Ahrefs adds AI crawler and rendering audit checks

Ahrefs announced more AI Discoverability checks in Site Audit on 19 November. The update flags AI-bot access, pages dependent on JavaScript rendering, and missing structured data. For a store team…

Will Slatton2 min read

Ahrefs announced more AI Discoverability checks in Site Audit on 19 November. The update flags AI-bot access, pages dependent on JavaScript rendering, and missing structured data. For a store team investigating weak visibility in AI search, these are useful technical questions to separate before changing content or buying more tools.

In its original announcement and demonstration, Ahrefs presents the checks as ways to find access and readability problems. The company’s demo includes an AI Discoverability score, but a high score should not be read as a promise that an assistant will cite a store or send it customers.

Start with representative pages

Our suggested first pass is a homepage, a collection page and several product pages, including one with substantial app-generated content. In Site Audit, review the AI Discoverability findings and record the exact URL, crawler and issue behind each warning. A site-wide score alone gives a developer little to fix.

For an access warning, inspect the applicable crawler rule and the actual response returned by the site’s delivery or security layer. Keep decisions about search access separate from decisions about model training. Do not remove every bot restriction simply to eliminate a warning.

Check what exists before scripts run

For rendering warnings, compare the initial page content with the finished storefront. Our editorial recommendation is to check whether essential product facts and policies remain available without relying entirely on a widget to insert them. The warning is a reason to investigate the affected content, not proof that every AI service fails to understand that page.

For missing structured data, verify what is already present and whether it matches the visible product information before adding more markup. Duplicate or contradictory output from a theme and an app is not a useful response to an audit warning.

Measure outcomes separately

After making a change, recrawl the affected URLs and keep a record of the original finding and the fix. Track actual citations, referral visits and commercial outcomes separately where those can be measured. Technical accessibility is one condition for discovery; this release does not establish a conversion lift or guarantee inclusion in an AI answer.

ℹ️
An audit score is a diagnostic aid. It does not guarantee citations, rankings or sales, and search-crawler access should not be confused with permission to train a model.
Ahrefs’ original 19 November 2026 demonstration. Sample scores are not merchant results. Watch the original announcement.
Filed underAI

More from AI

See all →