Skip to content

WWW Audits

Copy for agent Copied

WWW Audits helps you review individual website pages for technical and content signals that can make them easier for AI systems to understand. Use it to identify pages that may need clearer structure or more complete metadata, then share the findings with the person who maintains the site.

Open WWW Audits and choose New audit. Enter one valid URL per line. You can save the URL list to return to it later, or choose Save & Run to start processing it. The list is saved in this browser for your signed-in account, so it can be prefilled the next time you open the form there.

The usage indicator shows how many URLs remain in your allowance for the current billing period. The allowance renews when a new billing period starts. Every URL you submit counts, including URLs whose check later fails and URLs entered twice, so remove duplicates before running. If the list is over the remaining amount, remove URLs before running it. An empty list cannot be run. If any line is not a valid URL, the audit does not start until you correct it.

When a project is selected, the allowance and history cover that project’s audits, and everyone with access to the project sees them. Access depends on your plan and workspace or project access; trial plans and plans without audit allowance cannot open the audit workflow.

When an audit is running, the page shows a processing state. It takes about four seconds per URL. The audit continues if you close the page; it appears as Pending in the history until it finishes, and its URLs show as in progress. A successful run opens its results automatically. If the run fails, the page shows an error.

WWW Audits opens on your latest audit. A note at the top shows how many URLs you have used from this month’s allowance.

Four figures summarize the audit you are looking at:

  • Last audit — its date, number of URLs, duration and status.
  • Checks passed — passed checks out of the checks that apply to the audited pages, with the percentage.
  • Most common gap — the check that failed on the most pages, with how many pages passed it.
  • Avg. reading age — the average estimated reading age across the pages, next to the average recommended range.

Below them, the results table shows one row per URL and one column per check: ✓ passed, ✗ failed, — does not apply to the page or has no data. Reading shows the estimated reading age and Style D/D the share of declarative and descriptive sentences. Score counts the passed checks out of the eight checks that apply to that page; the reading age counts as passed when it falls inside the recommended range, while writing style is shown for information only. A check the page does not need, such as FAQ markup on a product listing without questions, is not counted, so it neither raises nor lowers the score.

Select a row to see what to fix on that page. To fix lists each failed check with what is wrong and how to fix it, for example a missing author or headings that skip levels. Below it, All good names the passed checks and Not applicable the checks that do not apply, followed by the writing style. Only one row is expanded at a time. A URL whose check failed shows error; expanded, it tells you the page could not be fetched and, when the audit recorded one, the reason.

The Audit history table lists every audit with its date, status, number of URLs, duration and passed checks. Select a row to show that audit’s results above; the selected row stays highlighted. Use Show more to load older audits when available.

You can export the audit you are viewing as CSV or XLSX with the export button next to its results. Exports are disabled while the audit is pending. Audits cannot be deleted. To check a page again after a fix, run a new audit; it uses the allowance again. Use an export to share the recorded results with a colleague who can make changes to the website.

The results table shows these checks for each URL:

  • How-To Schema, Organisation, Breadcrumb, and FAQ Schema report structured data found in JSON-LD, the most common format for it. Structured data added in another format can be reported as missing. FAQ Schema passes only when the page carries FAQPage markup, in JSON-LD or microdata; a visible FAQ section without that markup fails, because AI systems cannot read the questions as structured data. To inspect a page’s structured data in more detail, use Schema, which also keeps each result so you can compare a page before and after a fix; to create markup, use Schema Generator.
  • Content Structure reports whether the page has an H1 and a heading structure.
  • Crawlability reports the meta robots and canonical tags.
  • Authority Signals covers author information and outbound links, including reputable sources.
  • Reading Level estimates the reading age; it counts as passed when it is inside the recommended range. Writing Style reports declarative and descriptive sentences for information only.

A passed check means the audit detected the expected signal. A failed check is a prompt to inspect that URL and decide whether the page needs attention; it does not by itself prove that AI systems will ignore the page or that adding a signal will guarantee citations. The advice in an expanded row is built from the signals the audit found on that page. An audit only reads the pages; it does not change your Visibility results or monitored prompts.

Audit-level status can be Completed, Partial, Pending, or Error. A partial result means some URL checks succeeded and some failed. Review the individual URL results before deciding what to fix. A check that does not apply to a page, or returned no data, shows — and is left out of the score.

The form takes URLs, one per line. Add the specific pages you want reviewed, subject to the number of URLs remaining in your allowance.

Does a passed audit guarantee that my page will appear in AI answers?

Section titled “Does a passed audit guarantee that my page will appear in AI answers?”

No. The audit reports technical and content signals on the submitted pages. It does not measure whether a model used a page in an answer or guarantee a citation.

Save keeps the current URL list in this browser for your account. It does not start an audit. Choose Save & Run to submit the URLs for analysis.

Some submitted URLs may not have a complete result. Select the audit in the history and expand each URL marked error to see why the audit could not fetch it. Checks that do not apply or returned no data show —. Common causes are pages that need a login, block automated visitors, redirect elsewhere, or respond too slowly. Open the URL in a private browser window to check that it is publicly available.

Audit access requires the feature to be enabled and an eligible plan with audit allowance; trial plans are not eligible. Project-scoped access requires write permission for the selected project. Ask your workspace owner about plan access if the menu redirects you to plans.

  • Which submitted pages are missing structured data or clear heading structure?
  • Do the pages I care about declare crawlability and canonical information?
  • Which findings should I send to the person who maintains the site?
  • Did every URL finish successfully, or should I inspect partial results?