Logo

MonoCalc

/

Audience Overlap Calculator

Social Media

The audiences you are comparing

Enter each audience size and the number of people they share. Shorthand such as 120,000 or 1.2m works. Nothing is fetched and no account is connected — the page makes no network calls at all, so the overlap always has to be a figure you supply.

Three audiences switches to full inclusion–exclusion across seven regions
Analytics tools quote the same fact all three ways

That same overlap, the other two ways: 30,000 people · 25.0% of Instagram · 37.5% of YouTube

Only needed for the affinity index
Both spends are needed for the cost figures
Display currency only — nothing is converted

170,000 unique people

Circle areas are proportional to the audience sizes and the lens area is proportional to the overlap, so the picture is to scale rather than schematic.

Instagram — 120,000YouTube — 80,00090,000only Instagram · 52.9%30,000both · 17.6%50,000only YouTube · 29.4%
What that means
Of the 170,000 real people reached, 30,000 are in both audiences — 17.6% of everyone reached. Adding the two audience figures together would claim 200,000, overstating reach by 17.6%.

Real unique reach

170,000 people

What adding the two figures would claim

200,000 people

Only Instagram — 90,000

Shared — 30,000

Only YouTube — 50,000

Reach, counted honestly

Total unique reach

170,000

A + B − overlap

Duplicated audience

30,000

people counted twice

Unique to Instagram

90,000

Unique to YouTube

50,000

incremental reach B adds

Incremental reach

41.7%

how much YouTube grows the Instagram audience

Inflated-reach error

+17.6%

adding the two figures claims 200,000

Naive sum

200,000

A + B, the number to stop reporting

How similar the two audiences are

The same overlap divided by four different denominators. Each answers a different question, which is why media plans quote more than one.

% of Instagram

25.0%

overlap ÷ A

% of YouTube

37.5%

overlap ÷ B

Jaccard index

17.6%

overlap ÷ unique reach — the symmetric one

Overlap coefficient

37.5%

overlap ÷ smaller audience; 100% means containment

Affinity index — is this overlap more than chance?

Two audiences drawn independently from the same population already overlap by a predictable amount. This compares what you measured against that.

Overlap chance predicts

1,920

A × B ÷ population

Overlap you measured

30,000

your own figure, unchanged

Affinity index

1,563

15.6× the chance overlap

Heavily duplicated

at least twice the chance overlap

What that reads as
The two audiences are heavily duplicated — far more shared people than chance would produce. Either reallocate part of the budget to a channel that reaches somebody else, or keep the pairing deliberately and treat it as a frequency and retargeting buy rather than a reach buy.
IndexReadingRestated as a multiple
below 80More distinct than chancebelow 0.8× the chance overlap
80–120About what chance predicts0.8× to 1.2× the chance overlap
120–200Noticeably shared audience1.2× to 2× the chance overlap
200 and aboveHeavily duplicatedYouat least twice the chance overlap
These cut-offs are a convention, not a benchmark
The 100 baseline is arithmetic: it is the overlap two independent audiences of these exact sizes would have inside this population. The 80, 120 and 200 cut-offs are a reading convention rather than a measurement — they are simply 0.8×, 1.2× and 2× that baseline. No published table of normal overlap indices by platform or industry exists to cite, so none is shipped here, and you should read the index against your own campaigns rather than against an imagined norm.

Where do I get the overlap number?

This is where most people stall — not on the arithmetic. The overlap has to come from somewhere that already de-duplicates people, because nothing on this page can discover it.

Meta Ads Manager

Audience overlap reporting compares two saved audiences and reports the shared share of each.

Google Ads audience reports

Audience segment reporting shows how much of one segment appears in another across a campaign.

GA4 cross-channel reach

Comparisons and segment overlap in explorations give unique users per channel and in combination.

Influencer analytics platforms

Most creator-analytics products publish follower-overlap figures between two accounts.

Your own first-party data

If both audiences are lists you hold, the overlap is a straight de-duplicated match on email or customer ID.

These are pointers to product areas rather than menu directions, because every one of these interfaces is rearranged periodically. If none of them can give you a figure, an overlap you have estimated is still worth entering — the tool will tell you how much the answer moves as you vary it.

About This Tool

Audience Overlap Calculator – What Your Real Reach Is

A campaign that runs on two channels does not reach the sum of their two audiences. It reaches their union, and the difference is the people you paid to reach twice. Adding an Instagram audience of 120,000 to a YouTube audience of 80,000 gives 200,000 only if not one single person follows both, which is almost never true. Every reach report built by addition is overstated, and this audience overlap calculator tells you by exactly how much.

The audience overlap formula

Two sets need only inclusion–exclusion. Subtract the shared people once, because addition counted them twice.

unique reach = A + B − overlap overlap % = overlap ÷ union × 100 (the Jaccard index) coefficient = overlap ÷ min(A, B) × 100 (Szymkiewicz–Simpson) inflation = (A + B) ÷ union × 100 − 100

The worked example, step by step

The calculator opens on a real-shaped scenario: Instagram 120,000, YouTube 80,000, and 30,000 people in both. The union is 120,000 + 80,000 − 30,000 = 170,000 unique humans. Of those, 90,000 are Instagram-only, 50,000 are YouTube-only and 30,000 see the campaign on both. Those 50,000 are the incremental reach YouTube buys on top of Instagram, and as a share of the Instagram audience that is 50,000 ÷ 120,000 = 41.7% more people.

