Facebook ads frequency is the average number of times each person saw your ad: impressions divided by reach, both estimated by Meta. You can cap it only with the Reach and ThruPlay goals or with reservation buying, so a Leads campaign has no cap. In lead gen, treat frequency as a flag and let each creative’s CPL decide.
That’s less of a limitation than it sounds. Frequency is an average across everyone the ad reached, it climbs for reasons that have nothing to do with the creative, and in a lead-gen account the expensive mistake is reacting to the average instead of to the leads. This guide covers how Meta defines frequency, what a good level looks like, where caps exist, and how to read frequency next to the CPL of each creative.
What is frequency in Facebook ads?
Meta’s Marketing API reference for ads insights defines frequency as “the average number of times each person saw your ad” and reach as “the number of Accounts Center accounts that saw your ads at least once”. It adds that reach “is different from impressions, which may include multiple views of your ads by the same Accounts Center accounts”, and labels both metrics as estimated. Put the two together and frequency is impressions divided by reach: total views over the people who saw them.
Three consequences of that formula change how you read it.
It’s an average. A frequency of 3 can mean everyone saw the ad three times, or most people saw it once while a small group saw it a dozen times. The second case wears out a creative for that small group long before the average looks alarming.
It depends on the date range. Reach counts each person once in the period, so a 4-week frequency isn’t the average of four weekly frequencies. People who saw the ad in several weeks count once in the 4-week reach and every time in the impressions. Example:
| Period | Impressions | Reach | Frequency |
|---|---|---|---|
| Week 1 | 40,000 | 20,000 | 2.0 |
| Week 2 | 40,000 | 20,000 | 2.0 |
| Week 3 | 40,000 | 20,000 | 2.0 |
| Week 4 | 40,000 | 20,000 | 2.0 |
| Weeks 1-4 | 160,000 | 45,000 | 3.6 |
Every week reads 2.0, and the month reads 3.6, because many of the same people came back each week. Compare 7-day frequency with 7-day frequency, never a lifetime number with last week’s.
It depends on the level. An ad set’s reach counts a person once even if they saw two of its ads, so the ad set’s frequency can be higher than any of its ads’. Example: Ad A has 30,000 impressions on 15,000 people, Ad B 20,000 on 10,000. Both read 2.0. If 7,000 of Ad B’s people also saw Ad A, the ad set reached 18,000 people with 50,000 impressions: a frequency of 2.8. Fatigue happens to a creative, so read frequency per ad.
What is a good frequency for Facebook ads?
There isn’t one number. Meta’s developer reference defines the metric without setting any target for it, and a Leads campaign has no cap to aim at either. Any fixed threshold would be wrong for most accounts, because frequency depends on budget and audience size as much as on the ad.
The reason is arithmetic. Say you spend €1,000 a week at a €10 CPM: that buys 100,000 impressions. A plumber serving one town, where delivery finds about 30,000 people, ends the week at a frequency of at least 3.3. The same budget spread over 80,000 people across a region reads about 1.25. Same creative, same spend, a frequency nearly three times higher, and neither number says anything yet about whether the ad still works.
So a good frequency is one at which the ad’s link CTR and CPL are holding. Two practical defaults, as starting points rather than rules:
- Flag, don’t act. When an ad’s 7-day frequency is up by half on its first two full weeks, put it on the review list.
- Let the outcome decide. A high frequency with a steady CTR and a steady CPL means the audience is still responding. Leave it alone.
| Context | What frequency usually does | What to watch |
|---|---|---|
| Local business, small service area | climbs fast | CTR and CPL against the ad’s own baseline |
| Broad audience, national or regional | stays low for a long time | CPL; frequency is rarely the issue |
| Retargeting past visitors or leads | high by design: the list is small | cost per qualified lead, and list size |
| Budget raised on the same audience | rises with the budget | marginal CPL of the new spend |
Can you set a frequency cap on Meta ads?
Only in some setups, and a Leads campaign isn’t one of them. Meta’s Marketing API documentation for ad sets says writes to frequency_control_specs, the field that holds a cap, “are only available in ad sets where REACH and THRUPLAY are the performance goal”. A cap there is a maximum number of impressions per person over an interval, with both values between 1 and 90, and the Reach goal is defined as optimizing “to reach the most unique users for each day or interval specified”.
