Shoppers and business buyers now ask AI assistants for recommendations, and each answer names only a handful of companies. Most marketing teams assume their brand strength will carry over. Our audits suggest it often does not, and the reason is frequently outside marketing's control.
Discovery is moving into answers that name only a few companies.
A search results page offers ten links and room to scroll. An AI answer offers a short list and a reason. Being absent from that list means being absent from the consideration set, and conventional analytics cannot show a visit that never happened.
Brand fame does not transfer into AI answers.
We audited ten brands in a global apparel and footwear portfolio against our seventy-point AI readability checklist and tested branded and category questions across the major answer engines. The portfolio averaged 26.4 out of 70. No brand scored above 39. A leading outdoor brand won questions that used its name but lost the broader category questions, such as which winter jacket to buy, to review sites.
The biggest reason the brands were missing from AI answers was a security rule, not the content.
The most common blocker is invisible to marketing.
The largest single problem was not content. Infrastructure teams had configured CDN and firewall rules that returned access-denied responses to AI crawlers, and the commercial teams did not know. The rules were written to stop unwanted bots and blocked the systems that now shape buying decisions.
AI visibility depends on access, clarity, consistency and corroboration.
Answer engines must be able to reach your pages, understand them quickly, find the same facts wherever they look and confirm your claims through independent sources. Weakness in any one of these can keep a strong brand out of the answer.
Our approach measures the answers, fixes the evidence and retests.
We build a set of real buyer questions, including the category questions that never mention your name, and record which companies each engine recommends and which sources it trusts. We score priority pages, then split the fixes between the teams that own them: infrastructure for access, content teams for clarity, product data owners for consistency. The same questions are retested on a schedule to show whether visibility is improving. We never promise placement, because no one can.
AI visibility is now a shared responsibility.
Marketing cannot fix this alone. The security rule, the product feed and the website structure belong to different teams. Companies that treat AI visibility as a cross-functional program will appear in the answers that shape demand. The rest will not know what they are missing.