If you’re new to Qlik Sense, start with this Discussion Board and get up-to-speed quickly.
Hi,
i have created a line graph where it displays counts of open, which is delivered and opened on firstday for a particular batch hourly.
Dimension:Hour(formatted_opened_date)
Expression:COUNT(Distinct Aggr(IF(formatted_opened_date = Min(Total <batch_meta_data_id>formatted_opened_date),invitation_id),formatted_opened_date,invitation_id,batch_meta_data_id))
Now i want to do a accumulation line chart, So i changed expression to :
rangesum(above(COUNT(Distinct Aggr(IF(formatted_opened_date = Min(Total <batch_meta_data_id>formatted_opened_date),invitation_id),formatted_opened_date,invitation_id,batch_meta_data_id)), 0, Rowno()))
But this is not at all giving accumulated records.
Please help me on this!. I have attached a sample app.
I mean i need to show accumulation of opens% in line graph hourly for first day.
For this formula is (Opens / Sends).
i.e (opens count for first day for 2nd hour)/(sends count for first day for 2nd hour)
So i changed like this, this does not work
Aggr(rangesum(above((COUNT(Distinct Aggr(IF(formatted_opened_date = Min(Total <batch_meta_data_id>formatted_opened_date),
invitation_id),hour_first_opened_date,invitation_id,batch_meta_data_id)) /COUNT(Distinct Aggr(IF(formatted_send_date
= Min(Total <batch_meta_data_id>formatted_send_date ),
invitation_id),formatted_send_date ,invitation_id,batch_meta_data_id)) ), 0, Rowno())), batch_meta_data_id,
(hour_first_opened_date, (NumericCount)))
okay, so what is the issue you are running into when you do this?
Its shows more than 100% and percentages are wrong.
Not sure... can you share a qvf with the incorrect numbers and let us know what the correct numbers are....
Hi Sunny,
Below is qvf file. Percentages showing in open Rate chart are wrong
example: for 3rd hour: opens=85 sends = 4279 for batch = 96. So opensrate = 1.98%. But it shows 1.47%.
Note: when you select single batch at a time it shows correct value.
You have different dates in the two chart... I think you need a Canonical Date