens when people see the same creative over and over. The first time, your ad is new. By the fifth time, it is wallpaper. People scroll past, click less, and the cost of every lead or sale goes up.
Meta’s analytics team defines it at the level of the creative, meaning the visuals people actually see, not the ad. That detail matters. If you run the same video in three campaigns, people get tired of it across all three. Duplicating an ad does not reset anything.
The same study found that over 19% of impressions went to people who had seen that creative more than five times. Repetition was most common in campaigns optimized for link clicks and website conversions. That is exactly how most small businesses run Meta ads.
Why do Facebook ads stop working after two weeks?
There is no fixed expiry date. Two weeks is just a common pattern for small and mid-size budgets. Here is why it happens.
When you launch, Meta shows your ad to the people it thinks are most likely to act. Those people respond, and results look great. As days go by, those best prospects have seen the ad several times. Meta has to reach further out to people who are less interested, or keep showing it to the same people. Either way, response drops.
Meta’s researchers also tested a popular belief: that an ad needs a few views to “warm up” before it works. For direct response campaigns, they found no evidence of it. Clicks and conversions got steadily more expensive with each repeat view. Repetition still has value, but it works best spread across many different-looking ads.
Fatigue vs audience saturation
These are related but not the same. Creative fatigue means people are tired of this ad, and a fresh ad would win them back. Audience saturation means you have reached most of the people in a small audience, and even new ads will struggle. A local business targeting one small town hits saturation much faster than a national ecommerce brand.
| Creative fatigue | Audience saturation | |
|---|---|---|
| What is worn out | The ad itself | The pool of people |
| Typical sign | One ad declines while new ads in the same ad set do fine | Every ad declines, reach stops growing |
| Main fix | New, different creative | Wider geography, broader targeting or a new offer |
| Who sees it most | Accounts with few ads running for months | Local businesses with tight radius targeting |
What frequency is too high on Facebook ads?
This is the question everyone asks, and the honest answer is that frequency is a weak signal on its own. Meta’s own analysts say they do not recommend it for measuring creative fatigue, for two reasons.
- Frequency is reported per ad, ad set or campaign. Fatigue happens per creative, and the same creative can run in many places.
- Frequency is an average. If half your audience saw an ad once and the other half saw it nine times, frequency reads 5, but the second group is already fatigued.
That said, frequency is still useful as a trend. A cold prospecting ad set whose 7-day frequency keeps climbing while CTR falls is telling you something. Retargeting audiences are small, so higher frequency there is normal. Compare an ad set to its own history, not to a number from a blog post.
How do you diagnose ad fatigue in Ads Manager?
Fatigue rarely shows up in one metric. It shows up as a pattern across several, usually in this order.
- 1Break down your results by ad, not by campaign. Fatigue is an ad-level problem and a healthy ad can hide a dying one.
- 2Compare each ad’s last 7 days to its first 7 days. Look at CTR (link click-through rate), CPM, cost per result and conversion rate.
- 3Check the Delivery column. Meta flags some ads as “Creative limited” when cost per result is higher than your past ads, and “Creative fatigue” when it is about twice as high or more.
- 4Look at frequency for the same window. Rising frequency plus falling CTR is the classic fatigue signature.
- 5Rule out other causes: a broken landing page, a price change, a tracking problem, or a seasonal dip that hits every ad at once.
- 6If only older ads are slipping while newer ads hold steady, it is fatigue. If everything drops together, look at the audience, the offer or the season.
| Signal | What it usually means | What to do |
|---|---|---|
| CTR down, frequency up, CPM flat | Classic creative fatigue | Add new creative to the ad set |
| CPM up, CTR stable | More competition in the auction, often seasonal | Hold steady, check again in a week |
| CTR stable, conversion rate down | Landing page, offer or tracking issue | Test the page and your events, not the ad |
| All ads drop at once, reach flat | Audience saturation | Widen the audience or location |
| Cost per result 2x past ads | Meta “Creative fatigue” status | Replace or supplement the ad now |
Does adding new creative actually fix it?
