3 Paid Social Audience Targeting Strategies That Cut CAC

You are not paying for clicks. You are paying for intent. When that intent is missing, your Cost Per Acquisition (CPA) spikes while your Cost Per Click (CPC) stays flat. That divergence is the quiet killer of paid social budgets.

Most teams launch with "broad" targeting because it feels safe. It is not. It is a leaky bucket.

Why Broad Targeting Is Burning Your Budget

Consider a SaaS company in Austin running Meta ads. They launched with a broad interest target: "Marketing" and "B2B Software." Their CPC was $1.20. Solid. But their CPA hit $180. Why? The audience included students, hobbyists, and competitors. They were paying $1.20 to reach people who would never buy a $2,000/month platform.

This is the classic failure mode. As noted in The Core Click's breakdown of agency strategies, targeting is only one part of the equation, but the ad itself must give the audience a reason to listen. If the audience is wrong, the best creative in the world fails.

The specific failure modes:

1. Lack of Negative Audiences: You are bidding against your own customers. If you do not exclude existing email lists, you waste budget on people who already own the product.
2. The CPC/CPA Divergence: You see low CPCs and think you are winning. You are not. You are winning clicks from people with zero intent. Your CPA tells the truth. Your CPC tells a lie.
3. Wasted Spend: Without segmentation, a significant portion of your spend goes to non-converting segments. You are essentially paying for impressions that do not convert.

The fix is not "better creative." The fix is tighter audience definition.

Strategy 1: The Customer Match Feedback Loop

This is the foundation. You already have data on who buys from you. Use it.

Step 1: Clean and Upload Your List Export your CRM data. You need email addresses and phone numbers. Clean the list first. Remove duplicates and invalid formats.

Step 2: Create the Custom Audience In Meta Business Manager, go to Audiences > Create Audience > Custom Audience > Customer List. Upload your CSV. This creates your "Customer Match" list.

Step 3: Build Lookalikes Now, create Lookalike Audiences based on this list.

• B2B: Start with a 1% lookalike. This is the most similar 1% of the population. It is small, but high-intent.
• B2C: A 3% lookalike often works better. You need volume.

Step 4: Segment by Lifetime Value (LTV) Do not treat all customers the same. Create separate Customer Match lists for:

• High-LTV customers (top 10%)
• Average customers
• Low-LTV customers

Build lookalikes off the High-LTV list. This tells the algorithm: "Find more people like our best customers, not just any customer."

Minimum Requirement: You need at least 1,000 unique contacts for effective modeling. If you have fewer, you are working with noise.

This is not a one-time task. It is a loop. You feed new buyers into the list, and the lookalike refreshes.

Strategy 2: Job Title and Industry Stacking

For B2B, Meta is too broad. You need LinkedIn Campaign Manager.

The mistake is targeting a single attribute. "VP of Marketing" is too broad. "Software Industry" is too broad.

The Stacking Method:

Example: Target: "VP of Marketing" + "Software Development" + "51-200 Employees."

This is a specific, high-intent audience. They have the budget. They have the pain point. They are in the right industry.

The 10,000 Rule: LinkedIn requires a minimum audience size of 10,000. If you stack too many attributes, you drop below this limit. The campaign will not run.

Decision Rule:

1. Start with Title + Industry.
2. Check the estimated audience size.
3. If it is above 10,000, add Company Size.
4. If it drops below 10,000, remove Company Size or broaden the Industry.

This is LinkedIn targeting best practices in action. You are not guessing. You are narrowing.

Strategy 3: Creative-to-Audience Fit Testing

You have the right audience. Now you need the right message.

Worked Example: Here’s a hypothetical example: a 14-person HVAC company in Tucson runs two ads on two segments.

• Segment A: Homeowners, 25-40, interested in "Home Improvement."

  • Ad A: "Pain Point" — "Is your AC breaking down in the July heat? Fix it before the bill spikes."
  • Ad B: "Social Proof" — "Rated 4.9/5 by 500+ Tucson homeowners. See reviews."

• Segment B: Property Managers, 40-60, interested in "Commercial Real Estate."

  • Ad A: "Pain Point" — "Downtime costs you tenants. We keep your units cool with 24/7 monitoring."
  • Ad B: "Social Proof" — "Trusted by 50+ property management firms. See case studies."

The Math:

• Ad A on Segment A: 5% CTR. High.
• Ad B on Segment A: 2% CTR. Low.
• Ad A on Segment B: 1% CTR. Very low.
• Ad B on Segment B: 4% CTR. High.

Creative Fatigue: If you run Ad A on Segment A for 14 days, your CTR will decay. The audience has seen it. They are tired.

Solution: Segment your creative. Use Ad A for homeowners. Use Ad B for property managers. This delays fatigue. You are not repeating the same message to the same people. You are matching the message to the intent.

This is creative-to-audience fit in practice. You are not testing creative. You are testing fit.

Automating the Targeting Refresh Cycle

Static audiences decay. Within 21 days, your lookalikes and custom lists become less relevant. People change jobs. People buy. People stop buying.

The 30-Day Refresh Cycle:

1. Day 1: Upload new Customer Match list.
2. Day 30: Refresh lookalikes.
3. Day 30: Pause underperforming segments.

Autonomous Systems: Manual bid and audience management takes 5-10 hours per week. You can automate this.

The Framework:

1. Monitor: Track CPA by segment.
2. Rule: If CPA is 2x your target for 7 days, pause the segment.
3. Refresh: If a segment has not converted in 14 days, refresh the creative or the audience.

This is automate ad management without the agency overhead. You are not guessing. You are reacting to data.

Start with the Customer Match loop. Build your lookalikes. Stack your LinkedIn targeting. Test your creative fit. Refresh your audiences. Do this, and your CAC will drop. Not by a little. By a lot.

Related reading: Which Paid Social Targeting Optimization Software Actually Lowers CPA? · Precision Targeting Tools for Paid Social · How to Expand Audience on Paid Social Without Collapsing Your Return on Ad Spend

Frequently asked questions

Every post here was written by Vannor. Researched, written and reviewed by the Vannor Engine on our own infrastructure.How the engine works →