What is a good ad frequency, and when does it become bad

There is no single good ad frequency. Meta's own help center says frequency may average 1 to 2 per ad set or be much higher, depending on budget, audience size and schedule. What works in practice: on cold audiences, a 7-day frequency of 1.2 to 2.5 is comfortable and anything above 3 deserves a look; in retargeting, 3 to 6 is normal. Frequency only becomes bad when it climbs while CTR falls and cost per result rises. The number alone decides nothing.

If you need the definition first (impressions divided by reach, and why you can't add it up across campaigns), it's in the ad frequency glossary entry. This article answers a different question: which number to accept, which window to read and what to do once you've gone too far.

What Meta officially says

It helps to separate what Meta has published from what comes from running accounts. Meta's side first:

• The metric is estimated. Meta's frequency help page says it is calculated as impressions divided by reach, using sampled data. A tenth of a point between two weeks is noise.
• 1 to 2 on average, or much more. Budget, audience size and schedule drive it. Meta sets no ceiling.
• What to watch alongside it. Meta's advice is to monitor frequency together with results and ad relevance diagnostics. If performance drops as frequency rises, the audience may be fatigued, and the suggestion is to change the creative or the targeting.
• A default cap exists only in reservation. In the reservation glossary (the former "reach and frequency" buying type, renamed in October 2023), the default cap is 2 impressions every 7 days per person, and you can change it. Reservation is built for awareness and engagement campaigns that reach more than 200,000 people.

Meta also published a study, Effective Frequency from Facebook IQ, in July 2016. Across 11 brand campaigns, a cap of 1 impression per week captured up to 80% of the potential ad recall lift, and 2 per week captured 95% of the purchase intent lift. The same study says there is no magic frequency, and that new brands, short campaigns and peak seasons call for more repetition.

Two caveats: the study is from 2016 and measured brand campaigns, not leads or sales. Use it for the logic, not as a target for a conversion campaign.

What Meta does not publish is a frequency range for lead or sales campaigns bought at auction. From here on, this is practice.

The ranges I use, by audience

What follows comes from more than eight years of running ads for over 35 companies. It is not a Meta rule, and your account may behave differently. All ranges are 7-day frequency, read at the ad set level.

Retargeting tolerates more because the person already visited or engaged, and the list is small by definition. Repetition is the point. The limit shows up as irritation, and handling it is covered in retargeting that does not burn budget.

Prospecting tolerates less because every repeat impression is one that didn't go to someone new. On a cold audience, high frequency nearly always means the audience is too small for the budget. How the audience temperatures differ is in cold, warm and hot audiences.

The window changes the number: read 7 days, not lifetime

The most common reporting mistake I see is judging lifetime frequency. It only goes up. A campaign that has run for 60 days will always show higher cumulative frequency than any single week, even if nothing is wrong.

An example. Say a lead campaign in Texas has been live for 45 days:

Now the opposite: lifetime frequency of 3.1 on a two-week campaign, but the last 7 days show 4.4 against 2.2 the week before. Lifetime looks fine; the week shows repetition doubling. That is the case that needs action.

My rule: 7-day frequency, ad set level, compared with the previous 7 days. A campaign blends ad sets with different audiences, and the account blends prospecting with retargeting. At both levels the average hides the problem.

Signs that frequency has become a problem

High frequency with stable results is not a problem. Use this checklist in the same 7-day window:

1. Frequency rising week over week, not just high.
2. Link CTR falling while frequency rises.
3. CPM rising in this ad set only. If it rose everywhere, it's the auction.
4. Cost per result rising after the three above.
5. Ad relevance diagnostics slipping: quality ranking and engagement rate ranking moving down.
6. Negative comments such as "I've seen this a hundred times", common in retargeting.

The first three together are the signature of a tired creative. How to confirm it and swap without losing learning is in creative fatigue: when to swap, so I won't repeat it here.

What to do once you've gone too far

The fix depends on the cause, and the cause is almost always one of three.

Audience too small for the budget

The most common cause in local prospecting. Do the math before touching the creative. Example: an audience of 50,000 people, US$ 250 a day, a CPM of US$ 25. That buys about 10,000 impressions a day, or 70,000 a week. Even if Meta reached all 50,000 people, weekly frequency would already be 1.4; since it reaches only part of them, real frequency climbs to 3 or 4 within days.

A new creative buys you a week. What fixes it is a bigger audience (wider radius, fewer filters) or a smaller budget on that ad set.

Tired creative

If the audience is large and frequency still climbs while CTR falls, it's the ad. Launch a new angle next to the old one and let delivery shift.

Pushy retargeting

Shorten the list window (30 days instead of 180), exclude buyers and rotate ads. Someone who saw it ten times without coming back rarely returns on the eleventh.

What about a frequency cap?

The hard cap Meta documents belongs to reservation buying, for awareness campaigns. In lead or sales campaigns at auction, I don't rely on a cap: I manage repetition through audience size, budget and creative rotation.

How to measure it without fooling yourself

In Ads Manager, the path Meta points to is Columns: Delivery, which adds reach and frequency. From there:

• Set the date range to the last 7 days, then to the previous 7.
• Read at the ad set level.
• Put frequency, link CTR, CPM and cost per result side by side.
• Log the numbers every Monday. Four weeks of your own history beat any range in this article.
• Don't react to decimals: the metric is estimated.

Next step

Open the account today, filter the last 7 days and list the ad sets above the range in the table. For each one, check whether CTR fell and cost rose in the same window. If yes, figure out whether the cause is a small audience or a tired creative before ordering new ads. If no, leave it alone: high frequency with good results is money at work. To see where frequency comes from, read reach vs impressions, and to track these metrics weekly in one place, see the paid media report.