Meta Ads

What Are Custom and Lookalike Audiences? Reaching the Right People on Meta

Updated: 3 June 2026
All Topics
Short answer

A Custom Audience is a targeting group built from people who already know your business — past visitors, app users, or those who've messaged you before. A Lookalike Audience takes that list and lets Meta find new people who share similar traits. Used together, these tools let your ads reach both loyal customers and fresh prospects who genuinely resemble them.

Custom Audience: People Who Already Know You

Say you run a small boutique. People who visited your website or interacted with your Instagram posts have already noticed you once. A Custom Audience lets you introduce these people to Meta's ad system so you can target those who've shown interest — rather than starting cold with total strangers. Common sources include website visitors (tracked via Meta Pixel or Conversions API), existing customer lists (matched by email or phone number), app users, and people who've engaged with your Instagram or Facebook content.

Lookalike Audience: Finding People "Like Them"

Once your Custom Audience exists, Meta analyzes shared traits among those people — age ranges, interests, behavioral patterns — and finds others in your target country who fit a similar profile. You don't search for them one by one; Meta does the matching. You'll typically choose a percentage between 1 and 10: 1% means the closest match and usually performs best. As the percentage grows, the audience gets larger but less similar to your original list.

2× higher conversion rate

Meta reports that businesses combining Custom and Lookalike Audiences see roughly double the conversion rate compared to cold interest-based targeting alone.

  • Website visitor Custom Audiences — test 30-, 60-, and 180-day windows separately; recent visitors often respond best.
  • When uploading a customer list, include both email and phone number to improve the match rate.
  • Build Lookalike Audiences from a source of at least 1,000 people; smaller lists don't give Meta enough signal.
  • Always exclude your Custom Audience from Lookalike campaigns; otherwise you'll pay to re-target existing customers as if they were new.
  • A Lookalike built from buyers typically outperforms one built from all visitors — quality of the source list matters.
2026 privacy reality: Browser cookie restrictions have weakened Meta Pixel on its own. Without Conversions API — a server-side integration that sends customer data to Meta securely — your Custom Audiences may not fill up as expected. Ask your agency whether this integration is in place before investing in audience-based campaigns.
Practical starting point for small businesses: Upload your existing customer list (ideally 500–1,000+ contacts) and create a 1% Lookalike Audience from it. Split your budget — part to the Lookalike for new reach, part to website visitors for retargeting. These two campaigns complement each other well.

Frequently asked questions

Is it safe to upload my customer data to Meta?

Meta hashes (encrypts) uploaded emails and phone numbers before storing them — the raw data doesn't sit on their servers. That said, you need to have a valid legal basis under applicable privacy law (such as GDPR or local equivalents) to share this data, along with clear consent from your customers. Make sure your privacy policy mentions this type of data use.

What's the difference between a Lookalike Audience and interest-based targeting?

Meta can technically generate a Lookalike from a smaller list, but at least 1,000 contacts gives the system enough signal to build a meaningful profile. A list of 300–400 people will still work, but the audience tends to be less predictable. Combining your customer list with website visitors in one source can help bridge the gap.

What's the difference between a Lookalike Audience and interest-based targeting?

Interest targeting asks Meta to find people based on stated hobbies or topics — it's educated guesswork. A Lookalike Audience starts from a real data point: the actual traits your existing customers share. In practice, Lookalikes tend to perform more precisely than interest-based targeting because they're built on demonstrated behavior, not assumed preferences.

Need help with this?

Let's plan a path tailored to your business. First call is free, no commitment.