Closed eddierubeiz closed 2 months ago
@eddierubeiz this has failing test
Yup - hopefully it will pass once we have 2723 merged into it.
Yup, it's passing now.
In addition to our back-and-forth about what the data-attribute API shoudl be and how we should record it in GA4 -- i dont' want to merge this until we have actually verified that it's getting recorded in GA in a way that we can usefullly access in reports and consoles etc. And that whoever is accessing GA reports knows how to get to. This stuff isn't actually straightforward.
If we write code taht's trying to do something, but isnt' actually doing it in a useful way and none of the GA-reporting staff uses, and we don't realize it, it will just confuse us down the line.
(We also still have the option of NOT recording thigns that our GA-reporting-staff didn't actually ask for? Then we don't need to do the work to make sure it's being recorded in a way they can access usefully?)
It's possible you have already done this testing, but we just need to docuent it's been done here, for traceabilty and record-keeping!
We've decided we don't need to record the search phrase in Google Analytics after all, at least for now.
Ref https://github.com/sciencehistory/scihist_digicoll/issues/2606.
event_value
parameter, which we don't use anywhere else in the digital collections, to report the query string to GA. (This is a bit of an experiment.)data-analytics-value-css
.data-analytics-value-css
isn't specified.