An ad that sold well for three weeks suddenly costs twice as much per sale. Nothing changed in your settings, the budget is the same, the product is the same. In many cases the cause is Meta ad fatigue: the same people have seen the same ad so many times that they scroll past it, and Meta has to work harder (and charge you more) to find anyone new who will still click. This guide explains how to read frequency, the early signs that an ad is getting tired, how much frequency is too much for different audiences, and what to do about it without wrecking a campaign that still works.
In short: Frequency is how many times, on average, each person saw your ad. Fatigue shows up as rising frequency together with falling click-through and a rising cost per sale. Small audiences tire fastest. The fix is usually new creative, not a new campaign: add fresh ads to the same ad set, widen small audiences, and set a simple rule so you notice before your margin disappears.
What ad fatigue means on Meta
Ad fatigue happens when your audience has seen an ad often enough that it stops getting a reaction. The first time someone sees it, it's new. The fifth time, it's background. By the tenth, some people get annoyed, hide it or leave a grumpy comment.
On Facebook and Instagram this matters more than on most channels, because feeds move fast and your ad competes with friends' photos, videos and other brands. People don't consciously decide to ignore you; their thumb just keeps moving.
There are really two kinds of fatigue:
- Creative fatigue: the ad itself (photo, video, opening line) is worn out. A new ad to the same people can work again.
- Audience fatigue: the pool of people you're reaching is too small or already saturated. Even new ads get diminishing returns because you keep talking to the same few thousand people.
Knowing which one you have tells you what to fix.
How frequency is calculated
Meta defines frequency as the average number of times each person saw your ad, worked out as impressions divided by reach. Meta's own help article on frequency notes it's an estimate and should be watched together with your results.
A quick example with made-up numbers:
- Impressions in the last 7 days: 42,000
- Reach in the last 7 days: 14,000 people
- Frequency: 42,000 ÷ 14,000 = 3.0
Two things people often miss:
- Frequency is an average. A frequency of 3 can mean most people saw the ad twice while a smaller group saw it eight or ten times.
- The date range changes the number. Lifetime frequency on a campaign that has run for three months will look high even if this week is fine. Look at the last 7 days to see what's happening now.
To see it, open Ads Manager, go to the ads level, and add the Frequency, Reach, CTR (link click-through rate) and Cost per result columns. Then set the date range to the last 7 days and compare it with the 7 days before.
The warning signs of a tired ad
Frequency on its own doesn't prove anything. An ad at frequency 4 that still sells below your break-even cost per sale is fine. You're looking for a pattern, usually over one to two weeks:
| Signal | What it looks like | What it suggests |
|---|---|---|
| Frequency rising week over week | 1.6, then 2.4, then 3.3 | You're reaching mostly the same people |
| Link click-through falling | 1.4%, then 1.0%, then 0.7% | The ad is getting ignored |
| Cost per click rising | $0.60, then $0.85, then $1.20 | Meta has to show it more times to get a click |
| Cost per sale rising | Moves toward or above your break-even | Fatigue is now costing you money |
| Negative feedback | Comments like "I keep seeing this", hides | People are actively annoyed |
When the first three move together, fatigue is the likely cause. If cost per sale jumps but frequency is flat, look elsewhere: tracking problems, a price change, stock, a competitor's sale, or an auction-wide spike. Our guide on why cost per sale suddenly rises on Meta walks through those other causes, and the Meta ads metrics that matter explains each column.
How much frequency is too much?
There's no single safe number, and anyone who gives you one without asking about your audience is guessing. What you can do is use rough ranges as a prompt to look closer, and then trust your own cost per sale.
| Audience | Typical size | When to start watching closely (7-day frequency) |
|---|---|---|
| Broad or cold audience | Hundreds of thousands or more | Above roughly 2 to 3, especially if CTR is falling |
| Interest or lookalike audience | Tens of thousands to a few hundred thousand | Above roughly 3 |
| Retargeting (site visitors, cart abandoners) | A few hundred to a few thousand | Higher is normal; watch for falling CTR and complaints rather than the number itself |
| Customer list for win-back | Often small | Rises fast; rotate ads often and keep budgets small |
These are rules of thumb, not Meta rules. People who were on your product page yesterday expect to see you again, so retargeting tolerates more repetition. A cold audience owes you nothing and tires quickly. Our audience targeting guide explains how audience size affects delivery.
How to fix ad fatigue, step by step
Work through these in order. The first steps are cheap and keep your campaign's learning intact; the later ones are bigger changes.
1. Confirm it's fatigue
Compare the last 7 days with the previous 7 at the ad level. Rising frequency plus falling click-through plus rising cost per sale: fatigue. Only cost per sale moving: check tracking and the site first.
2. Add new ads to the same ad set
Instead of building a new campaign, add two or three fresh ads to the ad set that's tiring. That keeps the ad set's history and avoids starting from zero. Change something people actually notice:
- A new first frame or photo. The image is what stops the scroll.
- A new angle. If the old ad talked about price, try a benefit, a use case, or an objection ("Will it fit my small kitchen?").
- A new format. Photo to short video, single image to carousel, studio shot to someone using the product.
