Marketing Has a Measurement Problem - But Not the One We Think
By: Sonia Chung
Over the past two decades, marketing has been in a constant state of reinvention. Marketers digitized media, went mobile, and operationalized performance. Teams went from using third-party to first-party data; companies built data lakes. Attribution was adopted. AI was layered in. Each evolution promised greater precision and greater accountability.
For years, the industry rallied around four central ambitions: capturing richer, more granular data; connecting signals across channels and across the customer journey; allowing data to drive decisions; and enabling real-time optimization. In many ways, marketers have accomplished those goals. Digital transformation and consumer data gave us unprecedented visibility. Platforms have connected signals across touchpoints. Automation made optimization continuous rather than episodic.
And yet, despite this progress, marketers find themselves debating the same foundational questions:
Can we measure accurately? That requires separating real impact from noise across marketing activity, external factors, and baseline demand.
Are we truly driving incrementality? That requires understanding whether we are generating net-new sales/revenue or simply capturing existing intent.
Are we building long-term growth or merely short-term lift? Immediate spikes in performance are not enough — growth must be durable and sustainable.
The uncomfortable truth is that marketing does not have a data problem. It has an integration and interpretation problem. Marketers have become exceptionally good at collecting signals — especially in this new era shaped by AI. However, marketers are far less effective at connecting those signals in ways that inform enterprise-level growth decisions.
The Industry Has Become Structurally Siloed
The industry itself has become structurally siloed. Modern marketing is organized around platforms: Google, Meta, retail media networks, programmatic ecosystems, and CRM systems. Each platform comes with its own dashboard, its own attribution logic, and its own optimization model. These platforms are powerful, but they are also inherently biased. They are designed to demonstrate their own value, leaning into their own strengths. Marketers end up over indexing on single channel impact (slices of the lie) and not driving broader impact (the full pie).
Marketing organizations have mirrored that structure. Teams align around channels. Measurement aligns around platforms. Expertise becomes increasingly specialized and increasingly narrow. The result is that tactical campaign optimization thrives while strategic, holistic growth impact lags behind. We’ve all seen this: You have a high ROAS across your digital channels but still not hitting quarterly revenue goals.
Persistent tensions continue to define the landscape. Brand-building efforts are often pitted against performance marketing. Short-term efficiency metrics such as CPL, CAC, and CTR compete with longer-term measures like LTV and profitability. Organizations chase perfect certainty in measurement when, in reality, they are operating in probabilities. Data abundance creates overload, yet decision clarity remains elusive because the question is not how much data we collect, but which data truly matters.
Despite unprecedented access to information, marketing still struggles to justify investment at the leadership table. Conversations remain anchored in channel-level metrics rather than enterprise value impact — revenue durability, margin expansion, risk mitigation, and shareholder value.
Incrementality has emerged as the industry’s “holy grail,” not because we lack metrics, but because we are attempting to isolate causal impact within a multi-touch, omnichannel system that was never designed to deliver perfect certainty. Marketing is fundamentally a probabilistic growth engine, yet leadership often expects deterministic answers. That mismatch is where trust begins to erode.
Accepting Complexity Is the Competitive Advantage
The future will not belong to organizations that eliminate complexity; it will belong to those that embrace it. Marketing leaders must move beyond the pursuit of perfect attribution and instead build integrated growth systems. This requires understanding platform mechanics, developing bottoms up analyses to match the top-down goals, aligning measurement with financial drivers, and organizing data across both short- and long-term horizons. It also requires acknowledging that marketing operates in probabilities and design systems that manage risk and return accordingly.
In a world of infinite dashboards and accelerating AI capabilities, competitive advantage will not come from adding more tools. It will come from the ability to connect foundational data mechanics to business outcomes across the entire customer journey and across the full P&L.
The role of the CMO is evolving accordingly. Increasingly, it requires technological fluency as well as strategic acumen. While the industry was built platform by platform, it is now transforming in real time through AI and integrated systems. The next chapter of leadership will not be defined by who has the most data, but by who can translate that data into sustainable, enterprise-level growth.
More Insights
