LiveRamp and the Marketing + Media Alliance today published joint research showing that measurement errors most marketers never test for can make profitable advertising look like a loss, pushing budgets in the wrong direction. The report, titled The Missing Piece: Improving Confidence in Marketing Measurement, was presented today at the MMA CMO + CEO Summit in Santa Barbara, California.

The central claim is uncomfortable for an industry that has spent a decade chasing higher match rates and bigger datasets. According to LiveRamp, poor identity precision collapsed campaign return on investment by approximately 70% in the study's simulations, a distortion severe enough to get a working campaign cancelled. The problem was not too little data. It was data linked to the wrong person, or exposures that quietly went missing in a pattern that standard quality checks fail to catch.

LiveRamp, the San Francisco data collaboration company listed on the New York Stock Exchange under the ticker RAMP, ran the analysis in collaboration with the MMA using synthetic data and its own technical staff. The stated goal was to isolate common measurement problems and quantify what each one does to attribution, incrementality, and the budget decisions that follow. The announcement was made on July 20, 2026.

Two failure modes, unequal damage

The research splits measurement risk into two distinct categories, and the distinction matters because they behave differently and demand different fixes.

The first is missingness: real ad exposures that were served and seen but never recorded in the measurement system. The data has gaps that underrepresent what actually happened. The second is identity mismatch: exposures that were recorded but attributed to the wrong individual. Here the data looks complete, and the errors hide inside it.

The report frames these against two analyses that practitioners routinely conflate. Attribution asks which ads led to a conversion and estimates each touchpoint's contribution. Incrementality asks a harder causal question: how much did the advertising actually cause conversions that would not have happened otherwise. Identity resolution failures hit each one differently, and understanding which analysis a given tool performs determines how vulnerable its output is.

That framing echoes a recurring theme in industry measurement work. The Interactive Advertising Bureau and IAB Europe defined incrementality in commerce media guidelines released in November 2025 as the causal impact of marketing, the additional outcomes directly driven by a campaign compared with what would have occurred without it, explicitly distinguishing it from attribution and return-on-ad-spend calculations that show outcomes rather than causality.

The pattern of loss matters more than the volume

The missingness findings upend a common assumption. Random data loss, it turns out, is largely harmless to channel rankings. Removing 20% of impressions at random across all publishers preserved the baseline ranking entirely. Even a frequency-dependent scenario, where heavy users lost more impressions as cross-device tracking broke, stripped out 26.9% of impressions and still preserved the ranking.

The danger lies in non-random loss. When the missing impressions were concentrated among converters, the kind of loss caused by privacy controls or checkout flows that fail to fire a pixel, channel rankings reversed at an aggregate volume of roughly 1%. A rounding error in most audits produced a full ranking reversal.

Worse, the distortion was invisible to the diagnostics marketers rely on. The model's own accuracy metric, the area under the curve or AUC, actually improved under outcome-correlated loss even as the underlying ranking corrupted. A team watching its dashboard would have seen the numbers get better while the answer got wrong.

The simulations were not built on toy data. According to the report's methodology, they used real impression logs comprising 1.9 million exposures across 147,941 users and four publishers, paired with 12,956 transaction outcomes and a 1.82% empirical conversion rate. The attribution model was a logistic regression trained on per-publisher impression counts, with a baseline ranking across the four publishers of T over Y over G over M.

When 50% precision cancels a $1.50 campaign

The identity findings are where the headline number comes from, and the mechanism is worth spelling out because it explains why a high match rate offers no protection.

In the incrementality simulations, LiveRamp and the MMA held true campaign performance constant at a 25% lift and $1.50 return, then tested how different measurement designs read that ground truth under each failure mode. In the identity mismatch scenario, they set campaign reach at 30% and identity precision at 50%, meaning half of all matched users were linked to the wrong person.

