My company has tens of thousands of products, and millions of articles. Here's a pretty common request:
- I've attached a list of 100 products with their IDs. We think we can bundle them into a new sellable thing. Can you tell me how many unique visitors looked at them?
- I've attached a list of 1,000 article IDs affected by this issue. Can you tell me how many visits they had?
My workflow varies:
- If I need user/visit deduplication as in the above questions:
- Go into Ad Hoc Analysis (or Workspace), create a segment, filter hits where the product ID "contains any of" and paste the whole list. Cross fingers that the IDs are well-formed and not contained in other ids (e.g. abc not matching on abcde). Run report I need.
- Go into Ad Hoc Analysis (or Workspace), pull up the dimension, filter, and paste the list under "Any of these words". Hope for the same.
- If the list is very long or I need exact matches on IDs, create a classification and upload this list. This becomes very slow and adds a lot of junk to our report suite that others don't need and doesn't have long-term use.
- If the metric I need is already de-duplicated against the dimension (e.g. pageviews, only one product ever per pageview), I might just export the list and do a VLOOKUP in Excel. Doesn't work well when we're talking hundreds of thousands or more values.
- If I'm really crunched for time, write a python script that reads a CSV and builds a segment with that many rules. This works fine for me but not others.
Ideally, what I'd like is for Adobe to recognize that I am pasting in newline characters, and create a long array of inputs with those values in it. So, if in the advanced filter/search, I focused on the first value field, and I pasted my list, I'd actually get N rows with each row filled out.
Here's an example of how another tool I use handles it. I have entered the first two, but pasted a list after. Each becomes its own line item.
Category Overview.webm - Google Drive
