AI NewsGo-to-marketAnnouncement
Semrush names four layers an AI search strategy has to fix in order
Semrush splits the question of why a brand misses AI answers into four layers with different fixes.
Image: Semrush
Why it mattersA team that treats every AI answer gap as a content problem rewrites pages that AI never had trouble with.
A brand name does not appear in a ChatGPT answer. The team's first move is to rewrite a page. The page was never the problem. Semrush published a guide on 23 October. It asks the team to work out which layer of AI search broke first, before picking a fix.
The piece is bylined Alex Lindley, with Faizan Ali as a contributor. Semrush calls the shape an AI search strategy, and some teams call it AI SEO or agentic search optimization. The guide lists four layers that fail in different ways. Each layer has a different owner, and each needs a different kind of fix.
The four layers
Brand discoverability. Can AI systems reach the content at all. The guide names robots.txt, CDN, firewall, WAF, and bot-management rules as common blocks. It also notes a Google rule. A page must be indexed in Google Search and eligible for a snippet. Only then can it show in AI Overviews or AI Mode.
Brand clarity. Does AI describe the brand the way the brand wants. Semrush's Geertrui Laleman, who runs AI search optimization there, is quoted in the piece. She says AI answers used to describe Semrush as a traditional SEO tool with AI features added on. The team then fixed the sources the models were reading.
Brand authority. Does the brand show up when AI is picking a shortlist. The listed fixes are backlinks, topical depth, expert associations, and third-party mentions.
Brand trust. Will AI recommend the brand with no caveat. The listed fixes are sentiment, review quality, cross-source corroboration, transparent claims, and reputation consistency.
The numbers Semrush measured
Semrush ran a study of 3,981 brand appearances across ChatGPT, Google AI Overviews, Gemini, and Google AI Mode. In 62 percent of those appearances, the AI cited a source but never named the brand. ChatGPT alone cited brands in 87 percent of its appearances and named them in only 20.7 percent.
A separate Semrush study compared pages that AI cited with pages that rank in Google's top 20 for related keywords. The cited pages scored 32.83 percent higher on clear summarization. They scored 30.64 percent higher on E-E-A-T signals, and 25.45 percent higher on Q&A formatting. The guide notes that structured data here means lists, tables, and Q&A pairs in visible text, not schema markup.
What this changes
A team can now say which part of its AI search problem is broken. Rewriting a page helps clarity. Fixing a robots.txt or firewall rule helps discoverability. A rewrite will not fix either of the other two. The piece asks a reader to work out which layer broke first. It also asks them to measure citations, mentions, accurate descriptions, and recommendations as four separate numbers.
Source
This item was written by an AI system from the linked source. Reveneau is responsible for what it publishes.
Get AI News in your inbox
New developer tools, model and agent releases, and how teams are actually using them to release software. Short, and only when there is something worth reading.
