For ten years, CMOs have been promised that tech will solve measurement: yesterday attribution and dashboards, today AI.
For ten years, CMOs have been promised that tech will solve measurement. Yet many CMOs are discovering a more nuanced reality: AI hasn’t fixed measurement. It has made failed systems more dangerous, because it can generate unfounded confidence and precipitate bad decisions before they even have time to question them.
The measurement gap that no one dares to recognize
If you ask CMOs what they care about, measurement is no longer at the top of the list. They will talk about AI, talents or speed of execution. Yet the measurement challenge has not gone away. AI has simply made ignoring the fundamentals more costly, because measurement is the very infrastructure it relies on. This is where the problem lies.
The modern customer journey is fragmented between mobile apps, web, CTV, retail media and emerging platforms that did not exist five years ago. Too many systems still treat mobile as just another channel, not the connective tissue of the entire journey. Result: seemingly exhaustive data, but full of gray areas. Conversions seem disconnected, journeys appear linear when they’re not, and performance signals over-index on what’s easy to measure rather than what actually drives value. What has changed is that AI now relies on this void.
Why AI makes a bad metric worse
AI does not reason: it infers. And she does it based on the data we provide her. If this data is incomplete, biased toward certain channels, or lacking key behavioral signals, it amplifies the problem instead of correcting it.
AI systems are remarkably effective at creating a sense of certainty. They produce forecasts, recommendations and optimizations that appear accurate and robust. But confidence is not accuracy. When critical signals are missing, AI fills the gaps with hypotheses that strengthen over time. Budgets are reallocated, strategies become frozen, and the feedback loop becomes self-reinforcing, making any step back complex and expensive.
For a CMO, the fundamental question is simple: does our measurement infrastructure produce data reliable enough to drive automated decisions? Can you track a customer’s movement between different environments? Connect exposure, engagement and results across devices and channels? Distinguish real behavior from a modeled conjecture? Without it, AI becomes a multiplier of ambiguity, and the cost of failing measurement grows exponentially.
Mobile as a center of gravity
For the majority of consumers, the center of gravity is mobile. This is where the identity is strongest, the commitment strongest and the intention most readable, even if the final transaction is concluded elsewhere. This is also where marketers have learned to measure with less: fewer deterministic identifiers, stricter consent requirements, constant evolution of platforms. These mobile quality standards should serve as a benchmark for any omnichannel measurement.
In too many technology stacks, mobile measurement remains an afterthought: conventions inherited from the web era adapted to applications, rather than native standards designed for today’s constraints. Without a reliable anchor, omnichannel measurement is just a patchwork of proxies and assumptions, and AI optimizes for what platforms can easily observe, not what customers actually do.
The roadmap for the CMO of tomorrow
CMOs must properly sequence their AI adoption. It starts with direct questions about their measurement infrastructure:
Where are our blind spots between channels and devices?
Which decisions are based on modeled assumptions rather than observed behavior?
What data do we treat as a “source of truth”, and why?
Are our systems designed for automation, or just for reporting?
From there, the priority should no longer be on adding tools, but on strengthening the infrastructure on which AI will rely. This involves prioritizing data that reflects real customer behaviors, connecting journeys end-to-end rather than channel by channel, and preparing teams to work with AI as a decision partner.
CMOs are at a crossroads. They can continue to assemble reports by channel and feed AI with a partial view of reality, or make measurement the foundation on which their AI strategy truly rests. Those who make this choice are not just reducing risks: they are giving themselves the means to take full advantage of AI. The priority is no longer to add tools, but to strengthen the infrastructure, connect end-to-end journeys and prepare teams to work with AI as a decision-making partner. A solid measurement infrastructure, anchored in rigorous mobile standards, transforms AI into a real growth lever. This is the difference between an organization that experiences acceleration and an organization that drives it.