The result: relative lift, truly 25%, was measured at roughly 6.8%, a collapse of about 73%. Return on investment, truly $1.50, was measured at $0.43. On paper the campaign appeared to lose 57 cents on every dollar spent. The likely decision that follows such a reading is cancellation. A profitable campaign killed by a measurement artifact.

The report is careful to label the 50% precision figure an illustrative lower bound used to show how the mechanism propagates, not a typical condition for campaigns running on deterministic, people-based identity graphs. The mechanics are simple once stated. False positives, unexposed people wrongly labeled as treated, dilute the treatment group. False negatives, exposed people wrongly labeled as control, inflate the baseline. Both biases push measured results downward, and they compound.

This is the finding that reframes a decade of vendor marketing. Match rate, the share of records a system can link, has long been the headline metric identity providers compete on. The report's conclusion is blunt: a high match rate is insufficient if the matches themselves are wrong. Precision, not recall, is what governs measurement integrity. As the report puts it, matching that feels like progress may not be built for precision.

That distinction lands amid a wider reckoning over whether current systems can be trusted at all. Research covered by PPC Land in February 2026 found that broken measurement systems leave as much as $32 billion in value unrealized, with between 67% and 76% of buy-side decision-makers using incrementality tests, attribution, or marketing mix models yet consistently finding those approaches underperform against their core promises.

Which designs survive

Not every measurement method proved equally fragile, and the report's third finding points to which designs hold up under pressure.

Randomized controlled trials using an intent-to-treat design, counting everyone assigned to a test regardless of confirmed exposure, remained directionally correct across every missingness scenario tested. RCT intent-to-treat held return at the ground-truth $1.50 even in the outcome-correlated scenario that broke everything else. An RCT restricted to reached-only users, by contrast, drifted to $0.91 under the same conditions, because filtering to confirmed exposures inherits the identity graph's errors directly.

Quasi-experimental and modeled designs fared worst. In the outcome-correlated loss scenario, a design of that class read a truly profitable campaign, $1.50 return, as a loss of $2.65, flipping the sign entirely on a campaign that was working.

The report groups measurement approaches into a fidelity hierarchy. At the top sit gold-standard randomized experiments with placebo or ghost ads, resilient to many forms of missingness though still vulnerable to identity mismatch at the assignment stage. In the middle sit silver-tier quasi-experimental models comparing exposed against unexposed groups, susceptible to both failure modes and best suited to in-flight optimization rather than go/no-go decisions. At the bottom sit bronze-tier fixed-formula methods, last-touch, even-credit, U-shaped, time-decay, where credit is assigned by rule rather than evidence, and where the heuristics create distortions that missingness and mismatch only compound.

One caveat runs through the intent-to-treat resilience: its protection depends on test assignment staying in lock-step with the media owner's actual delivery. Identity mismatch at the assignment stage can still compromise even a well-designed RCT.

A diagnostic marketers can run

Rather than leaving the problem abstract, the report supplies a screening test. The Odds Ratio compares the odds of conversion for exposed users against the odds for unexposed users, producing a normalized view of lift relative to the organic baseline.

An odds ratio above 1.0 indicates exposed users are more likely to convert, the expected pattern when targeting and advertising work, though the report cautions this does not prove advertising caused the outcome, since good targeting alone pushes the index above 1. An odds ratio below 1.0 is a red flag for measurement, targeting, or selection problems, and often signals missingness. In the report's worked example, a 2-by-2 table of campaign figures produced an odds ratio of 1.32, meaning exposed users were roughly 32% more likely to convert, a directional signal only.

The report positions three concrete checks. First, reconcile publisher-reported delivery against measurement logs to assess missingness, paying particular attention to regional and trade-area delivery where audience files skew by ZIP code. Second, ask measurement partners how they validate identity precision, through people-based deterministic verification for example, rather than accepting match scale alone. Third, stress-test measurement for non-random gaps, because most audits simulate only random loss and therefore never probe the failure mode that matters most.

