The discussion revolves around a drop in logged-in user pageviews observed since June, reportedly starting with v3.1.0.beta5. yanokwa shared graphs from four forums showing the decline. Falco confirmed that bugs affecting logged-in pageview counts were fixed in late August, making September’s data accurate according to the definition of a pageview as an HTTP request with a 200 (OK) status code Sam Saffron. gassim seeks clarification on whether prior counts were incorrect and if historical accurate data can be recovered.
myserver.com/admin/reports/page_view_logged_in_reqs?chart_grouping=monthly&end_date=2023-09-30&mode=chart&start_date=2023-02-01 is the URL, if you want to compare.
We had a couple of bugs affecting pageview counts for logged-in users, which were ultimately fixed in the late part of August. The numbers of September you have are the correct ones, by our specific definition of what a “pageview” is.
If the correct views are from September, does that mean all the previous counts are incorrect?
So if I understand that correctly, the page views from September are based on the above definition. May I ask what the previous values represented? Is there a way to get the accurate values back?