What is one of the most frustrating situations a business leader can imagine? Employees leaving their jobs.
Not only does knowledge and experience get lost but also the time and resources invested in acquiring, selecting, and onboarding new candidates. That’s why companies need to answer the question: Which employee groups are at risk of leaving the company in the future?
Anyone with management experience might have a gut feeling about why individual employees quit but may miss larger patterns. One solution to get a holistic view is to bring experience and data together to enrich personal gut feeling with reliable and objective numbers.
Our organization started this journey with a pilot project called “employees@risk” that applied predictive analytics, which means using data to forecast people’s behavior. The main priority and challenge with this pilot was the development and the application of a predictive model that complies with the highest data protection and security standards. Therefore, an anonymized dataset containing more than 300 variables was used to identify key attrition drivers.
The resulting predictive model uses these attrition drivers to identify employee groups with a higher risk of voluntarily leaving our organization. The model becomes “predictive” because it is continuously fed with the latest data and automatically considers every single data change. Such models empower decision makers to foresee the attrition risk of employee groups, looking six months into the future.
The responsibility is to use the results in favor of employees. The data patterns can’t be used for individual cases because of anonymization, but serve as an additional and objective information source that leaders can use to back up their decisions.
The aim of every company is to retain its biggest asset – the people – and predictive models can contribute to better programs for employees. Data science can help create a smarter workplace, enabling HR organizations to use data to make intelligent, unbiased, and future-oriented decisions.
Everyone knows that people more often leave bosses, not jobs. This makes it crucial to understand why Everything You Know About Leadership Is Wrong.
