Custom Audience
A targeting option that lets you show ads to a specific group of people based on your own data. Your customer list, your website visitors, your app users, people who watched your videos. Instead of guessing who might be interested, you're targeting people who have already interacted with your brand.
Types of custom audiences
Customer lists. Upload emails or phone numbers from your CRM or email platform. The ad platform matches them against its user database. Match rates land between 30–70% depending on data quality and the platform.
Website visitors. Using your tracking pixel (Meta Pixel, Google Tag), you can build audiences of people who visited specific pages. Someone who spent time on your pricing page is a very different prospect than someone who read a blog post and bounced.
App activity. If you have a mobile app, you can target users based on their in-app behavior. People who installed but never completed onboarding, people who made a purchase, people who haven't opened the app in 30 days.
Engagement audiences. People who interacted with your content on the platform itself. Video viewers, people who opened a lead form, profile visitors, anyone who liked or commented on your posts. No website visit required.
Why custom audiences are useful
Retargeting gets specific. Instead of retargeting everyone who visited your site, you can show cart abandoners a "still thinking about it?" ad, show pricing page visitors a comparison chart, and show blog readers a top-of-funnel piece. Same strategy, much better execution.
Exclusions save money. Add your existing customers as a custom audience, then exclude them from your prospecting campaigns. Without this, you're paying to acquire people you've already acquired.
Reactivation campaigns. Upload your churned customers as a custom audience and show them ads about new features, updated pricing, or win-back offers. You already have a relationship with these people. Reigniting it is usually cheaper than finding new customers.
Lookalike foundation. Your best custom audiences become the seed for lookalike audiences. Upload your top customers and let Meta find people who look like them. The quality of your lookalike depends on the quality of the seed custom audience.
Platform differences
Meta has the most options. Customer lists, website visitors, app users, offline activity, engagement data. You can layer custom audiences together and combine them with interest targeting.
Google Ads calls it "Customer Match." Supports email lists and website audiences through Google Tag. Less flexible than Meta, but works well for search and YouTube campaigns.
LinkedIn accepts company lists and email lists. Website tracking works through the LinkedIn Insight Tag. Strong for B2B retargeting, especially when you layer on job title or seniority targeting.
TikTok supports customer uploads, pixel-based audiences, and engagement targeting. Still maturing, but improving fast.
Best practices
Segment by behavior, not just by visit. "All website visitors" is too broad. Create separate audiences for product page viewers, cart abandoners, blog readers, and pricing page visitors. Each group gets different messaging.
Use the right time windows. A 7-day website visitor audience is warm and recent. A 180-day audience is mostly cold. Match your retargeting urgency to how recently someone visited.
Refresh your lists. Customer email lists go stale. People change jobs, change emails, become irrelevant. Upload fresh lists monthly at minimum.
Mind the minimums. Meta requires at least 100 matched people in a custom audience. But for lookalike creation, you want 1,000–10,000 people in the seed audience for the algorithm to work well.
FAQ
Do I need consent to use customer data for custom audiences?
Under GDPR and most privacy regulations, yes. You need proper consent for uploading email lists and for pixel-based tracking. Your privacy policy should mention advertising as a use case for collected data.
What's the difference between a custom audience and a lookalike audience?
Custom audiences target people you already know (your data). Lookalike audiences find new people who resemble your custom audience (the platform's data). Custom is precise; lookalike is discovery.
Can I use custom audiences for prospecting?
Not directly, since they're built from existing contacts. But you use them as the foundation for lookalike audiences, which are your prospecting tool. You can also use custom audiences as exclusions to make sure your prospecting campaigns only reach new people.