I'm looking for the number of unique days that I s...
# gooddata-cloud
d
I'm looking for the number of unique days that I see a user with activity online. However, when I filter to last month I get numbers greater than 31 (which is impossible). I also get a double count of userids (a unique value). Can't figure this one out https://reporting.laivly.cloud/analyze/#/tenant-turo-v3/8a117a4c-846d-426c-8a40-8fe4b7b6c2ef/edit I think this is related to my question above actually. Its doing a ton of unions for no logical reason @Daniel Stourac
m
Hi Doug, please let us review this and we will get back to you with more details as soon as possible.
Hi Doug, I created a copy of your report which can be found here: https://reporting.laivly.cloud/analyze/#/tenant-turo-v3/c8b88c9c-fc49-4d35-afbf-935dd5ff3147/edit There I added the “Case opened - Date” to the report and it now displays the correct numbers and dates. However, when I then go ahead and add the Rollup (Total) to the report, I see it’s then showing the numbers greater than 31, etc as in your report. So I am going to go ahead and discuss this with my colleague Daniel Stourac who is currently assisting you with your previous case, as it seems this may be related.
d
Hello Doug! Dan is currently on holidays. The issue here is that
case_opened
dataset is in more datasets and they are not really connected within the report. So, they are several UNION and SUM... which are united and summed together. It doesn’t know which ones to take as unique, the count of all records (including possible duplicates) is returned. To overcome this, the metric has to be changed: add secondary parameter or add USING to the COUNT, like in here https://www.gooddata.com/docs/cloud/create-metrics/maql/aggregation/count/ Or you need to make the connection to the datasets, by adding another value in the report. For example, if you add
leg_metrics.user_id
it will give you the correct values.