Semrush published a study on September 8, 2026 measuring how much traffic AI tools actually send to manufacturing and industrial websites, and the answer is almost none. Across ten industrial categories, AI assistants and Google’s AI Mode combined accounted for just 0.48% of sessions between January and July 2026, even as AI Overviews expanded from covering 38% of the sector’s tracked search volume to 57% over the same period. Direct traffic and organic search together still account for nearly 80% of visits.
The gap matters because it separates two things marketers have been treating as one metric. Being mentioned in an AI answer is not the same as being clicked. The study’s second dataset, drawn from Semrush’s AI Visibility tracking across ChatGPT, Gemini, Google AI Mode and Google AI Overviews, found that the manufacturing brands named most often inside AI-generated answers were frequently not the brands those answers linked to as sources. A company can dominate the conversation an AI system has with a buyer and still receive none of the resulting traffic, because the citation and the mention are tracked, and rewarded, separately.
The original insight here is a budget one, not a technical one. Attribution models built for the click-through era already struggle with zero-click search; this data suggests industrial and B2B marketers are optimizing spend against a channel, AI-driven discovery, that is reshaping how buyers form opinions of a brand well before it shows up in any traffic report tied to a dollar figure. Measuring AI visibility separately from AI traffic, rather than folding both into one dashboard tile, is becoming the only way to see whether a brand is losing the conversation, the click, or both.
Source: Semrush
