Attrition rate calculation

Question & Answer
, , , , , , , ,
1

Hi ,
I have a requirement to improve an existing dashboard by incorporating an analysis of onboarding month trends and rolled-off month trends. I’ve added the same analysis in Arena. Now, I need to calculate the attrition rate based on dates, using the formula: number of employees who have left divided by the average number of employees .I’m not sure hoe to do this, can you please help me here.

Attrition

2

Hi @hellosai321,
So I understand that for employees who left, we’d want it to be based on the date range; but for the average number of employees portion; would that be calculated by the same dates or a wider range?

1 Like
3

Hi @Brett ,

The numerator will be the rollover count, while the denominator should represent the total number of active employees. For example, if the attrition count for January is 10, the calculation should be 10 divided by the number of active employees as of January.

4

Hi Team,

Is there any solution for it?

5

Hi @hellosai321,
I tried out a couple different potential solutions but with the way the data is setup, this will not be achievable.
Since all data for a given employee is setup within one row, the date fields aren’t able to reference any other field to compare against. So when trying to pull numbers for each month, it’s only able to recognize when someone’s onboard and rolloff dates are in the same month.
I’ve included some of the work throughs that my colleague and I tested for this in the attached Arena view.
Attrition rate calculation

For this to be possible, the dataset would need to be altered to include more than one row per person. Ideally something that would look like:

Name       Status       Action Date     Action Type 
Xx            Active        01/01/24       Onboard
...                                       Rolled Off

With a format like this and a date field that returns every consecutive month should make this possible with calc. fields.

6

Thanks @Brett

I understand it. Just checking if we have any workaround without modifying the data model?

7

Hi @hellosai321,
Unfortunately, I don’t see another work around that could potentially work without modification.