0

I have a report matrix with several columns, where each column shows the same kind of data but with different filters. Specifically, I'm showing some aggregate numbers, but each column shows it for a different time period (one may be daily, another weekly, another for some custom time period, etc). To avoid filtering for time period in each cell expression, I would like to set up some sort of a global filter per column.

At first I thought multiple datasets might be a good answer. I could create a new dataset for each column and filter the query accordingly, and then I could re-use the exact same expression from any particular row for all columns. Realizing that each report table is only bound to one dataset, I started looking for a way to make each column its own table, and just bind them together visually, so that all rows line up and are the same height.

Is there a way to do it this way, or should I try something entirely different? I could have a global dataset that contains all data and then filter per cell, but it seems a lot less efficient. Here is sample raw data and the report I would like to generate. Is there a way to do this without mucking around with the "Date" and comparing it to Today in every single cell?

Raw data and desired report

Egor
  • 1,622
  • 12
  • 26
  • Usually **1 dataset** is better for designing, maintenance and efficiency. Can you share your query, raw data and desired output. – Anup Agrawal Oct 22 '13 at 19:17
  • I can share a hypothetical but essentially identical scenario. Let's say I log phone calls from customers, which includes the datetime when the call was placed, duration, and area code. I want a report that has multiple rows with data like "number of calls from area code 123" or "number of calls lasting more than 5 minutes". For each row I want to see the numbers for yesterday, last week, last month, and last year (as columns). – Egor Oct 22 '13 at 19:25
  • You can use `UNION ALL` to join those groups in a single dataset. – Anup Agrawal Oct 22 '13 at 22:05
  • I don't understand how that would help. My issue is having to filter each individual cell to display aggregates for a specific time period, instead of filtering once per column. I'm trying to split a dataset into groups, rather than the other way around. – Egor Oct 22 '13 at 22:13
  • Without sample data and output it is difficult to understand and explain. – Anup Agrawal Oct 22 '13 at 22:20
  • I've added some sample data and the desired report. – Egor Oct 22 '13 at 22:39
  • Thanks for the sample data. One more question: Are the groups on left are predetermined or is there any logic? – Anup Agrawal Oct 22 '13 at 22:41
  • There is a mix. Some are row groups that can expand, while others are pre-determined. – Egor Oct 22 '13 at 22:47

0 Answers0