Case study card:
Predicting Employee Attrition
The supplier was briefed to predict employee attrition in order to prevent losing high-value staff members.
Typically around an organisation you will have a disparate set of databases such as salary databases, employee address databases, onboarding and recruitment records, etc. They are probably maintained by different departments.
The first step would be to find a way to unify the datasets so that for every current and past employee you can easily access all data about them. You want to know when someone joined the organisation, when they advanced pay grades, and when they left.
The supplier developed a machine learning model to predict which of the NHS’s 1.2 million employees were likely to leave the organisation.
They focused on junior doctors training pathways, increasing retention and satisfaction while reducing hiring costs.
20-day engagement for the NHS: find out why junior doctors were leaving the organisation. For context, the NHS invests £200k training a junior doctor from completing their medical degree, to qualifying as a consultant (e.g. cardiologist, urologist, etc) or GP.
This training can take 6 years. Doctors are prone to dropping out on the training pathway.
The project had several aims:
Developed a predictive model in Azure to identify people at risk of leaving
Developed recommendations to reduce attrition
Taught internal NHS analytics team how to use machine learning for this
Helped the NHS proceed to a Beta phase of this project
Delivered recommendations to reduce attrition
Left the analytics team with skills and recommendations to incorporate ML into workflows
UK National Health Service