The overlap itself reads three ways: 25.0% of Instagram, 37.5% of YouTube, and 17.6% of the 170,000 people actually reached. That last one is the Jaccard index, the single symmetric number for how alike two audiences are. And the naive 200,000 figure? It overstates reach by 17.6%, which is the same arithmetic seen from the other end.

Observed against expected: the affinity index

Subtraction tells you how big the duplication is. It does not tell you whether it is surprising, and that is the part most reports miss. Two audiences drawn independently from the same population overlap by a perfectly predictable amount: if a population P contains audiences of size A and B with no relationship between them, on average A × B ÷ P people land in both.

With a 5,000,000 addressable population that expected overlap is 120,000 × 80,000 ÷ 5,000,000 = 1,920 people. The measured overlap is 30,000 — an affinity index of 1,563, more than fifteen times what chance predicts. These two audiences are not two independent samples of a market; they are largely the same people, and the second channel is buying frequency far more than it is buying reach. An index near 100 would mean the opposite: overlap you could not have avoided, and two genuinely complementary channels.

Three channels, seven regions

Add a third audience and inclusion–exclusion grows a term: A + B + C − A∩B − A∩C − B∩C + A∩B∩C. The pairwise overlaps come off once each, then the triple overlap goes back on because it was removed three times and added three times. The calculator splits the result into all seven Venn regions and refuses inconsistent inputs by name — if the “only A” region computes negative, the overlaps you entered describe audiences that cannot exist.

High overlap is not automatically bad

Duplication is only waste when the goal is reach. If the campaign is built on frequency, retargeting or a sequenced message, the shared segment is the entire point — it is the only way any individual sees the message more than once. Decide what the money is for first, then read the number.

Measure the same people, the same way
Overlap figures are only comparable when both audiences are counted over the same window with the same definition. Mixing a 28-day reach against a lifetime follower count produces a number the arithmetic will happily process and no one should act on.

Frequently Asked Questions

Is the Audience Overlap Calculator free?

Yes, Audience Overlap Calculator is totally free :)

Can I use the Audience Overlap Calculator offline?

Yes, you can install the webapp as PWA.

Is it safe to use Audience Overlap Calculator?

Yes, any data related to Audience Overlap 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 audience overlap calculator work?

You type two (or three) audience sizes and the overlap between them, and every figure on the page is inclusion–exclusion over those numbers. Unique reach is A + B − overlap; the Jaccard index is the overlap divided by that union; the affinity index compares your measured overlap against the overlap two independent audiences of the same sizes would have inside the population you name. The Venn diagram is solved from the same three numbers — circle areas are proportional to the audiences and the lens area is proportional to the overlap. Nothing is fetched and no account is connected: the page makes no network calls at all.

What does audience overlap actually mean?

It is the number of individual people who appear in both audiences — the same human being counted once by each channel. If 30,000 of your 80,000 YouTube subscribers also follow your Instagram account, the overlap is 30,000 people, and every campaign that runs on both channels reaches those 30,000 twice. Overlap is a head-count, not a rate, which is why the same fact gets quoted three different ways: 30,000 people, 25% of Instagram, or 37.5% of YouTube all describe it.

Why can I not just add the two follower counts together?

Because addition counts the shared people twice. An Instagram audience of 120,000 plus a YouTube audience of 80,000 is 200,000 only if not one person follows both, which is almost never true. With a 30,000 overlap the honest number of humans is 170,000, and the 200,000 figure overstates reach by 17.6%. That inflation is exactly the error a campaign report carries when it sums per-channel reach, and it compounds with every extra channel you add.

Where do I get a real overlap number?

From whatever platform already de-duplicates people for you. Meta Ads Manager reports audience overlap between two saved audiences; Google Ads audience segment reporting shows how much of one segment appears in another; GA4 explorations give unique users per channel and in combination; most influencer-analytics products publish follower overlap between two accounts. If both audiences are lists you own, the overlap is just a de-duplicated match on email or customer ID. This tool cannot discover the number for you — there is no backend and no public API that exposes it.

What is the affinity index and what counts as a good value?

It is your measured overlap divided by the overlap that pure chance predicts, times 100. Two audiences of size A and B drawn independently from a population of P people share A × B / P members on average, so 100 means you measured exactly what chance predicts, 200 means twice that, and 50 means half. There is no published table of good values by platform or industry, and this tool deliberately ships none — the useful comparison is against your own other channel pairs. The 80, 120 and 200 cut-offs shown are a reading convention, not a benchmark: they are simply 0.8×, 1.2× and 2× the chance baseline.

Is a high overlap always bad?

No, and treating it that way is the most common misreading. High overlap is bad only when the goal is reach, because a second channel that lands on people you already reached buys no new humans. When the goal is frequency, retargeting or sequencing a message, high overlap is precisely what you are paying for — it is the only way to get a second exposure to the same person. Decide what the campaign is for first, then read the number: the same 65% overlap is a waste signal for an awareness launch and a success signal for a conversion push.