Ad frequency: how to calculate it, what is good, and frequency caps

Ad frequency formula with a worked example

Take impressions and reach from the same provider, the same level (campaign, ad set or ad) and the same complete period. Divide impressions by reach. Rearranging the same relation gives impressions from a planned reach and frequency, or the reach a number of impressions implies at a given frequency.

Worked example (illustrative numbers): a campaign records 24,000 impressions and a reach of 8,000 in a month. 24,000 ÷ 8,000 = an average frequency of 3.0. If the plan is to reach 10,000 people about 4 times each, you need roughly 40,000 impressions. If the target audience is 50,000 people, a reach of 8,000 is a reach rate of 16%. The calculator below solves for frequency, impressions needed or implied reach in your browser.

How to calculate ad frequency, step by step

Step 1: choose one provider, one level (campaign, ad set or ad) and one complete period, for example the last full week. Step 2: read impressions and reach for that exact scope from the same report. Step 3: divide impressions by reach. Step 4: compare the result with the same scope in earlier periods, not with another platform or a generic benchmark.

Most platforms also show frequency directly, already calculated for the scope in view. Calculating it yourself is useful when you export impressions and reach to a spreadsheet, or when you plan a campaign and need the impressions a reach and frequency target implies.

What is a good ad frequency?

There is no single good ad frequency, and any fixed number quoted without the objective, audience and period behind it should be treated with caution. The most cited idea is Herbert Krugman’s 1972 paper “Why three exposures may be enough”, which argued that a third exposure acts as the reminder and later exposures repeat it. “Effective frequency” planning grew from that work and is usually expressed as a range within a purchase cycle, below which exposure is too thin and above which it is waste. These were developed for television, and neither gives a cut-off that fits every digital campaign.

A practical way to find your own range: take eight or more complete weeks for one campaign, put weekly frequency beside click-through rate and cost per result, and look for the point where response starts to fall while frequency keeps rising. That point is specific to your audience size, creative and offer, and it moves when any of them changes. The table summarises how objective changes the reading; it contains no benchmarks because none apply across accounts.

Frequency vs reach vs impressions

The three measures describe the same delivery from different sides. Impressions count showings, reach counts people, and frequency is the ratio between them. Keep them in separate columns and read them together: rising impressions with flat reach means the same people are seeing the ad more often.

In media planning, frequency also feeds gross rating points: GRPs = reach as a percentage of the target audience × average frequency. 40% reach at a frequency of 3 is 120 GRPs.

An average frequency hides a distribution

A frequency of 3.0 does not mean each of the 8,000 people saw the ad three times. A plausible split is that 5,000 saw it once, 2,000 saw it four times and 1,000 saw it eleven times, which also averages to 3.0. The figures are invented to show how different the experience behind one average can be. Where the provider offers a frequency distribution breakdown, read it before deciding the audience is saturated.

Where to find ad frequency in Meta Ads Manager and Google Ads

Meta Ads Manager: open Columns, choose Customise columns and add Reach, Impressions and Frequency. Meta calculates frequency for the exact scope and dates in the view, so the campaign row is not the average of its ad set rows. In the Marketing API Insights endpoint the fields are reach, impressions and frequency. Reach and frequency buying lets you set a frequency cap for eligible campaigns.

Google Ads: for campaign types that support reach measurement, such as Video, Display and Demand Gen, add the Unique users and Avg. impr. freq. per user columns. Display and Video campaigns also let you set a frequency cap in campaign settings. Search campaigns do not report reach or frequency, because each impression answers a separate query.

Frequency cap: how to limit ad frequency in Google Ads and Meta

A frequency cap is a setting, not a measurement: it limits how often the platform tries to show an ad to one person over a period. Reported frequency can still differ from the cap, because reach is estimated, people use several devices, and the report period rarely matches the cap period.

Google Ads: Display and Video campaigns support frequency caps, Search campaigns do not. In Display campaigns you limit impressions per person per day, week or month, at campaign, ad group or ad level. In Video campaigns you limit impressions and/or views per day, week or month, at campaign level only; the cap applies to signed-in users, or to devices when people are signed out. When someone reaches the cap, Google stops showing them that campaign’s ads.

Meta: a frequency cap is available when the ad set’s performance goal is Reach or ThruPlay. In the Marketing API this is frequency_control_specs, with a maximum frequency and an interval of 1 to 90 days, for example at most 2 impressions every 7 days. Ad sets optimising for conversions or link clicks do not take a cap; there, monitor frequency in reporting and refresh creative or widen the audience instead.

When is frequency undefined, and why can’t daily figures be averaged?

When reach is zero, the ratio is undefined. Show it as unavailable, not as zero: nothing was delivered. Frequencies built from very small reach are also unreliable. Ten impressions over four people gives 2.5, but a single extra viewer changes it substantially.

Daily reach rows overlap: a person reached on three days appears in three rows. Adding daily reach overstates period reach, and dividing period impressions by that inflated total understates frequency. Averaging daily frequencies has the same problem in a different form. Use the provider’s period-level reach and frequency for the period you care about.

Does rising ad frequency prove creative fatigue?

A rise suggests exposure is concentrating on the same people, but it does not establish fatigue by itself. Fatigue is a claim about response, so look at click-through rate, cost per outcome and conversion rate over the same weeks. If frequency rose while response held, the audience may simply be small or the purchase cycle long.

Consider the purchase cycle and the creative context before recommending a change. A high-consideration service may benefit from repeated exposure; a one-off impulse product may not. There is no universal stopping point that fits every audience, and no number on this page is one.

How to investigate ad frequency in Adlane

Connect the Meta or Google Ads account and ask a bounded question such as “For the last eight complete weeks, show weekly impressions, reach, frequency, click-through rate and cost per purchase for this campaign in the account currency.” Review the evidence rows and check that frequency was read at campaign level rather than summed from ad sets.

If a sustained pattern suggests refreshing creative or widening an audience, Adlane prepares a proposal for your approval; it never changes the account from the conversation.

Questions, answered

Sources

• Google Ads: Measuring reach and frequency
• Google Ads: About frequency capping
• Meta Marketing API: Ad set reference (frequency_control_specs)
• Meta Marketing API: Insights
• Frequency (marketing), including Krugman’s three-exposure theory and effective frequency

Continue your review

• Reach vs impressions: what reach means in advertising
• What are ad impressions? Definition, formulas and examples
• CPM meaning: cost per thousand impressions, formula and calculator
• Meta ads frequency: what the average means and how to read it
• Creative fatigue in Meta Ads: test the hypothesis before acting