Lookalike audience

A lookalike audience is a group of people that Meta's system identifies as similar to a source audience you supply. The source is usually a custom audience built from customers, website visitors, engaged followers or app users. Meta looks for shared characteristics and builds a new audience of people who are not in the source but resemble it.

How it works

• Source audience: you choose who the lookalike should be based on, for example everyone who purchased in the past 180 days.
• Location: you choose one or more countries or regions.
• Audience size: expressed as a percentage of the population in the chosen locations, typically from 1% to 10%. Smaller percentages are closer matches; larger ones trade similarity for reach.

Example (made-up numbers): if a country has 10 million eligible users, a 1% lookalike would contain roughly 100,000 people.

Tips for better sources

• Quality matters more than size. A list of repeat buyers or high-value customers often makes a better source than all site visitors.
• Make sure the source is large enough for Meta to work with; Ads Manager will tell you if it is too small.
• Refresh sources periodically so they reflect recent customers.

Lookalikes today

As Meta has moved towards automation through Advantage+, lookalikes are increasingly used as a signal rather than a hard boundary. In some campaign setups, the system may expand delivery beyond the lookalike when it predicts better results. Many advertisers now test lookalikes against broad targeting to see which performs better for their account.