The other place frequency is planned is reservation buying, which Meta’s developer documentation files under reach and frequency. It lets you “plan and buy your campaigns with a fixed cost”: you reserve a prediction to “fix your CPM and have predictable reach”, with a frequency cap set in the prediction. It’s built for big audiences. The documented default minimum reach is 200,000 Accounts Center accounts, target audiences need at least 300,000, and it’s “available for certain ad accounts”.
| Setup | Frequency cap | What it optimizes for | Fit for lead gen |
|---|---|---|---|
| Leads campaign, any performance goal | none | leads, or conversion leads | the campaign you run; manage frequency through audience, budget and creative |
| Reach performance goal | yes: impressions per person over 1 to 90 days | unique reach | not for leads |
| ThruPlay performance goal | yes, through the same API field | video plays | not for leads |
| Reservation buying | yes, planned in advance at a fixed CPM | predictable reach on audiences of 300,000+ | awareness planning, not lead optimization |
Moving lead gen into a Reach campaign to get a cap trades the thing you want for the thing you can control. The Reach goal is built to reach as many different people as possible, not the people likely to fill in a form (Facebook ads objectives explains why the objective decides who sees the ad).
How to control frequency in a Leads campaign
Without a cap, you change the inputs that drive it:
- Widen the audience. Frequency is impressions over reach, so a bigger reachable audience lowers it at the same budget. For a local business that means a wider radius before anything else.
- Stop competing with yourself. Several ad sets aimed at the same people pile impressions on the same reach. Merge them; Facebook ads account structure covers how many ad sets a budget can feed.
- Exclude people who already became leads with a custom audience, so the budget goes to people who haven’t converted. The Advantage+ leads campaign guide covers which audience edits keep Advantage+ on.
- Feed new creative. Fresh concepts give the same people something new to respond to, which is the actual cure for fatigue.
- Slow the budget on an audience that’s already saturated, instead of pushing more impressions at it. How to scale Facebook ads has the step sizes.
Frequency tells you how often people saw an ad; only the CPL of each creative tells you whether it’s still paying. DropCPL syncs spend per ad from the Meta Ads API and counts leads on your own landing pages, so every creative × landing page pair has a real CPL to read next to Meta’s frequency. Start free →
Why Facebook ads frequency alone doesn’t diagnose fatigue
Frequency measures exposure. Fatigue is a change in response: the same people stop clicking, so each lead costs more. The two often move together, but frequency on its own gets it wrong in both directions:
- It rises without fatigue. A budget increase, a smaller audience or a retargeting list all push frequency up while the audience keeps responding.
- Fatigue starts below any threshold. A weak hook can lose its audience at a frequency of 1.8, and the average hides the minority who’ve seen it a dozen times.
- It’s not the outcome. In lead gen the outcome is a lead, and the decision is whether the next euro on this creative buys a lead under your target.
Example, two ads in the same ad set, week 1 against week 4:
| Ad | 7-day frequency | Link CTR | Landing page conv. rate | Real CPL |
|---|---|---|---|---|
| Ad A | 1.4 → 2.6 | 1.20% → 1.15% | 5.0% → 5.1% | €30 → €31 |
| Ad B | 1.3 → 1.8 | 1.10% → 0.70% | 5.2% → 5.0% | €28 → €46 |
A rule like “pause anything above 2.5” cuts Ad A, which is still on its original CPL, and keeps Ad B, whose CTR fell by more than a third while its CPL climbed 64%. The landing page conversion rate is flat for both, so whatever happened to Ad B happened before the click. That’s the pattern creative fatigue in Meta ads teaches you to confirm before you refresh anything.
How to read Facebook ads frequency next to CPL per creative
Once a week, per ad ID:
- Pull the last 7 days: frequency, link CTR, landing page conversion rate, leads you received and real CPL. Lifetime frequency blends every week since launch, so it can’t tell you about this one.
- Compare each ad with its own first two weeks, never with the account average.
- Use frequency and CTR to flag. They rest on thousands of impressions and hundreds of clicks, so they move first.
- Use CPL to decide. It rests on a few dozen leads at best, so give it two-week windows when an ad gets fewer than about 20 leads a week. The Meta ads creative testing framework shows how noisy small lead counts are.
- Check whether frequency is rising on one ad or on all of them. One ad points at the creative; every ad at once points at the audience or the budget.
| Frequency | Link CTR | Real CPL | Usual reading | What to do |
|---|---|---|---|---|
| up | flat | flat | the audience is still responding | nothing yet |
| up | down | up | creative fatigue | new hook or format; keep the original running while the new ad ramps up |
| up on every ad | down on every ad | up | audience too small for the budget | widen the audience or slow the budget |
| flat | down | up | the creative is losing appeal, or the auction got pricier | refresh the creative; check CPM on other ads |
| flat | flat | up | the problem is after the click | check the landing page, the form and the tracking |
The last row is the one frequency can never catch. When CTR holds and CPL rises, the creative is doing its job and the leak sits on the landing page or in the tracking, which is why the CPL you read should come from leads you counted, not from Meta’s attributed count (real CPL vs platform CPL).