The report is explicit about its own limits. Its findings are directional, drawn from synthetic-data simulations, and the magnitudes require real-world validation before any firm conclusions.

Why this matters for the measurement debate

The research arrives at a moment when the tools marketers use to prove advertising works are under sustained strain. Privacy regulation, third-party cookie deprecation, identity fragmentation, and cross-channel measurement gaps have scattered signal across disconnected systems, and marketers are absorbing all of that while preparing for AI-driven transitions.

The commercial subtext is unmistakable. The report's remedy is data collaboration, identity solutions and data clean rooms that enable unified, de-duplicated measurement, which is precisely LiveRamp's business. Christine Grammier, VP of Product at LiveRamp, stated in the announcement that "when it comes to understanding your customers across channels, and using identity and cutting-edge measurement tools, data collaboration is unparalleled for helping marketers use data with precision, and marketers can select solutions like LiveRamp to quickly and effectively resolve issues with their data precision." Grammier added that a strong data foundation is "critical for unlocking the best data to power your AI and agentic tools." A vendor-commissioned study concluding that the vendor's product category solves the identified problem warrants reading with that interest in view.

Vassilis Bakopoulos, SVP of Research and Insights at MMA, framed the two failure modes as blind spots. "Our research points to two blind spots working against them: missing data that can mask the channels that really perform, and identity matching that feels like progress but isn't built for precision," Bakopoulos stated. "Both erode confidence in measurement, and both can push budget decisions in the wrong direction."

Independent voices have raised parallel warnings about the infrastructure the report champions. The Federal Trade Commission cautioned in a November 2024 analysis that data clean rooms are not privacy-preserving by default, noting that despite the name, they "are not rooms, do not clean data, and have complicated implications for user privacy." The category the report presents as the fix carries its own scrutiny.

The timing also intersects with LiveRamp's corporate trajectory. Publicis Groupe agreed to acquire LiveRamp for $2.5 billion in an all-cash deal announced in May 2026, at $38.50 per share, a 30% premium to LiveRamp's prior closing price. The pending acquisition would fold one of the industry's larger neutral identity and clean room providers into a holding company, a shift with consequences for how independent the measurement layer remains.

Confidence in measurement itself has been slipping for some time. A TransUnion and EMARKETER study reported by PPC Land in October 2025 found 67.4% of marketers ranked proving incremental ROI to justify spend as their most pressing challenge, ahead of aligning metrics to business outcomes at 66.3% and improving cross-channel attribution accuracy at 55.1%. The MMA and LiveRamp findings sharpen that anxiety by naming a specific, testable mechanism through which the numbers on a dashboard can be confidently, invisibly wrong.

For an industry where a single misread signal can cancel a campaign that was quietly making money, the report's core message is narrow and pointed. The most dangerous measurement error is not the one that looks like an error. It is the one that looks like the truth.

Timeline

Summary

Who: LiveRamp, the San Francisco data collaboration company listed as NYSE: RAMP, in collaboration with the Marketing + Media Alliance, a nonprofit alliance of chief marketing officers operating in 16 countries with more than 825 corporate members.

What: A joint research report, The Missing Piece: Improving Confidence in Marketing Measurement, finding that non-random missing data and low identity precision can distort attribution and incrementality badly enough to make profitable campaigns appear unprofitable. In simulations, poor identity precision collapsed campaign ROI by approximately 70%, and a 50% precision scenario dropped measured return from $1.50 to $0.43.

When: Published on July 20, 2026, with findings presented the same day.

Where: Presented at the MMA CMO + CEO Summit in Santa Barbara, California, during a session on data loss and measurement.

Why: Marketers face data loss, privacy controls, identity fragmentation, and cross-channel measurement gaps while preparing for AI-driven transitions. The report argues that standard data-quality diagnostics fail to detect the non-random loss and identity errors that most distort budget decisions, and positions data collaboration through identity solutions and clean rooms as the remedy, a conclusion that aligns with LiveRamp's commercial offering.