Marketers ditch attribution for incrementality

By late 2026, the constraint on marketing spend was no longer a lack of channels but a lack of credible measurement. MarTech, citing Bitly’s Marketing Visibility Report, said the average marketing team now uses six different tools to measure performance, yet only 18% of marketers say they have a clear view of what’s actually working; the biggest visibility gaps sit in valuable digital environments, with organic social ranked first even though it is used by nearly three-quarters of marketers and remains difficult to connect to meaningful business outcomes.

That visibility shortfall became a budget problem under rising ROI scrutiny. NIQ reported that 84% of CMOs now answer to ROI as their primary metric and 74% face tighter scrutiny to prove it, while nearly 29% of US brands and agency marketers said 11% to 20% of their budget is currently at risk; at the same time, IAB found 61% say their MMM does not capture all channels, and Harvard Business Review said only 22% qualify as MMM leaders, with 42% unable to act on reports at all.

Attribution no longer settles the question because the signals it depends on are breaking down as automation speeds up decisions. Smarter Marketer captured both failures: 59% of online events are now missing due to privacy, ad blockers and tracking changes, and marketers now face 15 different measures of the same thing, making cross-platform truth harder to assemble; Paul Thinkinson warned that marketing does not drive everything and that claims of a 3,000% lift show how attribution can drift from causal measurement into overclaiming.

That matters more under ROI pressure and AI-led optimization, because credit-allocation metrics can steer spend before anyone verifies whether they caused growth. DesignRush cited Deloitte showing 58.8% of CMOs reporting greater pressure to prove marketing ROI to their CEOs, while Bdaily showed why that fails in Performance Max tests where switching off campaigns cut sales by only 30% of what Google attributed, meaning the remaining 70% would have happened anyway; that is why marketers are turning to incrementality testing, independent validation, and Marketing Efficiency Ratio, as Silverback Strategies did for CroppMetcalfe after the home provider was allocating 10% of its entire budget towards branded search but it was not clear if branded search was actually driving incremental results.

Incrementality is only reliable when it sits inside an external measurement frame, because platform-run tests still reflect platform incentives and partial visibility. On The WARC Podcast, the warning was explicit: independent validation is 100% key, especially when tools such as Meta’s incrementality products remain biased towards their own platform. Goafest 2026 sharpened the same point with Dhiraj Gupta’s line that the maker cannot be the checker and the checker cannot be the maker, followed by a call for independent, third-party verification to ensure advertiser confidence.

Even a sound lift test cannot answer the business question unless it is anchored to portfolio economics, which is why MER and break-even thresholds complete the stack. MarTech showed how isolated platform metrics can mislead, saying Meta reports a ROAS of 4.0 while measured iROAS is far lower, and tied testing to decisions with a 50% pre-advertising contribution margin and roughly a 2.0 incremental ROAS to break even, including rules such as Between 1.5 and 2.0, we’ll reduce spend 25%. Sleeping Barber warned marketers routinely lets a platform's attribution window become the de facto useful life of a brand impression, though nobody decided 28 days was the useful life of a brand impression, underscoring why Lifesight’s standard is to Measure true marketing impact with incrementality, MMM, and causal analytics in a privacy-first world.

Budget behavior is already moving from assumed efficiency to proven lift. Mexico Business News described a “large Mexican retailer” that “switched off its display campaigns across several states… for nearly two months” to learn “how much of their sales actually depend on this advertising,” and the independent test showed “total e-commerce sales in the test regions fell 19.23% below the expected scenario,” even though “ROAS had always said the campaign was efficient” and “had never said what portion of the business would evaporate without it,” a result that directly changes what a marketer would keep funding.

That shift is showing up in both intent and action, including among smaller advertisers: after a “Propel webinar on attribution in the warehouse,” the author reviewed registrations and found, “Half the people who joined were spending under $1M a year on ads.” A 2025 EMARKETER and TransUnion survey found “over a third of marketers plan to increase their investment in incrementality testing over the next 12 months,” while Exchange4Media reported that Truelift’s work led Sweet Karam Coffee to “change its advertising strategy, moving half of its quick-commerce ads to Meta,” after a six-week incrementality plan, and “the brand subsequently saw a 20% increase in sales in the following months,” with case-study material saying it “reallocated nearly 50% of its marketing budget to higher-performing channels.”