Gannett, the parent company of USA Today, cut staff and reorganized its audience team this month. The internal memo explaining the move did not point at a soft ad market or a platform algorithm tweak. It named the mechanism directly: "search traffic is under pressure, and platforms are increasingly keeping user experiences to themselves instead of sending them back to us." Search Engine Land reported the memo on September 4, and Nieman Lab corroborated it the same week.
That sentence is worth sitting with for a moment. "AI Overviews might be reducing click-through" has been a hypothesis for over a year, argued from aggregate data that a marketer could cite and a business owner could reasonably discount as noise. This is not that. A national media company with a dedicated SEO desk and a multi-year runway just reorganized its staff around the claim, in writing, in a memo one of its own people leaked.
It is tempting to read this as a publisher-specific story: ad-supported newsrooms have always been exposed to referral traffic in a way most businesses are not. But the mechanism the memo names, platforms keeping the user experience to themselves instead of sending a click back, does not care what kind of site sits on the other end. An AI Overview or an AI Mode answer that resolves a query fully is exactly as complete whether the underlying source was a newsroom or a founder's services page. We asked directly whether AI Overviews actually reduce clicks a while back, working from the same category of aggregate evidence Gannett had before this memo. An org chart is a stronger answer than an aggregate study, because nobody restructures a team over a trend they can wait out.
What is actually different here
Two things separate this from the usual "AI is coming for search" news cycle. First, it is a named company, not an anonymous survey respondent. Second, it is a resourcing decision, not a strategy memo. Gannett has run large-scale SEO operations for over a decade and has more data on its own referral patterns than almost anyone. If that team is the one being reorganized around this pressure, the pressure is not speculative.
For a founder-led business, the honest read is that Gannett had the budget, the traffic history, and the technical staff to notice this early and still ended up reorganizing under it. Most small business and independent sites have none of that instrumentation. The traffic drop shows up later and gets attributed to something else, a slow season, a Google algorithm update, a competitor. By the time it is named correctly, a large publisher will already have adjusted its whole operating model around the same cause.
None of this says AI answer visibility is solved by any particular technique, and this piece is not claiming one. It says the size of the company that just restructured around the problem is the actual news. A hypothesis that a hobbyist blog worries about is one thing. A hypothesis a newsroom lays off staff over is a different category of signal entirely.
Sources: Search Engine Land, September 4, 2026; Nieman Lab.
