Logo

MonoCalc

/

Ad Frequency Calculator

Social Media
Enter any two of the three and this fills in the third.
Total exposures. 1.8m and 1,800,000 both work.
Deduplicated people, not exposures.
Calculated below
People you could reach — not followers. Optional.
Both figures must cover it.
Optional; unlocks every money figure.
Display only — no conversion.

Frequency 6.00

over 30 days

exact

1,800,000 impressions ÷ 300,000 people reached · 60.0% of a 500,000 audience · GRP 360.0 · $14,400.00 at the CPM entered

What the average concealed (modelled)

Half the people you reached saw it 4 times or fewer, against a reported average of 6.00. The 12.7% who saw it 13 or more times absorbed 42.8% of your impressions — about $6,168.70 of $14,400.00. Only 36.7% of your target audience reached 3+ exposures.

23.8% of the people you reached (71,453) saw it exactly once.

If your platform reports a real frequency distribution, paste it in and nothing is modelled.
The most common planning convention, and this tool's default — a convention, not a finding.
How the spread was estimated
Random delivery would have reached 97.3% of the audience with these impressions; you reached 60.0%. Exposures were therefore more concentrated than chance — the normal case for auction-bought social — and are modelled as a negative binomial (Gamma-Poisson).

Who saw it how many times

modelled estimate
3+ exposures (effective)PEOPLE — how many saw it exactly this many times200k71k45k32k25k20k16k13k11k9.1k7.7k6.6k5.6k38kaverage 6.0median 4IMPRESSIONS — how many each group consumed71k90k97k99k98k95k91k87k82k77k72k68k771k012345678910111213+exposures per person

The two charts have opposite shapes, and that is the point: people thin out fast while the impressions pile up in the tail.

Exposures% of audiencePeople% of reachedImpressions% of impressions
040.00%200,00000.0%
114.29%71,45323.8%71,4534.0%
28.99%44,94515.0%89,8905.0%
36.47%32,34210.8%97,0265.4%
44.95%24,7388.2%98,9525.5%
53.92%19,5946.5%97,9705.4%
63.17%15,8755.3%95,2485.3%
72.61%13,0674.4%91,4675.1%
82.18%10,8823.6%87,0584.8%
91.83%9,1453.0%82,3064.6%
101.55%7,7412.6%77,4054.3%
111.32%6,5902.2%72,4894.0%
121.13%5,6371.9%67,6493.8%
13+7.60%37,99212.7%771,08842.8%

People counts are expected values, not headcounts of specific individuals, so displayed rounding can leave the column a person or two short of the reach total.

Effective reach

ThresholdOf audiencePeopleOf reached
1+60.00%300,000100.0%
2+45.71%228,54776.2%
3+36.72%183,60261.2%
5+25.30%126,52242.2%
10+11.59%57,95919.3%

One number, three framings — a share of the audience, a headcount and a share of the people you actually reached.

Impression concentration

perfectly eventop 10% took 36.9%share of reached people →↑ share of impressions

The heaviest-exposed 10% of the people you reached took 36.9% of your impressions; the heaviest 20% took 55.9%.

Delivery was heavily concentrated: a small slice of the audience took most of the impressions.

The same campaign by time window

Per day

1.21

at 10.0% reach

Per 7 days

2.31

at 36.3% reach

Per 14 days

3.49

at 48.1% reach

30-day campaign

6.00

at 60.0% reach

exact

Window figures assume delivery was spread evenly and the audience was stable across the flight. A burst-scheduled campaign breaks that assumption, so read them as a shape rather than a measurement.

About This Tool

Ad Frequency Calculator – What the Average Hides

Ad frequency is impressions divided by reach: the average number of times each person you reached saw the ad. Every platform reports it, and almost nobody reads it correctly, because the average is the least informative number in the whole distribution. This ad frequency calculator does the division, and then does the part the division hides.

The reach and frequency formula

The identity rearranges three ways:

  • frequency = impressions ÷ reach
  • reach = impressions ÷ frequency
  • impressions = reach × frequency

Reach counts deduplicated people; impressions count exposures. Every reached person contributes at least one impression, so frequency ≥ 1 always. A frequency below 1 is not a small number, it is an impossible one — usually reach and impressions swapped, or two figures pulled from different date ranges.

Where you also know the size of the audience you could reach, two more figures fall out: reach % = reach ÷ audience and GRP = impressions ÷ audience × 100. GRP is the one place the identity is visibly self-consistent, because it also equals reach % × frequency.

Why average frequency is misleading

Take a campaign that bought 1,800,000 impressions and reached 300,000 people out of a 500,000 audience. It reports a tidy frequency of 6.0. What actually happened is nothing like six exposures each:

  • 23.8% of the people reached saw it exactly once.
  • The median reached person saw it 4 times, not six.
  • The heaviest-exposed 12.7% — people who saw it 13 or more times — absorbed 42.8% of every impression bought, around $6,169 of a $14,400 buy at an $8 CPM.
  • Only 36.7% of the target audience cleared three exposures.

The gap between “frequency 6.0” and that is the entire reason this tool exists. A planner’s job is not to hit an average; it is to get enough people over a useful exposure threshold without burning budget on the small group that has already seen it fifteen times.

Effective reach and frequency capping

Effective reach is the share of the audience that saw the ad at least k times. It is worth stating three ways, because all three get asked for: on the campaign above, 3+ reach is 36.7% of the audience, 183,602 people, and 61.2% of everyone reached — three very different-sounding statements about one number.

Frequency capping attacks the tail. At a cap of five, roughly 810,069 impressions — 45% of the buy — went to people who had already passed it. Those impressions could in principle have gone somewhere new: with a hard cap of c, the same impressions cannot reach more than impressions ÷ c people, which here is 360,000 rather than 300,000.

A ceiling is not a forecast
impressions ÷ cap is an arithmetic upper bound under perfect delivery. No real platform achieves it, and a plan built on the ceiling will miss every time. Read it as the room a cap creates, not the reach it promises.

Frequency needs a time window

Frequency 6 over 30 days is about one exposure every five days. Frequency 6 over three days is somebody seeing the same ad twice a day. Dividing by the number of weeks is wrong, because reach over a shorter window shrinks too — properly rescaled, that 30-day frequency of 6.0 is 2.31 per week at 36.3% weekly reach and 1.21 per day at 10.0% daily reach.

Is three exposures the right target?

Genuinely unsettled. Krugman’s Why Three Exposures May Be Enough (1972) and Naples’ Effective Frequency (1979) established roughly three exposures per purchase cycle as the planning convention, which is where “3+ reach” comes from. Jones’ When Ads Work (1995) and Ephron’s recency planning (1997) used single-source data to argue the opposite: the first exposure does most of the work, so continuity and broad 1+ reach beat concentrated bursts. This calculator defaults to 3+ because it is the most common convention, shows the whole ladder from 1+ to 10+ regardless, and does not tell you which school is right.

Frequently Asked Questions

Is the Ad Frequency Calculator free?

Yes, Ad Frequency Calculator is totally free :)

Can I use the Ad Frequency Calculator offline?

Yes, you can install the webapp as PWA.

Is it safe to use Ad Frequency Calculator?

Yes, any data related to Ad Frequency Calculator only stored in your browser (if storage required). You can simply clear browser cache to clear all the stored data. We do not store any data on server.

How does this ad frequency calculator work?

It solves the identity frequency = impressions ÷ reach, and will solve it in any direction: give it any two of impressions, reach and frequency and it returns the third, plus reach as a percentage of your audience, GRP and the budget at your CPM. Add the size of the audience you could reach and it goes further, fitting an exposure model to your reach and impressions to recover the distribution the platform does not show you — how many people saw the ad once, twice, thirteen or more times.

Why can ad frequency never be lower than 1?

Because reach counts deduplicated people and impressions count exposures, so every reached person contributes at least one impression. A frequency below 1 is arithmetically impossible, and when a platform export appears to show one the cause is almost always reach and impressions swapped, or the two figures pulled from different date ranges. This calculator refuses that input and explains it rather than quietly clamping the answer to 1.

Why is average frequency misleading?

Because it is a quotient of two totals and says nothing about spread. On the worked example — 1,800,000 impressions and 300,000 people reached in a 500,000 audience — the average is 6.0, but 23.8% of the reached audience saw the ad exactly once, the median reached person saw it 4 times, and the heaviest-exposed 12.7% absorbed 42.8% of every impression bought. The average sits in a part of the distribution where relatively few people actually are.

What is effective reach, and is 3+ the right threshold?

Effective reach is the share of your audience that saw the ad at least k times. The tool defaults to 3+ because it is the most common planning convention, not because it is correct: Krugman (1972) and Naples (1979) argued roughly three exposures are the minimum useful level, while Jones (1995) and Ephron (1997) used single-source data to argue the first exposure does most of the work and continuity beats concentration. The research genuinely disagrees, so the whole ladder from 1+ to 10+ is always shown and no threshold is presented as the answer.

How do I choose a frequency cap?

Look at what a cap frees rather than at the cap number. For each cap from 1 to 20 the tool estimates the impressions delivered to people who had already passed it, prices them at your CPM, and shows the exact reach ceiling — impressions ÷ cap, bounded by your audience. On the worked example a cap of 5 leaves 810,069 impressions (45% of the buy) going to people already past it, against a ceiling of 360,000 people. The ceiling is an arithmetic bound under perfect delivery, never a forecast.

Why does frequency mean nothing without a time window?

Frequency 6 over 30 days is roughly one exposure every five days; frequency 6 over 3 days is somebody seeing the same ad twice a day. Dividing the campaign frequency by the number of weeks is wrong, because reach over a shorter window is smaller too. The tool rescales the fitted model instead, turning frequency 6.0 over 30 days into 2.31 per week at 36.3% weekly reach and 1.21 per day at 10.0% daily reach — assuming even delivery and a stable audience, which a burst schedule breaks.