Connected data
Search, analytics and edge data in one place
What Search Console, Analytics and Cloudflare each add to a crawl, what we read, what we never touch, and the questions only the combination can answer.
A crawl tells you what is wrong with a page. It cannot tell you whether anyone was on it. Three connections close that gap, and the point of having all three in one place is not convenience - it is that each answers a question the others cannot.
What each one brings
- Search Console - impressions, clicks and average position per page, straight from Google. This is the only source that knows what people searched for before they arrived.
- Analytics - sessions, engaged sessions, key events and revenue per page. The only source that knows what happened after they arrived.
- Cloudflare - requests, cached share, threats blocked and origin response times at the edge, including the traffic that never reaches your server or your analytics tag.
Why the combination is the point
Each of these is available on its own, free, from the vendor. What none of them will do is put a finding next to the number.
"This page is slow" is a technical fact and a low priority. "This page is slow and it is your third most visited page" is the same fact and the top of your list. The crawl produces the first half and the connection produces the second, and the report ranks by the pair rather than by severity alone.
- A page with findings and no impressions is a page to fix last, or delete.
- A page with impressions and a falling position is worth opening before anything the crawler flagged.
- A page the edge serves entirely from cache is one where an origin timing finding matters less than the number suggests.
How a page is matched across sources
By normalised URL: scheme, www and trailing slashes removed on both sides. That sounds obvious and it is the part that quietly goes wrong - a crawl URL carrying a campaign parameter has to match the bare path Analytics reports, or the busiest pages on any site with campaigns land in the "no traffic" column and look safe to ignore.
A page one source knows and the crawl never reached is still shown. Dropping it would be a traffic report that silently omits rows, which is the worst kind.
What we read, and what we do not
- Read-only, always. The scopes we ask for cannot change anything in your Search Console or Analytics property, and nothing we do writes to them.
- Per page, not per person. We read aggregate metrics keyed on URLs. No user identifiers, no session recordings, no cookies of yours.
- Only the property you pick. Connecting an account does not give us the rest of it, and disconnecting removes the stored token immediately.
- Tokens are encrypted at rest, and a connection whose token stops working is reported as broken rather than left looking healthy.
You do not need any of them
Every check, the health score and the whole report work with no connection at all - the crawl is the product. These make the ranking of what to fix better informed; they are not a condition of it.
Last changed 2026-08-10. Every check this refers to is listed on the checks page.