A growing measurement gap in digital advertising is raising new concerns across the marketing industry, as fresh research reveals that 80% of brand and agency marketers still optimize campaigns without verified purchase data. The findings highlight mounting pressure on advertisers, media platforms, and martech vendors to improve attribution accuracy as AI-driven advertising and privacy restrictions reshape how campaigns are measured.

New research from the Affinity Solutions Outcomes Marketing Council suggests the digital advertising industry is facing a broader credibility problem around campaign measurement and optimization. According to the study, four out of five marketers primarily optimize campaigns using signals other than verified purchase data, while 35% say those optimization decisions fail to hold up once reconciled against actual sales outcomes.

The report reflects growing tension across the advertising ecosystem as marketers attempt to balance speed, scale, attribution accuracy, privacy compliance, and AI-driven automation in increasingly fragmented media environments.

The research surveyed 210 senior brand and agency marketing leaders and found widespread skepticism toward current advertising measurement systems. Nearly 91% of marketers believe platform-reported results are overstated to some degree, according to related findings highlighted by Affinity Solutions.

The issue is becoming strategically important for enterprise marketers as AI-powered campaign optimization platforms gain influence across advertising and commerce ecosystems.

Modern digital advertising infrastructure increasingly relies on machine learning models built by platforms such as Google, Meta, Amazon, and Microsoft to automate bidding, audience targeting, creative optimization, and attribution modeling.

But marketers are questioning whether the underlying optimization signals accurately reflect real business outcomes.

According to the Affinity Solutions study, many organizations still rely heavily on proxy metrics such as clicks, attributed conversions, modeled outputs, engagement signals, and platform-specific reporting instead of verified transaction-level purchase data.

That disconnect creates operational inefficiencies across the advertising supply chain.

Industry analysts have increasingly warned that privacy changes, signal loss, cookie deprecation, cross-platform fragmentation, and walled garden ecosystems are weakening traditional attribution models.

Research from Bain & Company previously found that AI-generated search experiences and zero-click discovery environments are also reducing direct traffic visibility, complicating performance measurement even further.

The Affinity Solutions findings suggest many marketers are struggling to adapt.

The report identified several systemic barriers preventing broader adoption of purchase-based optimization strategies, including limited access to transaction data, data latency, privacy constraints, internal operational complexity, and budget limitations.

Long data-processing chains are also contributing to slower decision-making.

Nearly two-thirds of respondents reported three or more operational steps between a customer transaction and an optimization decision, increasing delays and reducing signal reliability for in-flight campaign adjustments.

The issue is becoming more urgent as generative AI transforms advertising workflows.

AI-powered media buying systems increasingly depend on large volumes of behavioral and transactional data to automate optimization decisions in real time. If underlying signals are inaccurate, delayed, or disconnected from actual purchases, the risk of systemic inefficiency increases significantly.

That has implications for both advertisers and technology vendors.

Enterprise marketing teams are under growing pressure from CFOs and executive leadership to prove measurable business outcomes tied to media investments, particularly as advertising costs rise across retail media, connected TV, programmatic advertising, and social commerce environments.

According to Marketing Week research, only about 39% of marketers currently measure whether campaigns are delivering broader business outcomes beyond engagement or conversion metrics.

The growing emphasis on outcomes-based advertising is helping drive interest in alternative measurement frameworks including incrementality testing, media mix modeling, retail media attribution, and verified transaction-based analytics.

Several martech and adtech vendors are positioning themselves around “outcomes measurement” infrastructure designed to connect advertising exposure directly to commerce activity.

The broader market shift also reflects changing advertiser expectations around transparency.

Advertisers increasingly want independent validation, clearer attribution methodologies, and real-time access to commerce-linked performance data rather than relying solely on platform-reported metrics.

Industry conversations on Reddit and professional marketing forums reflect similar concerns, with many marketers describing growing distrust around attribution systems and optimization models built on incomplete or modeled signals.

At the same time, AI is introducing both new complexity and new opportunities.

Advanced AI systems can improve predictive targeting, automate campaign optimization, and process larger datasets faster than traditional analytics environments. However, those systems still depend on reliable input data to generate accurate recommendations.

That dynamic is reshaping the broader advertising technology landscape.

Platforms focused on first-party data infrastructure, commerce media networks, AI-powered attribution, and identity resolution are increasingly becoming strategic priorities for enterprise advertisers navigating post-cookie digital ecosystems.

The Affinity Solutions study ultimately points to a larger transformation underway across digital advertising: marketers are moving away from measuring media performance based solely on engagement metrics and toward systems designed to connect media investment directly to verified business outcomes.