Why Marketing Measurement Is Broken Despite Tech Innovation – News Round Up: 10/17-10/24

by Jeff Licciardi

Your Attribution Dashboard Lies: Why Better Tech Made Marketing Measurement Worse – News Round Up: 10/17-10/24

Here’s the uncomfortable truth nobody wants to say out loud: Marketing measurement has gotten worse as the technology has gotten better. We’ve gone from simple, directional metrics to sophisticated attribution models that give us the illusion of precision while obscuring what actually drives revenue. This week’s industry news exposes the cracks in our measurement mythology—from retail media’s maturity crisis to AI ROI questions that nobody can quite answer.

The pattern is clear: We’re drowning in data but starving for insight. Every platform promises closed-loop attribution. Every vendor sells incrementality testing. Every dashboard shows “lift.” Yet CMOs are more confused than ever about what’s actually working. The tech evolved faster than our ability to use it honestly.

Retail Media’s Identity Crisis Reveals the Core Problem

Retail media was supposed to be the golden child of marketing measurement—closed-loop attribution paradise where you could finally prove that ads drove purchases. But as Digiday reports, the easy dollars are gone, and the industry is facing a maturity crisis that exposes the fundamental flaw in our measurement obsession.

The issue isn’t that retail media doesn’t work. It’s that we’ve been measuring the wrong things. Advertisers are hitting diminishing returns because platforms optimized for what’s measurable (last-click purchases) rather than what’s valuable (incremental demand creation). When everything is attributable, nothing is strategic.

The shift to “full-funnel” retail media is industry-speak for “we finally realized we’ve been measuring bottom-funnel harvesting and calling it growth.” Meanwhile, smaller networks are trying to differentiate, but they’re all selling variations of the same promise: better measurement, better targeting, better attribution. The tech keeps improving. The strategic thinking doesn’t.

AI ROI: The Questions Nobody Can Answer

Speaking of measurement problems, let’s talk about AI. Ad Age’s deep dive on where marketers are actually seeing AI ROI reveals something fascinating: The wins are almost entirely operational (efficiency, cost reduction, speed) rather than strategic (better outcomes, incremental growth, competitive advantage).

This matters because we’re making the same mistake we made with marketing automation in 2015. We’re measuring AI success by how much faster we can produce mediocre content, not whether AI fundamentally changes what’s possible. Marketers are warming to AI, but the creative challenges and legal risks remain unresolved because we’re optimizing for the wrong metrics.

The disconnect is stark: OpenAI has done a complete 180 on advertising, now embracing it after years of resistance. But what are we measuring when we evaluate AI-generated advertising? Click-through rates? Cost per acquisition? These metrics were designed for a different era and a different medium. We’re measuring AI with last decade’s KPIs and wondering why the results feel hollow.

CTV: The Attribution Black Hole Nobody Admits

Connected TV advertising is booming—growth projections remain strong for 2025—but here’s what the industry won’t say: CTV attribution is mostly theater. The “deterministic” matching everyone brags about is probabilistic at best, and the closed-loop measurement is closing loops that may or may not exist.

Q3 earnings from TV industry leaders show robust ad spend, but the measurement methodology hasn’t fundamentally changed from linear TV—we’ve just added more decimals to the Nielsen ratings and called it innovation. The technology got better. The truth didn’t.

The real tell is how the industry talks about CTV measurement: Always promising, never quite delivered. Every quarter brings new measurement partnerships, new attribution models, new cross-device solutions. If the measurement was actually working, we’d stop talking about it and start showing results.

Social Platforms: Where Measurement Goes to Die

The social media landscape offers the clearest evidence that better measurement tools don’t equal better understanding. Facebook’s declining relevance happened despite having arguably the most sophisticated advertising measurement system ever built. Turns out you can measure everything and still miss what matters: cultural relevance.

Meanwhile, TikTok has become Gen Z’s primary search engine, disrupting Google without obsessing over search attribution models. The platform won by creating value, not by measuring it better. The lesson everyone is ignoring: Sometimes the best measurement strategy is building something people actually want.

What This Means Going Forward

The marketing industry has confused measurement precision with strategic clarity. We’ve built increasingly sophisticated attribution systems that tell us exactly how we got here while providing zero insight into where we should go next. The dashboards are beautiful. The insights are shallow.

The solution isn’t better measurement technology—we have plenty. It’s better measurement philosophy. We need to stop measuring everything and start measuring what matters. That means accepting uncertainty, embracing directional metrics for brand building, and admitting that some of the best marketing can’t be precisely attributed.

Until then, we’ll keep getting exactly what we’re measuring: incrementally optimized mediocrity with excellent dashboards. The tech will keep improving. The results won’t.

Sources & References