Branch survey: marketers estimate losing 27% of ad spend to fraud, ~$3M a year
Roughly $3 million a year is the fraud loss the typical marketer now expects to absorb, according to a Branch survey of 455 marketing executives across North America, EMEA and APAC, released Monday. Respondents estimated they lose 27% of ad spend to fraud annually, even as artificial intelligence lowers the cost of faking clicks, installs and conversions. The report, "Ad Fraud In 2026: How Marketing Executives Are Responding," found that 87% of leaders have grown more concerned about fraud in the past 12 months and 98% have altered how they allocate or manage media budgets in response. Paid social topped the list of fraud concerns in every region, while connected TV ranked near the bottom despite showing higher invalid traffic rates than desktop or mobile web. Branch's Landis said new fraud methods, including agentic AI schemes and manipulated proxy metrics, are emerging faster than the industry can track them. "The closer you can measure the actual performance of your advertising, the more you avoid the aspects of fraud loopholes," he said.
Campaigns booked on trust in platform-reported metrics face a widening gap between what dashboards show and what fraud actually consumes, especially as CTV keeps drawing lower scrutiny despite carrying more invalid traffic than desktop or mobile. Budgets that once assumed click and conversion counts were roughly accurate now need verification layered on top, particularly as agentic AI and manipulated proxy metrics create fraud vectors the industry hasn't fully mapped. Measuring outcomes tied to actual revenue offers the more reliable defense as detection tools struggle to keep pace, especially since only 16% of surveyed marketers say AI currently helps them catch fraud.