Campaign setup, budget and delivery

How is frequency calculated in Facebook ads?

Divide impressions by reach and you have Meta's frequency: an average, for one period, at one level. The formula, the caps that exist and how to read it per ad.

Published

Frequency in Facebook ads is impressions divided by reach. Impressions count every view; reach counts the Accounts Center accounts that saw the ad at least once; Meta labels both as estimates. The result is the average number of views per person in the period you picked, which means it is a summary of everyone reached rather than a fact about anyone in particular.

The formula in one line

Meta’s Marketing API reference for ads insights supplies both halves. Frequency is described as the average number of times each person saw the ad, reach as the count of accounts that saw it at least once, with the note that reach is not the same as impressions because impressions include repeat views by the same account.

So: frequency = impressions ÷ reach. With no reach, there is nothing to divide by, which is why a metric can look blank on a very narrow audience.

Three reasons the same number can mislead

The division is simple; the reading is not. Three properties decide what the result means.

The first is that it is an average. A value of 3 fits two very different situations: everyone saw the ad three times, or most people saw it once and a handful saw it a dozen. In the second situation the creative is already tired for those people while the average still looks calm.

The second is the period. Reach is deduplicated inside the window, so frequency does not scale the way spend does:

WindowImpressionsReachFrequency
Week 140,00020,0002.0
Week 240,00020,0002.0
Weeks 1 and 280,00030,0002.7

Both weeks read 2.0 and the fortnight reads 2.7, because part of the audience returned. Compare seven days with seven days, and treat any lifetime figure as a different metric.

The third is the level. At ad set level, a person who saw two ads in the set is counted once. Two ads with 20,000 impressions each, on 10,000 distinct people each, read 2.0 on their own; if 4,000 people saw both, the ad set shows 40,000 impressions on 16,000 people, so 2.5. A tired creative sits at ad level, so that is where the number belongs.

Why no target frequency exists

Meta defines the metric without recommending a value, and a leads campaign has no cap to aim at either. The reason is that frequency follows budget and available audience more than it follows the ad.

Take a €1,000 weekly budget at a €10 CPM, which buys about 100,000 impressions. A plumber serving one town, where delivery finds about 30,000 people, ends the week at 3.3 or higher. The same money spread over 80,000 people across a region reads about 1.25. Identical creative, identical spend, a frequency nearly three times higher on one side, and neither number yet says whether the ad works.

SituationUsual patternWhat to watch
Small service areaclimbs within dayslink CTR and cost per lead against the ad’s own baseline
National or regional audiencestays low for weekscost per lead; frequency rarely explains anything
Retargeting past visitors or leadshigh from the start, because the list is smallcost per qualified lead, and whether the list is big enough
Budget raised on an unchanged audiencerises step for step with spendthe cost of the new spend, not the average

A workable definition: a good frequency is the one at which that ad’s link CTR and cost per lead are still holding. Two defaults make it operational. Put an ad on the review list when its seven-day frequency rises by about half against its first two full weeks. Leave an ad alone when frequency is high and CTR and cost per lead are steady. What frequency can and cannot tell a lead account goes further into that reading.

Where Meta lets you cap it

Not in a leads campaign. Meta’s ad set documentation says the cap field can only be written in ad sets where Reach or ThruPlay is the performance goal, with a maximum number of impressions per person over an interval of 1 to 90 days.

SetupCap availableWhat it optimizes forFit for lead generation
Leads campaign, any performance goalnoleads or conversion leadsthe campaign you actually run
Reach performance goalyes, impressions per person over 1 to 90 daysunique reachnot for leads
ThruPlay performance goalyes, same fieldvideo playsnot for leads
Reservation buyingyes, planned ahead at a fixed CPMpredictable reachawareness planning

Reservation buying is the second place a cap gets planned, and its scale is the tell: Meta’s documented minimum reach is 200,000 accounts, with target audiences of at least 300,000. Moving a lead campaign into a Reach campaign to obtain a cap swaps the result you want for the only control on offer. How to set a frequency cap covers the cases where that trade is worth making.

What to change instead, in a leads campaign

Without a cap, you change the inputs:

  1. Widen the audience, since a larger reachable pool lowers the ratio at the same spend. For a local business that means a radius decision before a creative decision.
  2. Stop bidding against yourself. Several ad sets aimed at the same people stack impressions on the same reach, so merge them and give the budget to fewer ad sets.
  3. Exclude people who already became leads, through a custom audience, so the spend goes to those who have not converted.
  4. Ship new concepts, because a fresh idea gives the same people a new reason to respond and is the real cure for a tired ad.
  5. Hold the budget on a saturated audience rather than buying more impressions into it.

Reading frequency per ad, next to cost per lead

Frequency measures exposure. Fatigue is a change in response. The two usually move together and sometimes do not, so read them side by side:

AdSeven-day frequencyLink clicks per impressionReal cost per lead
Ad A1.5 to 2.4about 1 click in 90, then 1 in 95€32 to €33
Ad B1.4 to 1.9about 1 click in 95, then 1 in 160€30 to €49

A rule of the kind “pause above 2.0” would cut Ad A, which is still producing at its original cost per lead, and keep Ad B, whose click rate lost roughly a third while its cost per lead climbed by about two thirds. Frequency ranked the two the wrong way round. Creative fatigue in Meta ads covers how to confirm the pattern before you replace a creative.

A routine that keeps it honest:

  1. Pull the last seven days per ad ID: frequency, link CTR, the leads you received, and the cost per lead built from them.
  2. Compare each ad with its own first two full weeks, never with the account average.
  3. Flag on frequency and CTR, since both rest on thousands of impressions and hundreds of clicks and move first.
  4. Decide on cost per lead, using two-week windows when an ad brings fewer than about 20 leads a week.
  5. Check whether frequency rose 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.

There is one case frequency cannot catch: CTR holds, cost per lead climbs, and the leak sits after the click, on the landing page, the form or the tracking. Reading that needs a cost per lead counted on your own leads. DropCPL counts leads on your landing pages and syncs spend per ad, so the figure next to Meta’s frequency comes from leads you received rather than from an attributed estimate. Cost per result explains how the two numbers differ.

Questions on this topic

All questions on this topic →

FAQ

Why is my 30-day frequency higher than my weekly numbers?+

Because reach counts a person once inside the period. The same people appear in several weekly reaches but only once in the monthly one, while impressions keep counting every view. A month can read well above any single week even when nobody changed anything.

Can I cap frequency in a leads campaign?+

No. A cap exists only for ad sets whose performance goal is Reach or ThruPlay, and for reservation buying. A leads campaign optimizes for leads or conversion leads, so the field is not available. You move frequency with the audience, the number of ad sets and the creative instead.

What frequency is too high?+

Meta sets no threshold, and its own definition allows for a wide range depending on budget, audience size and schedule. Use the ad's first two full weeks as its baseline and review a rise of about half against that baseline.

Does high frequency always mean creative fatigue?+

No. A bigger budget, a small service area or a retargeting list pushes frequency up while people still respond. Fatigue is a fall in response, so confirm it with link CTR over thousands of impressions and with cost per lead over enough leads.

Ready to find out what
converts, for real?

Free to start.

Start free →