Skip to content
PromotEdge

Home Glossary Lookalike audience

Glossary · Paid media

Lookalike audiencemeaning and examples

New people who resemble your customers: a way to find more of them.

Since 201565+ in-house team6 cities

Updated 9 Oct 2026

Brands we have worked with

  • Sai Angan
  • Rupa Peekaboo
  • Tea Junction
  • Zandu Vigorex
  • Fox & Mandal
  • Veedol
  • Tata Efee
  • Avadh Sugar & Energy
  • Taj Cement
  • Weichai
  • Rerasathi
  • Honest AI
  • Walton
  • Supertron
  • Balrampur Foundation
  • Fine Organics
  • Kopran
  • Bengal Aerotropolis Projects Limited
  • FIVB
  • Hanumanta Foods
  • Linc
  • Nitya Gold & Diamonds
  • Novozymes
  • PSPL
See all 122 clients
Definition

A lookalike audience is a group of new people who share traits with your existing customers. An ad platform builds it from a list or group you give it, then shows your ads to people who look like that group.

You give the platform a source, such as a list of your customers. It finds other people with similar traits and shows them your ads. It is a way to reach new buyers without guessing who they are.

What Meta says

On 9 October 2026 we read the Meta Business Help Centre page "About lookalike audiences". Meta uses the name lookalike audiences. It says they let your ads reach new people who are likely to be interested in your business, because they are similar to your existing customers.

  • Meta's system uses information such as demographics, interests and behaviours from your source audience to find similar people.
  • You choose a percentage range. Smaller percentages match your source more closely. Larger percentages make a bigger, broader audience.
  • The quality of the source matters. Meta says you may get better results from an audience made from your most valuable customers than from all of them.
  • Lookalike audiences are limited or unavailable for some ads about financial products and services, employment or housing.
  • Targeting options are limited for audiences that include people under 18.

The same page links to guides on value-based lookalike audiences and Advantage lookalike. We did not read those. Meta changes its tools often, so read the live page before you set up a campaign.

A made-up example in rupees

Say an online store in Surat has a list of 1,200 repeat buyers. This is a made-up example. It runs two ad sets, each with ₹10,000 for one week.

  • Ad set A targets a lookalike audience built from the repeat buyers. It brings 14 orders. Cost for each order: ₹10,000 ÷ 14 = about ₹714.
  • Ad set B targets people by interest alone. It brings 9 orders. Cost for each order: ₹10,000 ÷ 9 = about ₹1,111.

One week of results is small. The store repeats the test before moving more money. Always compare the cost of each order with what a customer is worth, as in CAC (customer acquisition cost).

How it works for an Indian business

A good source is your own record: buyers, repeat buyers or leads who became customers. A list of people who never bought is a weak source. Keep the list clean and make sure you have permission to use the data. Rules depend on the platform and the country, so check the platform's terms.

Common mistakes

  • Building the audience from a tiny or messy list.
  • Using all contacts, including people who never bought, as the source.
  • Choosing a very wide range and then wondering why results are weak.
  • Never refreshing the source as new customers come in.
  • Using it for ads that fall under a restricted category and getting blocked.

Lookalikes find new people. To bring back people who already visited, read about retargeting.

Related terms

  • Retargeting: ads for people who already visited your site.
  • CAC (customer acquisition cost): what you spend to win one customer.
  • Attribution: deciding which source gets credit for a sale.

To plan the audiences for your Meta campaigns, see our Meta ads work.

Sources

  • Meta Business Help Centre, "About lookalike audiences" (facebook.com/business/help/164749007013531), read 9 October 2026.

Our beliefChange is the only constant.

What people say

In their own words

“An aligned partner is always a welcoming step. Someone who can think deep through lines, understand the matrix of product management. Fortunate enough, PromotEdge has been a partner in progress for me. …”
Bobby SrinivasBobby SrinivasChief Manager Marketing & Brand Marketing at Utkarsh India
“As a progressive and emerging digital marketing company, PromotEdge is doing something right to enhance the experience of companies associated with them. …”
Anish Basu MallikAnish Basu MallikMarketing Executive at Unimark Group
“PRAN & PromotEdge have been working together for last 1.5 years. Throughout, the work experience is good and satisfactory. As a client, we found PromotEdge very professional and dedicated towards any project assigned to them. …”
Sharif Rezwan KabirSharif Rezwan KabirBrand Manager at PRAN RFL Group
Read all 20
No pressure. No sales pitch.

Tell us what’snot working.

Or leave your number. A strategist calls you in one working day.

We use these details only to call you back. Privacy notice.

Straight answers.

  • It is a group of new people who share traits with your existing customers. The platform builds it from a source you give it.

  • Meta says smaller percentages match your source audience more closely, and larger percentages create a bigger, broader audience. Choosing a very wide range is a common mistake that leads to weak results. Meta also says a source made from your most valuable customers may give better results than one made from all of them.

  • No. Meta says lookalike audiences are limited or unavailable for some ads about financial products and services, employment or housing. Targeting options are also limited for audiences that include people under 18. Using one for a restricted category can get your ads blocked, so read Meta's live page before you set up a campaign.

  • Start with a good source: your own buyers, repeat buyers or leads who became customers. A list of people who never bought is a weak source. Keep the list clean, refresh it as new customers come in, and make sure you have permission to use the data. Check the platform's terms.