Fragmented data, outdated models and long feedback loops make it more difficult to connect media spend to business results. Billions of dollars in investments are made based on incomplete information, often based on models that cannot take into account where consumers actually spend time. And as privacy changes and signal loss accelerates, the cracks are widening.
The report reveals a discrepancy between existing measuring instruments and where attention is actually directed. For example, 77% of marketers say gaming is underrepresented in their marketing mix models. Commercial media (50%) and the creative economy (48%) are also significantly overlooked. That kind of underrepresentation leads to underinvestment in the channels where consumers are most engaged.
Meanwhile, teams spend more time aggregating data in silos than generating insights. Measurement workflows are still largely manual and slow. The result: missed opportunities, misallocated budgets and marketing plans that don’t reflect real behavior.
The role of AI in fixing what’s broken
Amid all the dysfunction, marketers are hopeful that AI can bring meaningful change – not just to automate tasks, but also to rethink how measurement works.
According to the report, AI is expected to unlock $26.3 billion in media investment value by making measurement faster, more adaptive and strategic. The shift is already happening in three key areas:
- Speed and frequency: Marketers expect to move from annual or quarterly model updates to monthly, weekly or even real-time feedback loops. Incrementality testing, traditionally run a few times a year, is shifting to an always-on experimentation model.
- Strategy versus spreadsheets: As AI takes over routine data tasks like classification and cleaning, teams expect to be able to spend their time on higher-value work. The report estimates that this shift will yield $6.2 billion in productivity gains as marketers spend more time interpreting results and less time sifting through data.
- More access to advanced tools: AI helps democratize complex techniques such as multi-touch attribution and cross-channel lift analysis. These models have traditionally been reserved for advanced teams with the technical skills to manage them. With AI, more marketers can benefit from advanced insights without having to rebuild infrastructure from scratch.
About half of buy-side marketers are already scaling AI within their measurement programs. Many others are in early testing or proof-of-concept phases. It’s not surprising that analytics teams are the furthest ahead. They are more than twice as likely to adopt AI-based workflows as planning teams, largely because they are already working with machine learning models and large data sets.
Dig deeper: 3 mistakes in incrementality testing – and how to avoid them
That gap is closing. More than 70% of teams that have not yet scaled AI say they expect to do so by 2027.
What slows down adoption
While there is great excitement around AI, trust remains a major issue. Half of marketers expect legal, privacy or accuracy issues in the next two years. One of the biggest concerns is the ‘black box’ problem: when AI-driven insights cannot be explained or traced.
Risk tolerance also varies by role. Executives are focused on costs, ethics and impact on the workforce. Practitioners are more concerned about the implementation details – ownership, model management, and making AI work within existing workflows.
To manage these concerns, marketers are turning to contracts. About 37% of buy-side teams say they have already added AI-related language to partnership agreements, covering areas such as transparency, security and governance. That number is expected to double over the next two years, indicating that AI responsibility is rapidly shifting from theory to practice.
What marketers should do next
The IAB report outlines a clear action plan for marketers looking to modernize their measurement strategy without introducing unnecessary risks.
Urge for standardization and supervision
Shared industry standards – such as those being developed by IAB’s Project Eidos – can help ensure consistency and transparency between partners. Internally, marketers need to formalize human review processes, especially when AI is involved in budget or strategy recommendations.
Dig deeper: Consumers want less digital, more real world from brands by 2026
Modernize measurement methods
- For incrementality, replace one-time tests with a calendar-based approach and use AI to monitor when retesting is needed.
- For attribution, you must commit to rebuilding the model regularly and using AI to reconcile conflicting data signals.
- For MMM, validate input data before modeling and ensure the inclusion of channels that are often overlooked but increasingly important, such as CTV and retail media.
Break down the silos
Instead of treating attribution, incrementality, and MMM as separate models, marketers should use AI to reference results. Differences between models can reveal deeper issues and help teams converge on a more unified view of what really drives performance.
The shift is happening – with or without you
The measurement status quo is no longer sustainable. Marketers cannot afford to rely on systems that underrepresent key channels, delay insights, or lack transparency. AI offers a way out, but not as a layer on top of broken processes. To fully realize its value, marketers must rebuild their measurement frameworks with clarity, accountability, and adaptability at their core.
Energize yourself with free marketing insights.
MarTech is owned by Semrush. We remain committed to providing high-quality reporting on marketing topics. Unless otherwise stated, the content of this page was written by an employee or paid contractor of Semrush Inc.
#marketers #measurement #systems #inadequate #MarTech