I have a scenario with a website where starting a few months ago there was an entire URL restructuring of the entire website.
Ideally I'd like to use the native Google Analytics Connector in Tableau to extract 13 months of data to have a historical view of these URLs.
The problem lies within that the old URLs don't match the new URLs so when looking at a trended table/chart the data for the 'same' page do not line up.
e.g. Old URL: /about-us == New URL: /who-we-are
I do have a spreadsheet that can be used as a lookup table which contains two columns: Old URL and New URL
However I don't think regular joining will work because I have to either join the GA Connector Page dimension to the Old or New URL in the lookup table and this doesn't do anything for me.
Is there a simple calculation I can create in the GA Connector Data Source where I lookup the Page dimension against my lookup table source and then output the NEW URL?
I tried mixing dimensions from each source into a simple IF THEN calculation which forces me to use the ATTR function, but the problem is there isn't a consistent 1 to 1 matching, because there are numerous Old URLs that now map to a single New Page and this forces the ATTR function to just return a *
Just for reference [Page] = the new pages coming from the GA Connector and [Old Url] is obviously the old URL in my lookup table spreadsheet source.
IF ATTR([Page]) = ATTR([Old Data)].[Old Url])
ELSE ATTR([(Old Data)].[Old Url])
I can ETL this outside of Tableau, but for this solution/project I would ideally like to avoid it.
Anyone encountered this or something similar and have any pointers?
Thanks in advance!
I know you don't want to use external ETL tasks, but in this case it's the best way to fix the data. Tableau Desktop will want a field to join to get the data from the lookup table. In your ETL you can do conditional joining to change the format of the old URLs only.
Sorry, but I'd be using an ETL process to fix this before you use the data in Tableau