Difference between using (in the Field Metric) Count and Unique Count

[rosammb (Rosa)]

Hi,
I'm using 6.8.3 Elastic version.
I'm having problems when I create visualizes and using the Field Metric. If I use Unique count the result is different than if I use count. My feeling is that Unique count is the correct value but, if I generate the visualize monthly, the result is different than if I generate daily (and get the total).
for example:

Please help me because I'm going crazy with this

[tsullivan (Tim Sullivan)]

If there are duplicates of the ID, but they don't duplicate for the same day, they will be counted as unique on a day-by-day basis. That's why the first chart shows a larger number because it's just a sum of all the numbers in the column.

If duplicate IDs happen on different days, then the first chart says they aren't truly duplicate. But the second chart says they are duplicate because the second chart is not discerning records by a daily basis.

[rosammb (Rosa)]

Thanks for ur reply. Is there any way to solve this problem? (e.g. an index or a filter )
Mainly I use the monthly report and from my point of view, the result is not correct due the duplicity.
Many Thanks

[tsullivan (Tim Sullivan)]

This isn't a Kibana issue, the tables are simply showing you the nature of your data.

If the underlying data is wrong, then it's an issue to address before the data is indexed into Elasticsearch.

If the underlying data is right but people would be surprised to know that IDs are duplicated across different days, then I would take that as an insight that you can share with people.