How to calculate risk of loss
Last month I wrote about the job satisfaction model for employee retention. Now I have used the model to develop a model that allows you to calculate the probability of losing an employee, based on their personal retention and turnover factors.
In my mind the risk of loss is really the probability of the loss event occurring, i.e. a decimal number between 0 and 1. So the model really is an exercise in statistics, to work out what that overall probability is. For each of the factors in the job satisfaction model, we can use the data stored in any good HRMS system to give us more information.
Poor Pay — Leads To Dissatisfaction
We could compare the employee’s rate of pay with the market rate (using their compa-ratio). A low compa-ration means they are paid under the market rate; a high compa-ratio means they are paid higher than the market rate. Therefore if the employee’s compa-ratio is 1, this would be a factor for a neutral stance.
Also, since most compensation models allow for increase of pay over time, we could look at both compa-ratio and length of time in job code. Low compa-ratio combined with long service in job code would imply dissatisfaction; low compa-ratio/short service in job code implies neutral; high compa-ratio/long service in job code implies neutral (but also implies that the worker should be ready for promotion too, and if they’re missing out on promotion then they’ll be dissatisfied — see the paragraph below); high compa-ratio/short service in job code implies neutral.
Poor Compensation — Leads To Dissatisfaction
We could look at the average bonus versus the employee’s bonus. If they are under the average then they could be dissatisfied; if they are over the average then they would be expected to be neutral.
Lack Of Promotions — Leads To Dissatisfaction
We could look at the time since last promotion. Long time period implies dissatisfaction; short time period implies neutral.
Lack Of Job Security — Leads To Dissatisfaction
We could look at the number of voluntary terminations and involuntary terminations for the employee’s job code or job family, for the last 12 months. High number of terminations implies dissatisfaction; low number of terminations implies neutral. Higher proportions of voluntary terminations imply that current/recent employees are choosing to leave the organisation, so could imply job insecurity and hence dissatisfaction; higher proportions of involuntary terminations imply that the organisation is downsizing, so could again imply job insecurity and hence dissatisfaction.
Good Leadership Practices — Increase Satisfaction
We could use the employee’s rating of manager performance in 360 reviews. Scores in the top quartile or above a certain threshold indicate satisfaction; scores in lower quartiles or below a certain threshold indicate neutral.
This could also be used to assess the manager relationship (though it’s imprecise).
Recognition — Increase Satisfaction
We could use the data around awards given to the employee to capture if the employee has received awards/recognition in the last 12 months. If yes, indicates satisfaction; if no, indicates neutral.
Feedback And Support — Increase Satisfaction
We could use the existence of completed performance reviews to measure this one. If yes, indicates satisfaction; if no, indicates neutral.
Also we could measure the difference between the employee rating and the manager rating within the performance document. If they are the same, this implies neutral. If employee is higher than manager, implies dissatisfaction or neutral. If employee is lower than manager, implies neutral or satisfaction.
Clear Direction and Objectives — Increase Satisfaction
We could use the existence of worker goals, or goal plans, and/or existence of individual development plan. If they exist, implies satisfaction; if they do not exist, implies neutral.
Weighting
The metric for each factor can be weighted, since some factors will naturally be considered more important than others, depending on the organisation and their business goals.
Risk of Loss
The risk of loss will therefore be the weighted average of all the factors that influence loss of an employee.
Using this model should help HR departments and line managers gain a better understanding of the multitude of factors that influence an employee’s decision to leave. With the risk of loss to hand, line managers can act decisively to intervene in an employee’s decision to leave and be pro-active in making changes that will retain highly-valued employees.
