SEO researcher Lily Ray has documented a pattern of severe visibility drops for SaaS and B2B websites that place their own product at the top of “best of” lists. Now it has been confirmed that the damage is also spreading to AI response engines.
-
Affected websites saw their organic visibility drop by 30 to 50% within a few weeks, with a focus on blog subfolders and comparison pages, not homepages or product pages
-
The pattern suggests listicles where the publisher ranks its own product as a top choice, often with superficial “2026” title updates and little real testing behind the rankings
-
Ray warns that the impact is not limited to Google. Loss of visibility can extend to ChatGPT, Perplexity, and other AI systems that rely on the Google index for retrieval
What caused the drops?
Google hasn’t confirmed a new update by name, but the timing is consistent with the volatility that followed the December 2025 core update.
Ray’s research found that the most affected websites had a specific content pattern: mass-produced “Best In several cases, the AI-generated content was in a single subfolder, but the loss of visibility affected the entire domain.
Ray also asked a pointed question in her analysis: How many of these companies had actually used the competitors’ products that they had compared to their own? In most cases the answer was zero. Listicles created without real testing or direct use lose ground.
Why it matters beyond Google
Since AI Overviews, AI Mode, and Gemini come directly from the Google index, a decline in ranking there may mean fewer citations in AI-generated answers. ChatGPT, which relies heavily on Bing in addition to some Google signals, and Perplexity, which pulls multiple indexes in real time, are both subject to the same downstream effect. A brand that loses Google visibility risks losing AI visibility at the same time, rather than as a separate issue to be solved later.
What still works
Not every listicle is at risk. Objective third-party lists, i.e. those without an obvious self-promotional aspect, have remained stable over the same period. Analysts tracking the impact say the deciding factor is evidence: Real testing, honest negative results and first-hand experience trump pre-packaged rankings that crown the publisher’s own product every time.
Quick links:
