Report: 73% of B2B firms track AI referral traffic, only 34% trust the data

August 16, 2026 · mediapost.com

Seventy-three percent of B2B organizations say they track AI referral traffic, while only 34% say they highly trust the metrics behind it, according to a report cited by Kelsey White, vice president of client services at GNW Consulting, in an August 14 MediaPost column. White writes that reporting problems usually build up gradually, as platforms change, tracking definitions shift and reports get rebuilt to answer one-off questions, and that by the time inconsistencies surface, analysts have typically already found workarounds. She extends the argument to revenue attribution, where models built to explain which marketing and sales activities influenced a deal rest on assumptions that are not always obvious to the people reading the reports. "It's making longstanding ones much harder to overlook," she writes of AI's effect on those weaknesses.

Attribution models that assign credit across channels carry real risk when their underlying logic hasn't been challenged in years, and AI can wrap that risk in confident, polished language, exactly the dangerous combination White describes. Her point about revenue attribution matters directly for anyone weighing which channels get credit for a deal: an AI-generated summary is only as reliable as the definitions and data ownership behind it. Agencies and advertisers with disciplined measurement and clear ownership of their data are positioned to gain real efficiency from automation, and White notes that the organizations already getting the most value from AI often built that discipline well before this wave of tools arrived. Teams still leaning on tribal knowledge to reconcile numbers each week may instead find that AI simply speeds up the confusion.