Skip to main content
My website puts the four sources about this brand’s own site on one page. Each part links to the page that holds the rest: Crawler Visits, AI Visits, Crawlability and Own Pages.

Its own window

The toolbar’s Window reads 7, 28 or 90 days. The dashboard filters do not apply here. Each source ends on its own newest day. So each figure names the days of its source: the crawler logs, GA4 or the answers. When all sources cover the same days, the line under the site’s name states them once.

A source that is not connected

The crawler logs and Google Analytics need a connection. See Connect your site data. Until then, a note at the top offers the connection to an admin. The cards of that source state the connection that is missing, over illustrative bars or lines. A source that holds no figure shows its state in words, never a zero. The states are “Not connected”, “No data yet”, “Not reading”, “Not available” and “Not loaded”. A read that fails leaves the other sources on the screen.

The tiles

The AI sessions are a floor. AI Visits says why.

AI traffic and referrals

The chart has two views:
  • AI traffic draws the crawler requests against the AI sessions, each on its own axis.
  • AI referrals draws the sessions per AI assistant.
Grain sets what one point covers: a day, a week or a month. A name under the chart hides or shows its line. The three dots export the figures as CSV, save the chart as an image, or copy the figures as text.

Crawl access

The card counts the AI crawlers that this brand’s robots.txt allows, allows in part and blocks. Below, it lists each crawler under its purpose, with its operator and the platforms it feeds. The foot states the day of the read and which group of robots.txt decides. When the site has a site problem, the card states the first one and links to Crawlability for the others. When robots.txt did not answer, the card says so and shows no verdict.

Why AI requests you, and Request health

  • Why AI requests you splits the crawler requests by the purpose of the crawler: training, a person’s question, search or other. AI crawlers lists each crawler’s purpose.
  • Request health splits them by the answer the site gave: OK, Redirected, Not found and Failed. A page not found is no failed request, so the two stay apart.

Pages

The table lists each page of the site that an answer cites or reads. It also lists a page that a crawler requests, or that an AI assistant sends a visitor to. Rows switches between pages and folders. A column whose source did not deliver for the days shows a dash and says why. Own Pages says how a page is counted, and lists the same pages under the dashboard filters.

Where it appears