Yes, and Meta tested this directly. Its analytics team ran a two-phase split test across roughly 26,000 cases. After a week, they added a new creative to some fatigued ad sets and not to others. The more fatigued the original creative was, the bigger the improvement from adding a new one.
Note what they did: they added new creative to the existing ad set. They did not kill the ad set and start over. That keeps the learning Meta has built up and lets the new ad take over delivery gradually.
Why does creative variety matter more now?
Meta’s delivery system has changed. Its Andromeda retrieval engine, rolled out in late 2024, was built to handle a much bigger pool of ads and match each one to the right person. With broad targeting and Advantage+ now the default, the creative itself does a lot of the targeting.
More ads in the system means more competition for attention, and less patience from users. It also means near-copies do not help much. Five versions of the same photo with different headlines look like one ad to the person scrolling. They tire of all five together.
What counts as a genuinely new ad
| Change | Counts as new creative? | Notes |
|---|---|---|
| New headline or primary text only | Barely | Fine for testing copy, does little for fatigue |
| Same image, new color or crop | No | Looks the same in the feed |
| New hook in the first 3 seconds of a video | Partly | A cheap way to extend a winning video |
| Different format: static to video, or carousel | Yes | Reaches people who ignore your current format |
| New angle: price, proof, problem, testimonial | Yes | The strongest refresh |
| New person on camera or real customer footage | Yes | Especially strong for local services |
How often should you refresh Meta ad creative?
Tie the cadence to spend, not the calendar. The more you spend against the same audience, the faster people see your ads repeatedly. These are starting points we use, not rules from Meta. Adjust them to what your own data shows.
| Monthly Meta spend | Suggested cadence | New ads per refresh |
|---|---|---|
| Under $3,000 | Every 3 to 4 weeks | 2 to 3 new ads |
| $3,000 to $15,000 | Every 2 weeks | 3 to 5 new ads |
| $15,000 to $50,000 | Weekly | 5 or more, across 2+ angles |
| Local service in a small area | Every 2 to 3 weeks, regardless of spend | Rotate offers as well as visuals |
A simple refresh routine
- 1Every Monday, review ads by last 7 days vs prior 7 days. Flag any ad with falling CTR and rising cost per result.
- 2Keep a backlog of at least 3 ready-to-launch ads so you never scramble.
- 3Add new ads to the existing ad set. Do not duplicate the campaign or restart from scratch.
- 4Leave older ads on while the new ones ramp up. Turn them off only once the new ads carry the spend.
- 5Note which angles last longest. Build the next batch around what held up, not only what started strong.
How do you prevent ad fatigue in the first place?
- Launch with variety: at least 3 to 5 meaningfully different ads per ad set.
- Mix formats: short video, static images and carousels reach people differently.
- Use real footage: customers, staff and job sites age better than stock photos.
- Consolidate: fewer ad sets with more ads beat many ad sets each running one ad.
- Exclude recent converters from prospecting so you are not paying to show ads to existing customers.
- Track the right outcome: if Meta optimizes for cheap clicks, fatigue hides behind a low cost per click while real results fall.
Ad fatigue is not a sign that Meta ads stopped working for you. It is a normal part of running them. Businesses that plan for it with a steady creative pipeline keep costs stable. Businesses that wait for a crash end up rebuilding every month.
Frequently asked questions
Compare each ad’s last 7 days to its first 7 days. Falling CTR, rising frequency and higher cost per result, while newer ads stay healthy, point to fatigue. Meta may also show a “Creative fatigue” status in the Delivery column.
There is no single number. Frequency is an average and can hide people who saw your ad many times. Track it as a trend for each ad set alongside CTR and cost per result, and expect higher numbers in small retargeting audiences.
Neither as a first step. Add new creative to the same ad set and let it take over delivery. Duplicating shows people the same creative again, so it does not fix fatigue.
For most small business accounts, 3 to 6 genuinely different ads per ad set is a solid start. What matters most is that they differ in angle and format, not just in headline.
Often, yes. After a break of a few months, many people have forgotten it and new people have entered your audience. Updating the hook or first frame usually helps it perform again.
