Marketing measurement faces an identity crisis. The multi-touch attribution models that became standard over the past decade relied on cross-site tracking capabilities that are rapidly disappearing. Cookie deprecation, iOS privacy changes, and evolving regulations have degraded attribution data quality. Meanwhile, executives still expect marketers to prove ROI.

This moment of disruption is also an opportunity. The attribution models we’re losing were never as accurate as we pretended. Their false precision often led to poor decisions—overinvesting in lower-funnel tactics that were easy to measure while undervaluing brand and awareness activities that drove demand. The new measurement landscape can actually be better, if we approach it thoughtfully.

What’s Breaking

Several measurement capabilities are degrading simultaneously:

Cross-device tracking becomes unreliable as identifiers fragment. The same person on their phone, laptop, and work computer increasingly looks like three different people.

View-through attribution for display advertising loses signal as third-party cookies disappear. You can’t credit an ad view for a later conversion if you can’t connect the two events.

Multi-touch journey tracking becomes gapped. The complete customer journey view that underpinned sophisticated attribution models now has missing chapters.

Platform attribution becomes more siloed. Each ad platform reports its own view of conversions, but reconciling across platforms becomes harder.

The data that powered marketing mix optimization is becoming less complete and less reliable.

Emerging Measurement Approaches

Forward-thinking marketing teams are adopting new approaches suited to this environment:

Marketing Mix Modeling (MMM) Renaissance

MMM uses statistical analysis of aggregate data—spend by channel correlated with outcomes over time—to estimate channel effectiveness. It doesn’t require user-level tracking, making it privacy-durable.

Modern MMM has evolved beyond the quarterly reports of traditional approaches:

  • Faster refresh cycles enabled by better computing and automation
  • More granular inputs including creative types, audiences, and tactics
  • Bayesian approaches that incorporate prior knowledge and produce confidence intervals
  • Always-on dashboards rather than periodic studies

MMM can’t tell you about individual customer journeys, but it can reveal channel effectiveness patterns that inform budget allocation.

Incrementality Testing

Incrementality testing directly measures what happens when you change marketing activity. Hold out a portion of your audience from a campaign and compare outcomes to the exposed group. The difference represents incremental impact.

Types of incrementality tests:

  • Geographic holdouts: Compare regions with and without marketing activity
  • Audience holdouts: Randomly exclude segments from campaigns
  • Spend variation: Systematically vary spend levels and measure response
  • Platform-provided lift studies: Use ad platform experimentation tools

Incrementality testing provides high-confidence causal evidence but requires sufficient scale and patience to run properly.

Self-Reported Attribution

Ask customers how they heard about you. It sounds simplistic, but self-reported attribution captures influences that digital tracking misses entirely—word of mouth, podcast mentions, social media browsing, conference conversations.

Implement self-reported attribution through:

  • “How did you hear about us?” fields on forms
  • Post-purchase surveys
  • Sales conversation documentation
  • Customer interview programs

Self-reported data has biases (recency effects, socially desirable responses), but it provides signal about channels that digital attribution ignores completely.

Triangulation Approach

No single measurement method is perfect. The most robust approach triangulates across multiple methodologies:

  • Use MMM for overall budget allocation guidance
  • Run incrementality tests to validate MMM findings for high-spend channels
  • Incorporate self-reported attribution to capture unmeasured influences
  • Maintain digital attribution where it still works, while acknowledging its limitations

When multiple methods point in the same direction, confidence increases. When they conflict, investigation is needed.

Practical Implementation

Start with Clear Questions

Measurement should answer specific business questions:

  • How should we allocate budget across channels?
  • What’s the incremental impact of this campaign?
  • Which creative approaches drive better outcomes?
  • What’s the optimal spend level for this channel?

Different questions require different measurement approaches. Don’t try to build one model that answers everything.

Invest in Data Infrastructure

Quality measurement requires quality data:

  • Clean, consistent spend data across all channels
  • Reliable outcome data (conversions, revenue, pipeline)
  • Proper time alignment between spend and outcomes
  • Integration across marketing and sales systems

Data infrastructure gaps will undermine any measurement methodology.

Build Measurement Literacy

Ensure stakeholders understand what measurement can and can’t tell them:

  • All models have assumptions and limitations
  • Precision varies across channels and tactics
  • Directional guidance may be more achievable than exact ROI
  • Testing provides the highest confidence evidence

Unrealistic expectations for measurement precision lead to poor decisions when those expectations aren’t met.

Accept Uncertainty

The era of (falsely) precise attribution is ending. Effective measurement in the new environment means becoming comfortable with ranges, confidence intervals, and directional insights rather than exact numbers.

This is actually more honest than the false precision of click-based attribution. A range that’s accurate is more valuable than a point estimate that’s wrong.

The Path Forward

Marketing measurement is evolving from tracking-dependent attribution to a more diverse toolkit of approaches. This transition is challenging but ultimately positive. The new measurement paradigm can capture influences that digital tracking always missed while respecting user privacy.

Organizations that build sophisticated measurement capabilities—combining MMM, incrementality testing, and self-reported attribution with remaining digital tracking—will make better marketing decisions than those clinging to increasingly broken attribution models.

The measurement practices that serve you well in 2024 and beyond will look quite different from those of the past decade. Start adapting now.