Freeform Table - Grand Total is incorrect or not adding up correctly

Question

I am having a data element which records dimension values for a specific event. To get the total count of the values received for that dimension, I have created a calculated metric (Approx distinct count ( dimension_name )). Although, I can see that I am getting 37 Unique values ( Number of rows in the freeform table) for this dimension, the grand total header row for the freeform table shows that there are only 36 values ( 35 excluding NA values).

Jennifer_Dungan · Accepted Answer

Hi,Thanks for your suggestion. I don't think this is due to the data issue from Adobe. I validated other date ranges and I am still seeing the same issue.  If we consider that the "Approximate Distinct Count" is miscalculating, how can it me corrected? or is there any other to calculate the "Number of Unique Values" for a dimension? I was just playing with a dimension on our side... Based on the "breakdown" column, I should have approximately 479K rows, but the Approx. Count Distinct was only showing 475K.. so I was 4K short... So, I tried to make my own calculated metric. First I started with an IF statement: IF [    logical_test [        (Dimension X exists)        Occurrences   ]   value_if_true [       static number = 1   ]   value_if_false [       static number = 0   ]] This returned 1 on every row of my dimension (including the total), which is what I wanted... Then, I edited the metric again, wrapping the above IF statement inside a COLUMN SUM function. This now returns a match to my number of rows, i.e. 479K. Note, this will show on ALL rows, it won't be like the Approx Count Distinct where each row will show 1 and the total will show a total... every row is the column sum.

MandyGeorge · Answer

It looks like the metric you're using there is custom built. If you're able to share how that metric is built that could help give an answer.

But from what I can see in your screenshot, my best guess is that multiple CNs existed in the same visit or hit, and the way you've built out the metric is it's counting instances of at least one CN, so having multiple, it would be deduplicating it.

Sign up

Login with SSO

Login to the community

Login with SSO

Scanning file for viruses.

This file cannot be downloaded