Meta Ads Audience Targeting: Advanced Strategies That Lower Your CPL

Cost per lead is the number that keeps marketers up at night. You can have a beautiful creative, a compelling offer, and a perfectly optimised landing page — and still watch your CPL creep higher every week. More often than not, the culprit isn't the ad itself. It's the audience.

Meta's advertising platform gives you an extraordinary amount of targeting power, but most businesses only scratch the surface. They set up a broad interest-based audience, let the campaign run, and wonder why leads are expensive. The businesses that genuinely win on Meta are the ones that understand how to layer, sequence, and constantly refine their audience strategy.

At Praxxii Global, we manage Meta Ads campaigns across a wide range of industries and budgets. These are the advanced targeting strategies we use to bring CPL down — sometimes dramatically — for our clients.

Understand Why Your CPL Is High Before You Fix It

Before making any changes, it's worth diagnosing where the problem actually lives. High CPL on Meta usually stems from one of three places: your audience is too broad and attracting low-quality clicks, your audience is too narrow and you're hitting frequency fatigue, or you're targeting the right people at the wrong stage of the funnel.

Pull your campaign data and look at CPL by audience segment, placement, and ad set. If certain ad sets are consistently expensive with low conversion rates, the issue is almost certainly audience mismatch. If your CPL was low but has recently increased, frequency is likely the issue — your audience has seen the ad too many times and is no longer responding.

With a clear diagnosis, you can apply the right fix rather than guessing.

1. Build Lookalike Audiences from Your Best Customers, Not Just Any Customers

Lookalike audiences are one of Meta's most powerful tools, but the quality of your lookalike depends entirely on the quality of your seed audience. Most advertisers create lookalikes from their full customer list, which often includes low-value or one-time buyers who don't reflect the customers you actually want more of.

Instead, build your seed audience from your highest-value customers specifically. This means filtering your customer list to include only those who:

• Have made repeat purchases or renewed their contract
• Have a lifetime value above a certain threshold
• Converted within a short time frame (fast decisions signal high intent)

Upload this refined list to Meta as a Custom Audience, then create a 1% lookalike from it. A 1% lookalike gives you Meta's tightest match — the people whose online behaviours most closely resemble your best customers. As you scale, you can expand to 2–3% lookalikes, but always start tight.

2. Layer Lookalikes with Interest and Behaviour Filters

A 1% lookalike audience alone can still contain a lot of noise. You can sharpen it significantly by layering additional targeting on top — narrowing the audience to people who also match specific interest or behaviour signals.

For example, if you're running ads for a B2B SaaS product, your targeting might look like this:

• Lookalike of your top customers (1%)
• And job titles: Marketing Manager, Growth Lead, Head of Marketing
• And interests or behaviours: small business owners, active business travellers, engaged shoppers

This intersection approach reduces your audience size but increases relevance dramatically. More relevant impressions means higher click-through rates, lower CPMs, and a lower CPL overall. Be careful not to layer so many filters that your audience becomes too small to exit the learning phase. Aim for an audience size of at least 200,000–500,000 for most campaigns.

3. Use Engagement Custom Audiences as a Mid-Funnel Bridge

Many businesses run top-of-funnel awareness ads and bottom-of-funnel conversion ads but have nothing in between. This leaves a gap where warm, interested prospects fall away simply because they weren't quite ready to convert after a single exposure. Meta lets you build Custom Audiences from people who have engaged with your content — these are your mid-funnel bridge. You can create audiences from:

• People who watched at least 50% of one of your videos
• People who opened or completed your Lead Form
• People who engaged with your Instagram profile in the last 30, 60, or 90 days
• People who visited your Facebook page and interacted with posts

Run a separate retargeting campaign targeting these warm audiences with a more direct conversion offer — a free consultation, a limited-time deal, or a case study. Because these people already know your brand, your CPL from this segment will typically be a fraction of your cold audience CPL. Our Meta Ads management services include full-funnel audience architecture as standard — it's one of the biggest levers we pull to reduce CPL for new clients.

4. Stack Website Retargeting by Page and Recency

Not all website visitors are equal. Someone who visited your homepage and left immediately is very different from someone who spent three minutes reading your pricing page. Treating them the same way in your retargeting is a common mistake that drives up CPL.

Segment your website retargeting audiences by:

Page visited:

Pricing page visitors, service page visitors, and blog readers should each get different ads tailored to where they are in the decision process.

Recency:

Visitors from the last 7 days are far warmer than visitors from 60 days ago. Create separate ad sets for 0–7 days, 8–30 days, and 31–90 days, with decreasing bid levels as recency drops.

Exclusions:

Always exclude people who have already converted. Showing acquisition ads to existing customers wastes budget and creates a poor experience. This granular approach to retargeting ensures the right message reaches the right person at the right moment — the formula for low CPL.

5. Leverage Advantage+ Audience — But with Guardrails

Meta's Advantage+ Audience (formerly Broad Targeting with AI optimisation) has become increasingly powerful as Meta's algorithm has improved. When given enough conversion data, it can often outperform manually defined audiences by finding conversion signals you wouldn't have predicted. However, using Advantage+ without any constraints is a risk. We recommend using it with what Meta calls "audience suggestions" rather than leaving it completely open. Provide a suggested audience — your lookalike or interest-based targeting — and allow Meta to expand beyond it when it identifies better opportunities. This hybrid approach gives the algorithm room to optimise while keeping your spend from drifting into completely irrelevant territory. Monitor CPL closely in the first two to three weeks and be prepared to tighten the guardrails if the algorithm starts pulling in low-quality leads.

6. Test Audience Segments Scientifically with A/B Splits

One of the most reliable ways to lower CPL over time is systematic audience testing. Rather than running multiple audience variations within the same campaign (which causes Meta's algorithm to compete with itself), use Meta's built-in A/B Test tool.

Test one variable at a time:

• Lookalike vs. interest-based audience
• 1% lookalike vs. 3% lookalike
• Broad demographic targeting vs. narrowed demographic targeting
• Retargeting window of 14 days vs. 30 days

Each test should run long enough to reach statistical significance — typically a minimum of seven days and at least 50 conversions per variant. Document your findings and build on them. Over time, this compounding of small improvements leads to dramatic reductions in CPL.

7. Refresh Your Creative Before You Refresh Your Audience

Here's a counterintuitive truth: if your CPL is climbing, the audience often isn't the first thing to fix. Creative fatigue — where your audience has seen your ads so many times they stop engaging — is frequently the culprit, and it's often misdiagnosed as an audience problem. Check your frequency score. If the average person in your audience has seen your ad more than three times in the past seven days, creative fatigue is almost certainly dragging down performance. Before restructuring your entire audience strategy, introduce new ad formats and fresh creative angles. Fresh creative against the same audience can restore strong CPL figures quickly and cheaply. Only if new creative fails to recover performance should you look at audience restructuring.

Putting It All Together

Reducing CPL on Meta Ads is not a single action — it's a system. It requires building quality seed audiences, structuring your funnel properly, retargeting with precision, giving the algorithm the right signals, and testing continuously.

The businesses that master this process don't just lower their CPL — they build a scalable, predictable lead generation machine that compounds in value over time as each optimisation builds on the last.

If your Meta Ads are delivering leads at a cost that makes growth feel difficult, it's worth getting a professional eye on your audience strategy. Reach out to the team at Praxxii Global for a free audit and let us show you exactly where your targeting can be tightened — and what that could mean for your cost per lead.