What is lookalike audience?
A lookalike audience is a group of people an ad platform assembles because they resemble a source list you provided, such as your customers or your highest value buyers.
Worked example, not a result
The seed list decides everything
Say a store uploads its full buyer list of 42,000 people and builds an audience from it. The result resembles the average customer, which on most stores means it leans towards whoever bought the cheapest introductory product, because that group is the largest.
Now build a second audience from a narrower seed: the 9,500 customers who ordered more than once or who spent above a chosen level. The people in that audience resemble buyers who came back and spent more, which is a different and usually more valuable group.
The lesson is that the platform builds what you ask it to build. It cannot tell which customers you wish you had more of, and the default of feeding it everyone quietly asks for more of what you already have too much of.
Audiences also go stale as the seed ages, so the list is worth rebuilding on a schedule. The counts above are an example.
Why it matters
Lookalike audiences are one of the main ways a store buys new visitors on feed based platforms, so they sit directly on the visitors lever. What makes them useful is not the technology, it is the quality of the customer data you feed in, which is something a store controls entirely.
They also connect back to the repeat purchase and order value levers in a way most stores miss. Building the seed from repeat buyers or from high value customers means the traffic arriving is more likely to behave the same way, which improves lifetime value rather than just order count. Building it from everyone optimises for volume of buyers rather than quality of them.
For a store owner the decision is which customers to describe. Write down what your best customer looks like, build the seed from those people, and refresh it as the business changes.
Common mistakes
- Seeding from everyone who ever bought. The audience then resembles your average customer, including the discount hunters and the one time buyers.
- Building from a list too small to be meaningful. A thin seed produces a vague audience, and the results are indistinguishable from broad targeting.
- Never refreshing the source. Customer behaviour changes with the range and the season, and an audience built from an old list keeps chasing an old business.
- Uploading customer data without a lawful basis and a clear privacy notice. The list is personal data whatever the platform calls it, and that obligation does not move.
Related terms and answers
Your plan is free.
A person reads it. You keep the plan. No logins, no call first.