Detailed Targeting

Detailed targeting is an ad set setting on Meta that lets you narrow your audience beyond location and age: by demographics (e.g. education, job industry, life events), interests (e.g. fitness, real estate, entrepreneurship) and behaviors (e.g. devices used, travel habits). Meta derives these traits from profile information and activity on its platforms.

How it works

• Include: People must match at least one of the selected traits.
• Narrow further: With a second condition, people must match traits from both groups (AND logic).
• Exclude: People with certain traits won't be reached.

With Advantage+ audience, or the option to expand detailed targeting, Meta may also deliver ads outside your selection when it expects better results there. Over the past few years, Meta has removed many options related to sensitive topics like health, religion or political views.

Example

A tax software provider tests two ad sets with $2,000 each: detailed targeting on "self-employed" and "small business" against a broad audience with only location and age. The interest ad set brings in 50 leads at $40, the broad one 70 leads at $28.57. In the CRM, the interest ad set produces 8 customers, the broad one 7. So a customer costs $250 from the interest ad set and $286 from the broad one.

Detailed targeting in lead generation

In lead generation, the key question isn't which audience brings the cheapest leads, but which brings the most customers. Interest audiences are often more expensive but sometimes deliver better-fit leads. Broad audiences work well when the algorithm learns through the Conversion API which leads actually buy.

LeadMetrics attributes leads, deals and revenue to the ad set they came from. In the dashboards you can compare different targeting approaches by closed deals instead of leads alone. The guide to social media KPIs shows which metrics work best for this.

Common mistakes

• Audiences that are too narrow: Stacking several AND conditions makes the audience small and expensive.
• Treating interests as facts: The assignment is Meta's estimate, not reliable information.
• Never comparing: Test detailed targeting against a broad audience instead of using it out of habit.