AdWhiz
返回术语表

Lookalike Audience

定义

A Lookalike Audience on Meta Ads is a targeting option that finds new users who share similar characteristics with your existing customers. Created from a source audience like your customer list or website visitors, Lookalike audiences use Meta's data to identify people most likely to convert, making them one of the most effective prospecting tools in social advertising.

How Lookalike Audiences Work

You provide Meta with a source audience (at least 100 people from the same country), and the algorithm analyzes hundreds of data points including demographics, interests, behaviors, and engagement patterns. It then finds users in your target country who most closely match these patterns. You choose a percentage size from 1% (most similar, smallest audience) to 10% (broader, less similar). A 1% Lookalike typically contains 2-3 million people in the US. The quality of your source audience directly determines the quality of the Lookalike.

Choosing the Best Source Audience

The best source audiences are high-value customer lists: repeat purchasers, high-LTV customers, or qualified leads. Pixel-based audiences like purchasers or add-to-cart users also work well. Avoid using page likes or broad website visitors as sources since they include too many low-intent users. Aim for a source audience of 1,000-10,000 people for optimal results. Value-based Lookalikes (where you upload purchase amounts alongside customer data) let Meta prioritize finding users similar to your highest-value customers, not just any customer.

Lookalike Size Strategy

Start with a 1% Lookalike for the highest quality match, then test expanding to 2-5% for more reach. Stacking multiple Lookalike sizes in separate ad sets lets you compare performance at each level. For retargeting-heavy funnels, combine Lookalikes with interest targeting for layered precision. In many accounts, 3-5% Lookalikes outperform 1% because Meta's algorithm has more room to optimize within a larger pool. Test both narrow and broad approaches for your specific account before committing budget.

Lookalike Audience Insights with AdWhiz

AdWhiz analyzes your Lookalike audience performance across campaigns, comparing conversion rates and costs between different source audiences and percentage sizes. The audit identifies which Lookalike audiences deliver the best cost per acquisition and recommends source audience improvements. For accounts not yet using Lookalikes, AdWhiz evaluates your existing customer data and suggests the most effective source audiences to create.

常见问题

Meta requires at least 100 people from the same country in your source audience. However, for best results, aim for 1,000-10,000 people. Larger source audiences give Meta more data points to identify patterns, but extremely large sources (100,000+) may dilute the signal by including too many low-value users.

Start with 1% for the most precise match to your source, then test 3-5% for broader reach. Many advertisers find that 3-5% Lookalikes perform nearly as well as 1% with significantly more reach. The optimal size depends on your budget, market size, and how broad your product appeal is.

Refresh your source audiences quarterly by uploading updated customer lists. Pixel-based Lookalikes update automatically as new data flows in. If your customer profile has changed significantly (new product lines, new markets), create new Lookalikes from recent customer data rather than relying on stale historical sources.

设置一次，自动运行。

连接广告账户，让 AI 接管——出价、预算、关键词和创意测试。免费开始。

留下邮箱，我们团队将协助你配置和解决遇到的问题。

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.