Ad targeting: is it really that precise?

Ad targeting is one of the biggest selling points of digital marketing, but how accurate is it really? In this episode, we challenge the assumption that ad targeting is as precise as marketers believe. We discuss the pitfalls of audience segmentation, the over-reliance on machine learning, and the growing importance of creative in a world where targeting is becoming less effective.

Watch the full episode below or listen to the conversation here. Prefer to skim? Check out the top takeaways below.

Top takeaways from this episode

Is ad targeting as precise as marketers think?

• Digital advertising promised the ability to serve the right message to the right person at the right time, but in practice, it's far less accurate than we assume.
• Many audience signals are unreliable—interest groups are often based on outdated or vague data points, and conversion-based targeting can be easily misled.
• Marketers are shifting toward broad audience targeting, letting machine learning find conversions instead of relying on predefined interest groups.

Is AI making ad targeting better or worse?

• AI-driven targeting, like Meta’s Advantage+ and Google’s PMAX, prioritizes efficiency but often relies on low-quality or misleading audience signals.
• Smart bidding strategies can drive results, but they also lead to waste—some marketers unknowingly pay exorbitant CPMs for single impressions on low-quality placements.
• The black-box nature of automated bidding means advertisers don’t always know where their ads are being placed or how targeting decisions are made.

How should marketers approach ad targeting today?

• Simplification is key. Instead of over-complicating targeting with excessive parameters, broad targeting with strong creative can often drive better performance.
• Be wary of exclusion lists. In programmatic, exclusion lists may not be enough. Inclusion lists (pre-vetted, high-quality sites) can better prevent wasted spend.
• Beware of conversion-based targeting pitfalls. Machine learning may optimize for clicks or page views, but that doesn’t always translate to true business impact.

Is creative more important than targeting?

• As tracking and audience precision become less reliable, strong, targeted creative is becoming one of the best ways to “self-select” the right audience.
• AI can assist in creative production, but it lacks the human ability to take big creative risks and drive brand differentiation.
• Over-personalization of creative through AI may backfire—brands should focus on distinctiveness over hyper-customization.

How can marketers measure if their targeting is working?

• Incrementality testing remains one of the best ways to determine if your targeting is truly driving results (we wrote a whole blog on incrementality here if you’d like to check it out).
• Excluding existing customers from campaigns doesn’t guarantee you're only reaching new ones—tests show that customer exclusions only shift targeting slightly.
• Don't take audience insights at face value – Just because an interest group performs well in a campaign doesn’t mean it represents your actual audience.

Additional resources

• As promised, here is a link to The Ehrenberg-Bass Institute for Marketing Science book Kurt mentioned: How Brands Grow.