What is audience overlap on Meta?
Audience overlap is the share of one audience that is also in another. On Meta, two ad sets overlap when the people they target are partly the same. To calculate it, divide the number of people in both audiences by the size of the audience you start from, then multiply by 100. The result depends on that starting point: if 240,000 people are in both a 2,000,000-person audience and an 800,000-person audience, that is 12% of the first and 30% of the second. Meta says overlapping audiences can lead to poor delivery: when ad sets from the same advertiser would enter the same auction, Meta enters only the one with the best performance history, so your ads do not compete against each other. A high overlap therefore holds back one of your ad sets. This tool reads the higher of the two percentages, the one for the smaller audience, and tells you whether to keep both ad sets, exclude one audience from the other, or merge them.
How does Meta’s Audience Overlap tool work?
Meta’s tool compares the audiences listed in your ad account. Its help page gives three steps:
- Go to Audiences in your ad account.
- Tick the boxes next to the audiences you want to compare, up to 5.
- Click Actions, then Show audience overlap.
The first audience you tick becomes the selected audience; the others are comparison audiences. For each one, the tool shows the Overlap, the number of Accounts Center accounts in both, and the % overlap of selected audience, that number divided by the size of the selected audience. Change the selected audience and the percentage changes, while the number of people in both stays the same. Meta adds that the tool only gives useful information for audiences of at least 10,000 Accounts Center accounts. Meta does not publish an overlap percentage at which you should act.
- Meta Business Help Center: About overlapping audiences. Read on October 9, 2026. The page shows no update date.
What audience overlap percentage is too high?
Meta gives no threshold, so this tool uses a rule of thumb. It reads the overlap of the smaller audience: in Meta’s own explanation, that is the audience whose ad sets are more likely to have delivery problems.
| Overlap of the smaller audience | What to do | Decision |
|---|---|---|
| Under 20% | Most people are unique to each ad set. Leave both as they are. | Keep both |
| 20% to under 50% | Exclude the smaller audience from the ad set that targets the larger one, so each ad set keeps its own people. | Exclude |
| 50% or more | Half or more of the smaller audience is shared. Combine the two into one ad set with one budget. | Merge |
These lines are Adwize’s rule of thumb, not a Meta rule. Move them if your own tests say otherwise, and check delivery in Ads Manager after each change.
What does an overlap calculation look like?
These are example numbers, not a benchmark. Ad set A targets an interest audience of 2,000,000 people. Ad set B targets a lookalike audience of 800,000. Meta’s tool shows 240,000 people in both.
| Step | Calculation | Result |
|---|---|---|
| Share of A also in B | 240,000 ÷ 2,000,000 × 100 | 12% |
| Share of B also in A | 240,000 ÷ 800,000 × 100 | 30% |
| People in A or B | 2,000,000 + 800,000 − 240,000 | 2,560,000 |
| Decision | 30% ≥ 20% | Exclude |
The smaller audience, B, shares 30% of its people with A. That is between 20% and 50%: exclude audience B from ad set A. Ad set B keeps its 800,000 people, and ad set A stops bidding for the 240,000 it shared with B.
Can I compare interests instead of audiences?
Yes, as a first check. Paste the interests of each ad set and the tool counts the ones they share, as a share of the shorter list. It is a rough guide, not a count of people: two different interests can reach many of the same people, and Advantage+ audience can deliver beyond the interests you entered. When the interest overlap looks high, compare the audiences themselves in Meta’s Audience Overlap tool to get the real numbers.