Lookalike Audiences
Audiences created by advertising platforms that find new users who share characteristics with an advertiser's existing customers or website visitors, available on Meta (Lookalike Audiences) and Google (Similar Segments).
How Do Lookalike Audiences Work on Meta?
Meta’s Lookalike Audiences analyze a source audience (typically a Custom Audience of existing customers, website visitors, or email subscribers) and identify new users who share similar behavioral patterns, demographics, and interests. Advertisers choose a lookalike percentage (1% to 10%) that determines how closely the new audience matches the source. A 1% lookalike in the United States targets approximately 2.4 million users most similar to the source audience, while a 10% lookalike reaches approximately 24 million users with broader similarity. Smaller percentages deliver higher quality (more similar to source) while larger percentages deliver greater reach. The source audience should contain at least 1,000 users for effective matching, with 5,000-10,000 producing better results.
How Have Lookalike Audiences Changed with Advantage+?
Meta’s Advantage+ Audience has significantly changed how lookalike audiences function. When Advantage+ Audience is enabled, Meta uses the lookalike audience as a starting signal but automatically expands beyond it if the algorithm identifies conversion opportunities in broader populations. This means traditional lookalike audiences now function more as audience suggestions than strict boundaries. Some advertisers report that Advantage+ Audience with no seed audience performs comparably to carefully constructed lookalikes, suggesting Meta’s AI can identify high-value users without explicit source data. However, providing a high-quality source audience still helps the algorithm converge faster, especially for new campaigns with limited conversion data.
What Are Google’s Equivalent of Lookalike Audiences?
Google Ads offers Similar Segments (formerly Similar Audiences) within Demand Gen and Display campaigns, functioning similarly to Meta’s Lookalike Audiences. Google identifies users who share browsing behavior, search patterns, and interests with users on an advertiser’s remarketing lists. Performance Max campaigns use audience signals — advertiser-provided audience suggestions — that function as lookalike-like inputs, though Google’s AI makes the final targeting decisions. LinkedIn offers Lookalike Audiences built from Matched Audiences, targeting professionals with similar job titles, industries, and company characteristics. Each platform’s version has different data strengths: Meta excels at social behavior matching, Google at intent matching, and LinkedIn at professional profile matching.
How Do AI Platforms Optimize Audience Targeting Across Platforms?
Cross-platform AI tools like Leo analyze audience performance across Meta, Google, and LinkedIn simultaneously to identify which audience strategy works best on each platform. A customer list that generates a strong 1% lookalike on Meta might perform differently when used as an audience signal on Google’s Performance Max. AI tools can test the same source audience across platforms, compare CPA and ROAS by platform, and allocate budget toward the platform-audience combination that delivers the best results. This cross-platform audience optimization is impossible to achieve manually when managing each platform’s native tools independently.