Facebook Ads Audience Targeting in 2026: What Still Works
Facebook Ads targeting has changed more in the past three years than in the previous decade. Between privacy regulations, platform updates, and the rise of AI-driven optimisation, the targeting strategies that worked in 2022 are often outdated or counterproductive in 2026.
This guide covers what still works, what's changed, and how to build Meta Ads audiences that actually deliver results in the current landscape.
What's Changed in Facebook Ads Targeting
Before diving into strategies, it's important to understand the landscape shifts:
Interest Targeting Has Been Reduced
Meta has removed thousands of detailed targeting options related to sensitive topics (health, politics, religion, sexual orientation) and consolidated others. Many of the hyper-specific interest categories that advertisers relied on simply no longer exist.
Broad Targeting Is Now Competitive
Meta's machine learning has improved to the point where broad targeting (minimal audience restrictions) often outperforms detailed targeting for established accounts with strong conversion data. The algorithm learns who converts and finds more people like them — without you needing to specify interests.
Advantage+ Audience Is the Default
Meta has introduced Advantage+ audience, which uses your targeting inputs as "suggestions" rather than hard constraints. The algorithm can expand beyond your defined audience if it identifies likely converters outside your parameters. This is enabled by default in new campaigns.
Targeting Options That Still Work
1. Custom Audiences
Custom audiences remain the most powerful targeting tool in Meta's arsenal. These are built from your own first-party data:
- Website visitors: People who've visited your site, viewed specific pages, or taken specific actions (requires pixel and ideally CAPI)
- Customer lists: Upload email addresses or phone numbers to target existing customers or suppress them from prospecting
- Video viewers: People who've watched a percentage of your video content on Facebook or Instagram
- Engagement audiences: Users who've interacted with your Facebook page, Instagram profile, or ads
- App activity: Users who've taken actions within your app
Custom audiences are the foundation of effective retargeting strategies and serve as seed audiences for lookalike audiences.
2. Lookalike Audiences
Lookalike audiences find new people who share characteristics with your existing customers. Despite reduced data availability, they remain effective when built from high-quality seed audiences:
- 1% lookalikes: Closest match to your seed audience. Best for conversion campaigns with limited budgets.
- 1–3% lookalikes: Broader reach with slightly lower precision. Good for scaling.
- 3–5% lookalikes: Very broad. Best used as targeting suggestions within Advantage+ audience rather than hard constraints.
The quality of your seed audience matters more than the percentage. A 1% lookalike based on your top 100 customers will outperform a 1% lookalike based on all website visitors.
3. Detailed Targeting (Interest and Behaviour)
While reduced, interest and behaviour targeting still has its place:
- Best for: New accounts with little conversion data, awareness campaigns, and niche markets
- Stack interests: Combine multiple related interests to create more qualified audiences
- Use as suggestions: With Advantage+ audience, your detailed targeting inputs guide the algorithm's initial exploration
4. Broad / Open Targeting
Counterintuitively, running campaigns with no targeting restrictions beyond country and age can deliver excellent results for accounts with:
- Strong conversion history (50+ weekly conversions)
- Complete tracking (pixel + CAPI with high event match quality)
- Compelling creative that naturally pre-qualifies the audience
Building a Targeting Strategy for 2026
The Audience Funnel Approach
Structure your targeting in layers that mirror the customer journey:
- Top of funnel (prospecting): Broad targeting or wide lookalikes. Let your creative do the qualifying. Optimise for engagement or landing page views initially, then shift to conversion optimisation once you have data.
- Middle of funnel (consideration): Engagement custom audiences (video viewers, page engagers, content consumers). Serve educational content and social proof.
- Bottom of funnel (conversion): Website visitor custom audiences, cart abandoners, past purchasers (for repeat purchase). Direct response creative with strong CTAs.
Audience Exclusions
Equally important as who you target is who you exclude:
- Exclude recent purchasers from prospecting campaigns (unless you sell consumables)
- Exclude existing customers from lookalike audiences used for acquisition
- Exclude specific retargeting audiences from prospecting to avoid overlap and inflated costs
Advanced Targeting Tactics
Value-Based Lookalikes
Instead of building lookalikes from all purchasers, create a seed audience of your highest-value customers (top 20% by lifetime value). The resulting lookalike will find prospects more likely to become high-value customers, not just one-time buyers.
Engagement Layering
Combine engagement signals to create highly qualified audiences. For example: people who watched 50%+ of a product video AND visited your website in the past 30 days. This intersection is a warmer audience than either signal alone.
Creative as Targeting
In an era of broad targeting, your creative is your targeting. An ad that speaks directly to new mothers, for example, will naturally attract that audience even without interest targeting. Use specific language, imagery, and scenarios that resonate with your ideal customer to self-select the right audience.
Measuring Audience Performance
Track these metrics at the audience level to understand what's working:
- CPA by audience: Which audiences deliver customers at the lowest cost?
- Frequency: Are you oversaturating any audience? Frequency above 3–4 per week typically signals fatigue.
- Audience overlap: Use Meta's audience overlap tool to ensure your campaigns aren't competing against each other.
For benchmark metrics to compare against, refer to our Facebook Ads benchmarks guide.
Effective audience targeting in 2026 requires a blend of first-party data, algorithmic trust, and creative strategy. If you want expert help building a targeting approach tailored to your business, our Meta Ads management team can help. Book a free strategy session via our Calendly to discuss your audience strategy.
Frequently Asked Questions
Is interest targeting still effective on Facebook?
It can be, particularly for new accounts or niche markets. However, for accounts with sufficient conversion data, broad targeting and lookalike audiences typically outperform interest targeting. Use interests as Advantage+ suggestions rather than hard constraints for best results.
How many conversions does Facebook need to optimise targeting?
Meta recommends approximately 50 conversions per week per ad set for optimal learning. Below this threshold, consider optimising for a higher-funnel event (add to cart, lead) and working back to purchase optimisation once volume increases.
What's the best lookalike audience percentage?
For direct response campaigns, start with 1% lookalikes based on your best customers. As you scale and need more reach, expand to 1–3%. Above 3%, the audience becomes so broad that you lose most of the similarity advantage. Read our complete lookalike audience guide for detailed strategies.
Should I use Advantage+ audience or original audience settings?
For most advertisers, Advantage+ audience (the default) delivers better results because it gives the algorithm flexibility to find converters beyond your defined parameters. Test it against original audience settings with a controlled experiment if you want to validate for your specific account.