73% of Marketers Fail ROI: ANA Masters 2026

Let’s be blunt: despite all the money we pour into marketing, a whopping 73% of enterprise marketers can’t accurately measure campaign ROI. That’s not my opinion, it’s from a 2025 Interactive Advertising Bureau (IAB) report. This exact problem was all anyone could talk about at the recent ANA Masters conference, and it highlights a huge gap between what we spend and the results we can actually prove. So how do enterprise teams finally fix this measurement mess and show their value?

The Data Disconnect: 73% Struggle with ROI Measurement

The IAB’s 2025 “State of Measurement” report laid it all out: nearly three-quarters of us can’t confidently pin ROI to our work. This isn’t some academic exercise. It screws up strategic planning, it gets budgets cut, and it shapes how the rest of the company sees the marketing department. I see it constantly with the large teams I work with, they’re running on a patchwork of disconnected tools and siloed data, making a complete picture of campaign performance impossible. I had one major retail client whose e-commerce platform data and in-store POS systems were totally separate, creating a massive blind spot on any omnichannel campaign. Without one source of truth, figuring out if a sale came from a digital ad or an in-store sign is just a guessing game.

The problem is almost always a fragmented data infrastructure. Big companies inherit a tangled mess of legacy systems, new platforms, and niche tools that all spit out their own data. Trying to connect it all requires a serious investment in data engineering and integration layers. It also means you have to change your thinking. Marketing measurement can’t be something you staple on at the end. You have to design it into the campaign from day one. That means defining success, identifying the data points you need to track it, and making sure the tech is in place before you launch. Skip that foundational work, and any attempt at sophisticated attribution or predictive analytics is a waste of time.

Attribution Model Adoption: Only 28% Use Advanced Methods

A late 2025 study from eMarketer showed that only 28% of large companies have gotten past basic last-click or first-click attribution. Given how complicated customer journeys are today, that statistic is alarming. A customer might see a social ad, click a search result, open an email, and then finally buy something days later. Last-click just gives all the credit to that final touchpoint, completely ignoring everything that came before. This is how budgets get misallocated, with awareness-building channels getting starved of funds because they don’t drive the final click. I’ve watched marketing directors kill their display ad spend for this exact reason, only to see their overall conversions tank because the top-of-funnel pipeline dried up.

People think implementing multi-touch attribution is this huge project that requires ridiculously expensive software and a platoon of data scientists. I disagree. While the fancy algorithmic models are great, even a simple linear or time-decay model gives you way more insight than last-click ever will. The real barrier isn’t the tech. It’s organizational inertia and a deep-seated fear of what the data might reveal about channel performance. Moving to a better model means you have to be ready to challenge old assumptions and maybe shift huge chunks of the marketing budget around. That often puts you face-to-face with internal stakeholders who are very attached to their channel’s metrics. The trick is to start small. Pilot a new attribution model in one region or on one product line, prove its value, and then build from there.

The Talent Gap: 65% Report Shortages in Measurement Expertise

A 2024 Nielsen report on marketing effectiveness found that 65% of enterprise marketing leaders admit they have a major talent gap when it comes to data analytics and measurement. This is not a shock. The demand for people who can actually interpret complex data and turn it into a real marketing strategy has exploded, and the supply hasn’t kept up. You have a lot of traditional marketers who don’t have the stats background, and you have data scientists who don’t have the marketing context. This creates a bottleneck that stops companies from getting real value out of their data.

Hiring more data scientists helps, but the real solution is building data literacy across the whole department. That means training programs that teach marketers how to read a dashboard, question the data, and ask for the right reports. And as mentioned in the takeaways, setting up a dedicated marketing operations team is a big deal. This group can centralize the expertise, acting as the bridge between the heavy-duty data infrastructure and the strategic marketing work. These teams manage the data pipelines, ensure data is clean, and build the reports, which frees up campaign managers to actually be creative and strategic. Without that specialized function, measurement just becomes an annoying, ad-hoc chore instead of a strategic weapon.

Budget Allocation: 40% Still Rely on Gut Feeling

An early 2026 Statista survey found that about 40% of enterprise marketing budgets are set based on “gut feeling” or just doing what was done last year, not on actual data. That figure, while maybe better than it was a decade ago, is still shockingly high for companies of this size. Flying by intuition might work out occasionally, but it’s not scalable and it’s incredibly risky. This is how inefficient spending gets baked into the budget year after year, and it makes it impossible to react quickly when the market or customer behavior changes.

The problem is usually a lack of confidence in the data. If marketers don’t trust their measurement systems, they’re going to fall back on what feels safe or what they think has worked before. This just proves why strong data governance and transparent reporting are so important. When I’m advising enterprise clients, a big part of what we do is build a single, accessible dashboard that shows the key performance indicators (KPIs) clearly and consistently. It needs to be updated constantly (ideally in real-time) and become the single source of truth for every budget conversation. On top of that, you need a rigorous test-and-learn process. Dedicate small parts of the budget to experiments with very clear goals and measurement. This is how you slowly build trust in data-driven decisions and prove the value of moving away from just spending based on a hunch. For CMOs, getting a handle on Google Ads data accuracy is a non-negotiable part of this.

The Future of Measurement: A Focus on Business Outcomes

The big takeaway from the ANA Masters conference is that the future of enterprise measurement isn’t about clicks and impressions anymore. It’s about showing a direct line to core business goals. We have to shift our KPIs away from marketing-only metrics (like traffic or engagement rates) and toward things that the CFO and CEO actually care about: customer lifetime value, sales qualified leads, pipeline velocity. So instead of just reporting that your email open rate was 25%, you need to be able to show how those opens led to website visits, which led to form fills, and which in the end became closed-won deals.

Getting to that level of integration isn’t easy. It means you need a deep understanding of data warehouses, your CRM, and even the company’s enterprise resource planning (ERP) platform. It also forces marketing, sales, and finance to speak the same language. Too often, marketing talks about “conversions,” sales talks about “closed deals,” and finance talks about “revenue.” Getting everyone to agree on definitions and building cross-functional dashboards is a huge project, but it’s the only way marketing earns a permanent seat at the strategy table. Without it, you’ll always look like a cost center instead of a growth engine. And digging into the problems with AI attribution and its blind spots makes these challenges even clearer.

Look, getting to sophisticated measurement in a big company isn’t a quick project. It takes real investment in tech, talent, and process. But even more than that, it takes a cultural change where data isn’t just a nice-to-have, but the absolute foundation for every decision. The companies that really commit to this are the ones that will win.