Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!
So I have a table that is showing our trailing 12 month retention which is made up of these pieces and looks like this:
I then attemp to break this down further by employee size per company. I'll just show one size group here since all the others are set up identical. Here is also the raw data I am basing this off of:
Attempted to created a measure that would count the number of employees per company:
Then the retention based upon employee size:
So as you can see, my pink graph above has spikes at 1/31 of each year. This is the date where we drop most termed clients so thats why it spikes. I was expecting when calculating the same but taking employee size into consideration instead of just start and end dates, that each group on this graph would spike too, but it is only spiking on 1/31/2017. I feel it has to do with my 'Employee Count' measure I created to count the number of employees per company. After doing some digging around, I discovered the starting period number is what is throwing everything off, but I'm not sure why. For reference to the below screenshot, 'Cservice' is referencing the pink graph, 'Employees' is the yellow graph which are my datasets. Far right is the difference in calculations for each column between the datasets (they ideally are supposed to match or be pretty close):
I'm not sure why the 2017 numbers are pretty identical which is what I expect, but the other numbers are miscalculating. Is there a better way to achieve what I'm trying to do? Or what formulas need changing to set this up better? Thanks!
@Anonymous,
Would you please share the complete data of your table for us to test? You can share the data via Private Message.
Regards,
Lydia
User | Count |
---|---|
128 | |
108 | |
99 | |
65 | |
62 |
User | Count |
---|---|
137 | |
115 | |
102 | |
71 | |
61 |