Hi,
Let’s say I have a dataset with columns: job_id | day | sum_clicks. I need to find a unique job_id for the period specified by the filter. I’ve applied an insight by categorizing my ‘job_id’ column. This approach works well, except when certain filters are applied with large date ranges. In such cases, an error occurs: ‘unique values: too many different categories have been found to be able to calculate correctly.’
I tried another approach by creating a calculated field using various variants of the ‘count_distinct’ function with KPI. However, for some reason I don’t understand, all of them seem to calculate non-distinct values, but rather count all values. In addition, the value does not respond to filters.
There are Screenshots with Insight:
and a screenshot of a calculated field using ‘count_distinct(job_id)’. I applied a filter where ‘sum_clicks’ for the entire time period should be greater than or equal to 100. I can see the correct rows in the left table, but I expected to find ‘2’ unique values in the KPI, whereas it’s displaying ‘6’, which includes all the jobs.
So, what can you recommend to me? Thank you in advance