I've got a table of measurements that are collected weekly, in which measurements could be missing or late.

 id name date measurement period 1 customer 1 Jun 1, 2017 8875 rolling half year 1 1 customer 1 Jun 8, 2017 8908 rolling half year 1 1 customer 1 Jun 16, 2017 8934 rolling half year 1 1 customer 1 Jun 22, 2017 8958 rolling half year 1 1 customer 1 Jun 29, 2017 8996 rolling half year 1 1 customer 1 Jul 6, 2017 9018 rolling half year 2 1 customer 1 Jul 13, 2017 9046 rolling half year 2 1 customer 1 Jul 20, 2017 9065 rolling half year 2 1 customer 1 Jul 27, 2017 9097 rolling half year 2 2 customer 2 Jun 2, 2017 532 rolling half year 1 2 customer 2 Jun 8, 2017 541 rolling half year 1 2 customer 2 Jun 15, 2017 548 rolling half year 1 2 customer 2 Jun 22, 2017 556 rolling half year 1 2 customer 2 Jun 29, 2017 559 rolling half year 1 2 customer 2 Jul 7, 2017 562 rolling half year 2 2 customer 2 Jul 13, 2017 567 rolling half year 2 2 customer 2 Jul 20, 2017 569 rolling half year 2 2 customer 2 Jul 27, 2017 586 rolling half year 2

Is there a way to select all first values of the "rolling half year 2" ?

• Bart Jonk
• Janice Janczyn

Hi Bart,

Happy to hear you solved the question yourself.  Here's another approach that uses Actions instead of functions & formulas.

If you want to group measurements by name, selecting only those in the rolling half year 2 period, you could for example build a table with a hidden data column.

Column 1 - points to the 'name' column in your data
Column 2 - points to 'measurement'
Hidden Data - points to 'period'

Then do the following steps:

1. Click the 3 dots menu to the right of Hidden Data (in the component hierarchy), select Filter and set the Condition to Match 'rolling half year 2' (exclude the quotes).
2. Click the 3 dots menu to the right of Column 1 and select Group. All other columns in the table are automatically aggregated, now you need to set the Aggregation method for the other column.
3. Click the 3 dots menu to the right of Column 2 and set Aggregation > First.