Table of Contents
Intro:
Three years ago, the standard advice for Meta Ads targeting was to build detailed interest stacks, layer demographics and create tightly defined audience segments that matched your buyer persona. Today that advice is largely obsolete — and the accounts still following it are paying more per lead than the accounts that have moved to broad targeting. Here is why Meta’s algorithm has made interest targeting less effective and what you should be doing instead.
What Changed:
Meta’s algorithm has become significantly more powerful at identifying purchase intent signals independently of advertiser-defined targeting parameters. When you define a narrow interest stack, you are telling Meta to only look for your buyer within a specific subset of the population. When you use broad targeting, you are telling Meta to find your buyer anywhere in the entire eligible population.
At sufficient ad spend, broad targeting almost always finds buyers that interest targeting misses. The algorithm has access to billions of behavioural data points that no interest stack can fully capture. Your buyer might not follow any of the Facebook pages or have any of the interests you associate with your product. But they exhibit purchase intent signals in their browsing behaviour, purchase history and engagement patterns that Meta’s algorithm identifies without you telling it where to look.
When Interest Targeting Still Makes Sense:
Interest targeting remains useful in two specific situations. First, when your pixel has no conversion data and the algorithm has nothing to learn from — a brand new account needs some directional signal to prevent the broadest possible audience interpretation. Second, when your product is genuinely niche and the audience is small enough that broad targeting would spread spend too thin before finding the relevant segment.
For most established accounts with 30 or more conversion events per month, broad targeting or Advantage+ Audience will outperform manually defined interest targeting.
How to Test the Transition:
Run a direct comparison. Duplicate your best-performing interest-targeted ad set and change only the targeting to broad or Advantage+ Audience. Keep the creative, copy, landing page and budget identical. Run both for two weeks with sufficient budget to generate at least 20 conversions per ad set. Let the data decide rather than the assumption.
The targeting debate is mostly over. Meta’s algorithm wins. The new competitive advantage is not who you target — it is what you show them when you get there.
Final Thoughts:
Letting go of granular interest targeting feels counterintuitive. It goes against the instinct to control every variable. But Meta’s advertising system is fundamentally a machine learning platform and machine learning requires freedom to explore. The more you constrain it, the more you limit its ability to find the buyers that your manual targeting would never have identified. Release control of the audience and invest the energy you were spending on targeting into creative strategy instead. That is where the leverage is in 2026.
Contact me for more.
FAQS
Advantage+ Audience is Meta’s fully automated audience selection system where you provide optional audience suggestions and Meta decides who sees the ad based entirely on its own signals. Broad targeting means you leave the audience fields largely empty — no interests, no detailed targeting — but you may define basic demographic parameters like age range or location. Both approaches give the algorithm significantly more freedom than interest-based targeting. Advantage+ Audience gives it the most freedom of all.
For a brand new account, provide some directional signal through Advantage+ Audience suggestions rather than launching completely cold with broad targeting. Without pixel conversion data, the algorithm has nothing to learn from and complete broad targeting can result in spend going to irrelevant audiences during the early data-gathering phase. As the pixel builds 30 to 50 conversion events, you can progressively remove the directional constraints and allow broader exploration.
Broad targeting is less effective for highly niche B2B products because the audience of qualified buyers is genuinely small relative to the total addressable population. Showing a cloud infrastructure management platform ad to everyone on Facebook will produce low conversion rates regardless of how good the algorithm is. For niche B2B, LinkedIn’s job-title targeting or a tightly defined Meta audience based on company size, job function and seniority will outperform broad targeting. Broad targeting works best when the addressable audience is large enough for the algorithm to find meaningful patterns.
Give it at least three weeks before drawing conclusions. When you switch to broad targeting, the algorithm enters a new exploration phase as it learns which segments of the broader population convert for your offer. CPL often increases temporarily during this period before falling below the interest-targeted baseline. If CPL is still higher after 30 days and 50 conversions, consider testing Advantage+ Audience rather than reverting to interest targeting, as the algorithm may need more explicit directional guidance.
Yes and this is often the right approach. Broad targeting with exclusion audiences removes the segments you know are not buyers — existing customers, people who have already converted, irrelevant geographic areas — while allowing the algorithm full freedom within the remaining population. This gives you the benefits of broad exploration without spending budget on audiences that cannot convert. Exclusion audiences are the one type of audience control that almost always improves broad targeting performance.
Leave a Reply