The Fraud Files: When Good Metrics Tell Bad Stories
August 12, 2026 · By Paula Chiocchi · 4 min read
By Paula Chiocchi · August 12, 2026
In the first installment of The Fraud Files, we explored the staggering scale of digital advertising fraud and why billions of marketing dollars may never reach real human audiences. But that raises an even bigger question. If fraud is this widespread, why haven’t more marketers noticed? The answer may lie in the very dashboards marketers rely on every day. In my second conversation with Dr. Augustine Fou, we shifted our focus from fraud itself to something even more fundamental: measurement. Because if the metrics guiding your decisions are flawed, every optimization, budget allocation, and performance review built on those metrics becomes questionable. It’s an uncomfortable conversation—but an essential one.
When Measurement Becomes Misleading
Marketing has never had access to more data. Every campaign generates click-through rates, viewability scores, engagement metrics, attribution reports, and conversion dashboards. These numbers create the impression that digital advertising is incredibly measurable. The problem, as Dr. Fou explained, is that measurement isn’t always the same as accuracy. Many of the industry’s most trusted metrics were developed years ago to solve legitimate problems. Over time, however, they evolved into optimization targets—and once a metric becomes the target, bad actors quickly learn how to manipulate it. That’s why campaigns can appear highly successful on paper while delivering disappointing business results.
The Attribution Problem Few People Question
Perhaps the biggest example is attribution. Most marketers are familiar with attribution models that assign credit to various marketing channels throughout the customer journey. In theory, this helps organizations understand what’s driving conversions. But as Dr. Fou pointed out, many attribution models don’t necessarily measure causation—they measure correlation. If several platforms each claim credit for the same conversion, it’s entirely possible to end up with attribution reports that significantly exceed the actual number of sales your business generated. That should prompt every marketing leader to ask an important question: Are we measuring what actually caused the sale—or simply what happened nearby?
Optimizing Toward the Wrong Goal
One of the themes that surfaced repeatedly throughout our discussion is that algorithms simply optimize for whatever they’re told to optimize. If success is defined by clicks, algorithms will chase clicks. If success is defined by viewability, they’ll maximize viewability. Unfortunately, fraudsters understand those objectives just as well as marketers do. Bots generate clicks. Fraudulent sites create artificially high engagement. Fake inventory can produce exceptional-looking performance metrics—all while delivering little or no genuine business value. The result is a dangerous cycle: campaigns appear to improve, algorithms allocate more budget to those “high-performing” environments, and marketers unknowingly invest more heavily in fraudulent inventory. The dashboard gets better – the business doesn’t.
Looking Beyond the Dashboard
One point I found especially compelling was Dr. Fou’s reminder that interim metrics should remain just that—interim. Clicks, engagement rates, viewability scores, and other campaign indicators can certainly provide useful direction. But they should never replace business outcomes. Ultimately, marketers should be asking questions like: Did sales increase? Did qualified pipeline grow? Did customer acquisition improve? Did we reach new buyers? Was revenue actually impacted? Those are the metrics that determine whether marketing is creating value. Everything else is simply a clue—not the final answer.
The Incentives Matter
We also discussed something that’s rarely talked about publicly: incentives. Many agencies are compensated based on media spend. Platforms benefit when budgets increase. Vendors often optimize toward delivery metrics that make campaigns appear successful. None of that necessarily means anyone is acting in bad faith. But it does mean that relatively few participants in the ecosystem are rewarded for questioning whether those metrics truly represent business performance. That’s why marketing leaders themselves have to ask the difficult questions. No one has more at stake than the organization funding the campaign.
Healthy Skepticism Creates Better Marketing
If the first installment of The Fraud Files challenged us to rethink where our advertising dollars go, this episode challenges us to rethink how we evaluate success. Good marketers shouldn’t become cynical about data. They should become more curious. Dashboards are valuable. Attribution models are useful. Performance metrics absolutely have a place. But none of them should be accepted without asking whether they reflect real business outcomes. The best marketers don’t simply measure more – they measure what matters.
Next in The Fraud Files... If measurement tools can be misleading, what about the technologies designed to protect your brand? Next, we’ll investigate brand safety, verification vendors, and why some of the industry’s most trusted safeguards may actually be sending advertising dollars to the very places they’re supposed to avoid. Listen to the full conversation with Dr. Augustine Fou on the Marketing Influence Podcast for an even deeper look at the hidden forces shaping today’s digital advertising ecosystem.