Article

HR analytics and employee turnover

HR analytics is essentially based on the well-known management maxim that what cannot be measured cannot be (efficiently) managed, applying this maxim on human resources. In the last few years, an extension has been added in the form of more advanced analytical procedures that have a better potential to bring deeper insights and recommendations with greater effect. But whether you use only basic reporting or some more advanced analytics, the objective is always the same – try to use data and their analysis to suitably influence the individual HR processes that help organisations achieve their strategic goals. This is illustrated by the diagram below which depicts the mechanism connecting HR processes with (not just) the financial results of the organisation.

image

HR analytics helps us optimise the set-up of this mechanism by allowing us to find answers to certain key questions, such as:

  • Which channels bring us the best candidates?
  • What characteristics differentiate successful candidates from unsuccessful ones?
  • What factors contribute to successful onboarding?
  • Which KPIs have the strongest link to the company’s financial results?
  • Which training sessions are most likely to lead to improvement of work performance?
  • Which interventions have the biggest impact on well-being or work-life balance perceived by employees?
  • What increases or decreases the employees’ engagement level?
  • Where does the organisation have isolated silos and bottlenecks that prevent efficient communication and cooperation between individual employees, teams or whole departments?
  • Who represents hidden talent that needs to be detected and further developed?
  • Where can resistance be expected with respect to planned changes in the company and who can instead be their ambassador or catalyst?
  • Which factors contribute to employee turnover and what specific employees are at a high risk that they will leave the company within the next few months?

This last method of use of HR analytics often represents one of the first types of HR analytic projects that companies start with, for an easily understandable reason. The unwanted departures of employees often lead to high direct and indirect costs, so even a relatively moderate decrease in employee turnover can represent significant savings that will be appreciated by any company management. The urgency of this problem is additionally exacerbated by the current stage of the economic cycle with an all-time low level of unemployment, which, together with various on-line job search platforms, motivates many people to look for a new position where, as they hope, work will be more meaningful, more interesting and better paid and the colleagues will be nicer and the bosses more inspiring.

Given the acuteness of this problem which plagues quite a few companies, it is not surprising that a large number of various studies have addressed and keep addressing the topic of turnover. For example, late last year an extensive meta-analysis was published by Rubenstein, Eberly and Lee, who synthetised the results of more than 300 individual studies concerning turnover predictors. We would therefore be justified in asking what new findings can be brought by HR analytics focused only on the employee turnover in one organisation. Has everything significant regarding this topic not been discovered already? (For more on this question, see e.g. this inspiring article by Thomas Rasmussen). It is true that it is not very probable that you will discover a completely new turnover factor while analysing your own data. On the other hand, it is also true that each organisation is, in some ways, unique, so certain retention factors will probably be more important for the organisation and others less so. This information about the relative importance of the individual retention factors is then key for setting up a retention plan, and HR analytics can be very helpful in this respect.

With the use of this dashboard you can try out on your own how useful the outputs of such an HR analytic project focused on employee turnover could be for you. The dashboard contains information that helps (not only) the management to answer a variety of key questions that are at the beginning of every effective employee retention plan, such as:

  • How many employees leave us each year?
  • Which groups of employees leave the most often?
  • What is the external benchmark? Are we doing about as well as the competition in the industry?
  • Does the current turnover rate represent a serious problem for us and is it therefore worth trying to resolve it?
  • Why do people usually leave their jobs in general?
  • What factors contribute to the departures of our employees specifically?
  • What pro-retention measures are available in general?
  • What pro-retention measures should we select given the probable reasons why our employees leave?
  • What employee groups should we focus on especially in terms of prevention of turnover?
  • Which specific employees are at a high risk of leaving and what specific retention factors should we focus on during regular stay interviews with them?

As evident from the list of questions above, the dashboard contains information that can be used in the decision-making process not only of HR managers, but also HR business partners or even team-leaders and line managers of the individual teams and departments. In addition, the dashboard contains a variety of technical details about the prediction model used and the actual data underlying all the presented visualisations and analyses. With their help, HR/Business analysts can e.g. look for the optimal way to set up the scoring algorithm in order to maximise the positive effect of the pro-retention measures, or they can search in the available data for some additional useful information. For more details, see the dashboard below (it is possible that depending on the current version of your browser, you may encounter problems launching the dashboard; if this happens, try to launch the dashboard in a different browser).

image