A leading healthcare technology company sought to improve Resource Management Planning by reducing last-minute candidate dropouts.
By applying predictive analytics to hiring data across locations and designation levels, the organisation aimed to move from reactive hiring to more proactive, data-led workforce planning.
Listen
The Challenge
The client needed to:
- Minimise last-minute candidate dropouts
- Reduce repeated sourcing cycles caused by non-joining
- Improve resource planning affected by joining uncertainty
- Address operational inefficiencies linked to hiring disruptions
- Use data-led intelligence to track candidate behaviour and trends
In essence, the organisation required predictive insight to anticipate joining outcomes and plan staffing more effectively.
Act
Insight-Led Diagnosis and Design
Feedback Insights designed and developed a predictive analytics model to forecast the likelihood of a candidate joining.
- Reviewed industry practices to identify relevant predictors
- Identified 18 variables spanning candidate profile, resume source, and internal processes
- Designed the model to account for differences across locations and designation levels
The model provided a structured, scalable approach to predicting joining outcomes and supporting hiring decisions.
Accelerate
- Assigned weights to variables using discriminant analysis
- Conducted separate analyses by designation level and key locations
- Categorised candidates as hot, warm, or cold using quartile analysis
- Built and integrated the predictive model into the client’s system
- Applied the model using real-time data, including “Time to Join” data points
Business Outcomes
- 80% accuracy in predicting candidate joining
- Identified critical variables influencing a candidate’s intention to join
- 20% improvement in joining metrics enabled by analytics-led intervention
- Improved resource planning and operational efficiency
