Demographic targeting feels basic compared to behavioral signals and AI-driven optimization. But the right demographic combinations still significantly impact ROAS — both directly and by giving Meta's algorithm better starting points.
Why Demographics Still Matter
Even with Meta's sophisticated machine learning, demographic targeting serves important functions:
- Faster learning: Narrowing demographics helps algorithms find patterns quicker
- Budget efficiency: Excludes demographics that rarely convert
- Creative relevance: Enables demographic-specific messaging
- Compliance: Some products require age restrictions
- Testing: Isolates demographic variables for learning
Age Targeting Strategies
Optimal Age Ranges by Product Type
- Impulse purchases ($20-50): 25-44 typically performs best
- Subscription products: 25-34 shows highest retention
- Luxury/premium: 35-54 has more purchasing power
- Tech/gadgets: 18-44 for adoption, 35-54 for premium tiers
- Health/wellness: Varies significantly by category
Age Range Width
Balance reach against relevance:
- Narrow (10-year span): Higher relevance, faster learning, limited scale
- Medium (15-20 years): Good balance for most products
- Wide (25+ years): Maximum reach, let algorithm optimize
Gender Targeting Insights
When to Target by Gender
- Always target: Gender-specific products (menswear, skincare for men/women)
- Consider targeting: Gift-heavy categories during holidays
- Test first: Products with assumed gender appeal
- Stay broad: Most consumer products perform well with all genders
Gender-Specific Creative
Rather than excluding genders, create gender-specific ad creative within broad targeting. Let the algorithm match creative to audience. Learn more about creative-based targeting in our broad targeting guide.
Location-Based Combinations
Geographic Layering
- Country level: Broadest reach, varies by market maturity
- State/region: Useful for regional products or shipping constraints
- DMA/metro: Target high-density markets with better unit economics
- Radius targeting: Local businesses, event promotion
Location + Income Proxies
Meta doesn't offer direct income targeting, but location serves as a proxy:
- Target zip codes with higher median household income for premium products
- Exclude low-performing regions after initial testing
- Layer with interest signals for better income correlation
High-Performing Combinations
Ecommerce Standard
- Age: 25-54
- Gender: All (unless product-specific)
- Location: Top-performing metros + national
- Combined with: Engaged shoppers behavior
B2B Lead Generation
- Age: 28-55
- Gender: All
- Location: Business hubs and tech corridors
- Combined with: Job title or business decision maker behaviors
See our B2B targeting guide for more.
Local Services
- Age: 30-65 (homeowners)
- Gender: All
- Location: Radius around service area
- Combined with: Homeowner interests, income proxies
Subscription/SaaS
- Age: 25-44
- Gender: All
- Location: English-speaking markets
- Combined with: Technology interests, early adopter behaviors
Testing Demographic Variables
Isolation Testing
Test one demographic variable at a time. Use the scientific method approach:
- Create identical campaigns with single variable changes
- Age test: 18-34 vs 35-54 vs 55+
- Gender test: All vs Male vs Female (if relevant)
- Location test: Metros vs national vs specific regions
Analyzing Results
- Compare CPA and ROAS across demographic segments
- Look at conversion rate differences, not just volume
- Factor in LTV differences by demographic
- Consider scale potential of each segment
Common Demographic Mistakes
Over-Narrowing
Don't stack too many demographic restrictions. This limits Meta's algorithm and reduces auction efficiency. Our guide on narrow audience signs explains how to diagnose this.
Assumption-Based Targeting
Test your assumptions. Products often perform well outside expected demographics:
- Gaming products selling to 35-54 age range
- Skincare appealing across genders
- Premium products performing in unexpected regions
Ignoring Lifetime Value
Some demographics have higher initial CPA but better LTV. Optimize for long-term value, not just first-purchase ROAS.
How ROASPIG Helps
Demographic optimization requires testing and analysis. ROASPIG streamlines the process:
- Demographic Performance Analysis: Break down ROAS by demographic segments
- A/B Testing Framework: Structured demographic isolation tests
- LTV Integration: Factor lifetime value into demographic decisions
- Creative Matching: Generate demographic-appropriate creative at scale
- Audience Recommendations: AI-suggested demographic combinations
The Bottom Line
Demographic targeting isn't obsolete — it's foundational. The best advertisers use demographics strategically: narrow enough to improve efficiency and enable relevant creative, broad enough to give Meta's algorithm room to optimize.
Test your assumptions, analyze results by demographic segment, and remember that the right combination varies by product, market, and campaign objective. Demographics are a starting point, not a destination.
Frequently Asked Questions About Demographic Targeting
It varies by product, but 25-54 is the most common high-performing range. Impulse purchases often perform best with 25-44, while premium products see better results with 35-54.
Only if your product is genuinely gender-specific. For most products, leaving gender broad and using gender-specific creative within broad targeting performs better.
Test one variable at a time using identical campaigns. Create separate ad sets for different age ranges or locations, run them simultaneously, and compare CPA and ROAS after sufficient data.
Meta doesn't offer direct income targeting in most markets. Use location-based proxies (affluent zip codes) or layer with interests that correlate with income levels.
Over-narrowing restricts Meta's algorithm and reduces auction efficiency. The algorithm often finds high-value users outside your assumptions. Test broader targeting against narrow.