Lookalike audience
Why it matters
Most small teams can only retarget a limited number of warm visitors. A lookalike audience is a way to reach new people without guessing at interests and job titles. The platform finds patterns a person would not spot. The result depends almost entirely on the source list. Built from best-fit customers, it works. Built from everyone who ever gave an email address, it finds more of the same average person.
How to apply it
- Build the source from your best customers, for example those who stayed longest or paid the most, not the whole contact database.
- Use a source that is large enough. Meta's minimum is 100 people from one country, and larger lists of a few thousand tend to give better matches.
- Start with a narrow audience, then test a broader one against it and compare cost per qualified lead.
- Pair it with a clear offer, because targeting cannot rescue a weak message.
- Refresh the source every few months as the customer base changes.
- Uploading customer data needs a lawful basis under GDPR and similar rules, so check consent before sending a list.
What it is
You give the ad platform a source list, such as the emails of your best customers. The platform matches those people to its own users, studies what they have in common, and finds other users who share those traits. The result is a new, cold audience that looks like your customers without being on your list. Meta's version is called a Lookalike Audience, and other platforms offer similar tools under different names.
On Meta, the size is set as a percentage of a country's population. A 1 percent audience is the closest match to the source, and larger percentages are broader and less alike.
Common mistakes
Using a source list of low-value customers. Judging results on clicks alone when the goal is customers. Letting the audience overlap with the retargeting audience, so that the two campaigns bid against each other.