Small tweaks like a new headline on the same photo rarely reset fatigue, because people recognize the picture before they read anything. For ideas, see Meta ad creative that sells.
3. Pause the worn-out ad, not the ad set
Once the new ads are delivering, pause the tired one if its cost per sale is above your break-even. Leave anything that's still profitable running, even at a high frequency.
4. Widen a small audience
If fatigue keeps coming back within days, the audience is probably too small for your budget. Options: widen the age range or locations, combine similar interests, use a broader audience and let Meta find buyers, or lower the budget so you don't burn through the same people.
5. Exclude people who already bought
Showing a "buy now" ad to someone who bought last week is pure waste and drives up frequency. Exclude recent purchasers (with a customer list or a purchase audience from your pixel) from campaigns aimed at new buyers.
6. Set an alert or an automated rule
Ads Manager lets you create automated rules, and Meta has an example rule to pause an ad set when its frequency is above a number you choose. A rule that only notifies you is often safer than one that pauses automatically, because a high-frequency ad set that still sells profitably shouldn't be switched off.
Worked example: the planter that stopped selling
A hypothetical store sells a $32 ceramic planter and keeps $15 per sale before ads, so $15 is its break-even cost per sale. It runs one ad set at $25 a day to an interest audience of about 120,000 people.
| Week | Frequency (7 days) | Link CTR | Cost per sale | Status |
|---|---|---|---|---|
| 1 | 1.5 | 1.6% | $9.80 | Good |
| 2 | 2.2 | 1.3% | $11.40 | Fine |
| 3 | 3.1 | 0.9% | $14.60 | Close to break-even |
| 4 | 3.9 | 0.6% | $19.20 | Losing money |
Week 3 is when to act, not week 4. The store adds two new ads in week 3: a short video of someone repotting a plant, and a photo of three planters on a windowsill with a new opening line. It keeps the ad set and budget the same, and pauses the original ad when its cost per sale crosses $15.
What can happen next:
- What if the new ads bring cost per sale back down? It was creative fatigue. Plan to refresh ads every few weeks at this budget.
- What if the new ads start well but tire within a week? It's the audience. Widen it or lower the budget.
- What if the new ads never get delivery? Meta keeps favoring the old ad. Pause the old one so the new ones get a chance, or test them in a separate ad set (see how to A/B test Meta ads).
- What if you want to scale while this is happening? Raising the budget on a tired audience makes fatigue worse. Fix the creative first, then read our guide on how to scale Meta ads.
How to prevent ad fatigue
- Launch with more than one ad. Two to four ads per ad set gives Meta options and spreads impressions.
- Keep a creative pipeline. Shoot a few extra photos and short clips every time you photograph products, so you always have a fresh ad ready.
- Match budget to audience size. A big budget on a small audience burns it out fast.
- Look at 7-day frequency every week. Not lifetime, not today.
- Separate retargeting from prospecting. They tire at different speeds and need different messages; our retargeting guide shows how to split them.
How Meta Ads Copilot Max helps with ad fatigue
Meta Ads Copilot Max is a Chrome tool (a Tampermonkey script plus a web app at adscopilotmax.com) that reads your campaigns in Ads Manager and explains them in plain words. For fatigue specifically:
- High-frequency and low click-through alerts. Copilot scores the account from 0 to 100 and raises alerts when frequency climbs or click-through drops, so you see the week 3 problem before week 4.
- Judged against your margin. You enter your profit per sale and Copilot compares each campaign's cost per sale with it, so you don't pause something that's still making money just because frequency is high.
- Fresh ads from your product page. Open the product page and Copilot writes three new versions of the ad, and can make a short video if you connect a video model.
- Fix with Claude. Tick the alerts and Claude in Chrome (required) does the clicking in your browser. Changes stay paused until you approve them.
You can try Copilot free for 7 days. Start your free 7-day trial and get your key by email.
Frequently asked questions
What is a good frequency for Meta ads?
There's no single good number. For cold audiences, many advertisers start watching closely once 7-day frequency goes above about 2 or 3. Retargeting audiences often run higher without problems. The real test is your cost per sale: if it stays below your break-even, the frequency is fine.
How long does it take for a Meta ad to get fatigued?
It depends on budget and audience size. A large budget on a small audience can tire an ad in days; a modest budget on a broad audience can run for weeks or months. Watch the weekly trend in frequency and click-through rather than counting days.
Should I duplicate the ad set or campaign to fix fatigue?
Usually not as a first step. Duplicating starts a new learning period and can compete with the original for the same people. Adding new ads to the existing ad set is a gentler fix. Duplicate only if you also want to change the audience or the optimization goal.
Can I set a frequency cap on Meta ads?
Frequency caps are available with some buying types and objectives, such as reach and frequency buying and the reach objective. For most sales campaigns there's no direct cap, so you control frequency through audience size, budget and fresh creative, and you can use an automated rule to notify you when frequency gets high.
Does high frequency always mean I should pause the ad?
No. High frequency is a reason to look, not a reason to stop. If the ad still sells below your break-even cost per sale, keep it running and add new ads alongside it. Pause when cost per sale goes over your limit and click-through keeps falling.
