Search Intelligence
Search intelligence platform Your crawl, Search Console and Analytics data in one model of every page.
Search Intelligence in QueryRev AI is one model of every page of your site: what the crawler found, what Search Console and Google Analytics 4 recorded, and how each has changed over time. It holds sixteen months of Search Console history, Analytics 4 by landing page and a crawl that keeps versions.
A case, start to finish
A sample travel site ships a template release on 2 September, and its city pages start naming the country page as their canonical. Follow what the crawl, Search Console and Analytics 4 show, from the first changed page to the recovery.
Every figure in this case is read from data QueryRev stores: the crawl, Search Console and Analytics 4. A finding like this becomes a typed decision with a value range.
What it stores
A question about search is usually a question about time: what changed, and when. So every source is stored day by day and version by version, never as a picture of this week alone.
Search Console, in full detail
Sixteen months on first connection, then every day: clicks, impressions and position by page, query, country and device. Clicks whose queries Google does not disclose are shown as hidden by Google, so page totals match what Search Console itself shows.
Analytics 4, by landing page
Organic sessions, engaged sessions, key events and revenue for each landing page and day. Revenue from Analytics 4 is labelled GA4-reported until Revenue Intelligence joins it to your store or CRM.
A crawl that keeps versions
QueryRevBot obeys robots.txt and reads each page as a search engine does: status, redirects, title, headings, canonical, robots directives, hreflang, structured data and internal links, rendering JavaScript where a page needs it. A new version is kept only when the page changed.
In the loop
Search Intelligence is the first step of the QueryRev AI platform: the measured starting point every later step reads. A decision cites its evidence from here, a policy check reads the state of the page from here, and verification compares the days after a change with the days stored before it.
In QueryRev AI each page is a managed asset in the Search State Graph, with its role, its intent, the queries and AI answers it appears in, the product it sells and the revenue it earns.
The inventory is also what a plan is measured in. Pricing counts managed URLs, the canonical and working pages of the sites you connect, never tokens or AI credits. Terms such as canonical and managed URL are defined in the glossary.
Questions
Short answers first. Ask anything else at business@queryrev.com.
What is search intelligence?
Search intelligence is a single, current model of how a website stands in search: every page with its technical state, the queries it appears for, its clicks and conversions, and how each has changed. In QueryRev AI it is built from your crawl, Google Search Console and Google Analytics 4.
Which data does QueryRev AI read?
Google Search Console, Google Analytics 4 and its own crawl of your site. Access to Google is read-only. Search Console is stored by page, query, country and device for sixteen months back, Analytics 4 by landing page and day, and the crawl page by page with versions.
How often is a site crawled?
Once a week in full, and sooner for pages that change. QueryRev reads your sitemap every hour, so on a site whose sitemap carries modification dates a changed page is read again within the hour. The crawler identifies itself as QueryRevBot, obeys robots.txt and paces itself at about one page a second.
Is it a technical SEO crawler or a rank tracker?
Neither on its own. The crawler records what a technical audit needs and Search Console supplies positions, but the value is the join: one record per page that holds both, with history, so a change in the HTML can be set beside the change in search that followed it.
Does Search Intelligence use AI?
No. Everything it stores is pulled from Google or computed from the crawl. QueryRev Intelligence, the decision layer, reads this model to explain findings and draft decisions, and it never supplies a figure: every number is computed from your data.
Keep exploring