5 Reasons to Start with Predictive Employee Turnover Analysis – HR Analytics
As Prof. Baesens stated in a previous blog post on predictive employee turnover analysis, employee churn is a major problem for many firms these days. Prof. Baesens: “Great talent is scarce, hard to keep and in high demand. Given the well-known direct relationship between happy employees and happy customers, it becomes of utmost importance to understand the drivers of employee dissatisfaction. In doing so, predictive analytics can be a core strategic tool to help facilitate employee engagement and set up well targeted employee retention campaigns“. In our 2 most recent posts, we’ve explained the benefits and challenges of starting with predictive employee turnover analysis. In this blog post, we will explain the unique advantages of doing predictive employee turnover analysis.
The past few months we’ve experienced a strong interest increase in predictive employee turnover analytics (PETA). No wonder, by doing predictive analytics the best-managed companies are discovering the power to predict who will leave, perform, promote or sell. And underneath their employee turnover data (and all other HR data), lies unparalleled prediction potential. And as line managers will get used to work with predictive analytics, HR will get under pressure to move along with it and create meaningful business value by providing deep insights in drivers of e.g. employee turnover (or any other business related topic). Initial results of doing PETA were nothing short of astounding: depending on the project, 20 to 50% more accurate turnover insights were reported. Imagine this in the context of Strategic Workforce Planning! It turns traditional (descriptive) turnover analysis (or projections) to an inferior ‘product’ definitely. Why is that?
1. 3-Dimensional analysis
Instead of reporting the employee turnover of the last month/quarter/year per department, talent cluster (e.g. high potentials) or business unit (what we call a 1-dimensional view), PETA looks at employee turnover in a 3-dimensional way:
- 1st Dimension: historical patterns –> what were turnover patters in the past years?
- 2nd Dimension: shifts in those historical patterns –> how did these patterns evolve over the past years?
- 3rd Dimension: relationships in the data –> what new/existing relationships exist between turnover and employee factors?
Standard 1-dimensional reporting systems tell you what happened “yesterday” which has little value in a fast-changing context. On top, reporting or HRIS systems are not capable of discovering patterns and the relationships in these patterns. By using predictive algorithms such 3-dimensional insights however can lead to a significant improvement in studying employee turnover and beats the degree of insights of any traditional, descriptive approach by 20 to 50%.
2. Using predictive software to study large historical datasets
Most HR professionals don’t go any further as ‘the past month, quarter or year’ to study turnover. They use averages and large aggregations instead of diving deep in the historical and individual patterns of turnover. “The value of analytics is derived from identifying a group of people who – in aggregate – will tend to behave in a certain way”, says Eric Siegel in his book ‘Predictive Analytics’ (1). This identification process, which studies the characteristics of every single employee over a certain period (mainly a few years) is not possible without the use of predictive software. It’s time to face reality: line managers really want to know what problems or opportunities will occur over the next week or months so they can work to prevent them. That’s where PETA can really add tremendous value.
3. Predicting future turnover risk
One of the hottest areas in business is making predictions for important organizational risks. Unfortunately, most HR departments simply don’t have expertise with calculating risk. Assessing impactful problems such as key employee turnover, work accidents or absenteeism can benefit enourmously of predictive analytics compared to the use of historical reporting. As you can see in the table below, PETA calculates risk scores for each individual and aggregates this into high risk clusters, such as age groups, high potential groups, salary groups, etc. Almost without exception, the usual HR turnover metrics only include a single number (i.e. our turnover in division ‘X’ was 6%). But such a single number doesn’t tell you at all where you stand with regards to future turnover risks. On the basis of PETA’s risk predictions, organisations will be able to start specific retention campaigns however.
4. Asymmetric clustering
Instead of the usual reporting clusters such as department, division or business unit, PETA constructs what we call assymetric clusters based on the risk profiles of these automatically generated clusters. This could be a gender group, a salary group (within a certain, non-company specific salary bandwidth), a tenure cluster, a performance cluster, etc. Again, this creates an enourmous advantage versus the usual approach within HRIS or reporting software. It combines both the risk assessment (see point 3) as well as the clustering/classification of specific (and often unrelated) target groups for retention campaigns. Such a clustering exercise has also tremendous value to assess the turnover of critical talent groups within the context of strategic workforce planning.
5. What-if scenario’s
Finally, advanced analytical processes such as PETA allow decision-makers to develop models which allow them to try out different alternatives and to vary the constraints and the assumptions in order to see how the results would change. With the pace of change accelerating, there are more unexpected events than ever that are both high in business complexity and risk. These decisions require rapid and accurate human judgment – facilitated by scenario modeling driven from predictive analysis. Scenario modeling capabilities in HR will empower front-line decision makers and executives to make decisions when the business requires it. A though challenge for HR…
HR managers with no vested interest in anything other than “getting the turnover analysis job done” won’t change behaviors quickly. Here’s one more prediction from myself: the number of HR Managers not willing to jump on the predictive band wagon, will be decimated in the coming 3 to 5 years! PA is disruptive HR innovation and it will confront HR’s own self-definition: will I accept or resist to work with the machine? HR, your identity is at stake! Get on with it!
About the author
Luk Smeyers is an experienced senior HR executive who has lead complex transition projects for compelling Fortune 500 companies, such as PepsiCo, Starbucks and Nielsen. He has managed at board level through business start-ups, turnarounds, acquisitions, divestitures, joint-venture creation, down-sizing and restructuring on a pan-European basis. Till 2007, Luk served as Senior Vice President and CHRO Europe for Nielsen, the world biggest marketing research and data company. In 2008, Luk started a cutting-edge HR Analytics consultancy, together with academic partner Dr. Jeroen Delmotte. Luk is widely recognized as one of the few European top HR analytics experts. He is revered as a leading thinker, educator, influencer and is a well-known content contributor, blogger, columnist and author of many articles. He is an invited speaker at international conferences and helps clients set a higher ambition for strategic HR intelligence, leading consultative projects in the Benelux countries with such organizations as ING, KPN, ABN-AMRO, Philips, Rabobank, UWV, RealDolmen, Acerta, NS, BASF, Besix, Strukton, Bekaert, Randstad, Eandis, AG, Postnl, AON, Raet, etc.
2 Related articles: 7 Benefits of Predictive Retention Modeling and 4 Challenges with Predictive Employee Turnover Analytics
Interested in using predictive HR analytics as a key component in your HR strategy? Get in touch with CEO and co-founder Luk Smeyers for more information ([email protected] or via Google+). Or follow iNostix on Twitter and/or Facebook for exciting international articles on HR analytics.