The detailed study of the motivations that lead an employee to leave a company is an advantageous position for talent management departments seeking to attract/retain the best professionals for their organizations.
Determining turnover rates, identifying behavioral patterns and segmenting workforces according to objective criteria to enhance productive efficiency is a challenge that analytical practices can already address with high success rates.
Needs
-Visualize employee turnover rates.
-Identify those employees who are most likely to leave the company.
-Analyze locations with their turnover levels.
-Navigate through the information to perform the analysis.
Starting data
The work was based on 50,000 records of 10 years of historical data, which included information on active employees and those who had left the company.
The fields analyzed were:
Employee’s age
Gender
Seniority
City
Employee’s unit
Reason for leaving
Employee’s position
…
Work performed
The project has been carried out using a cyclical method based on CRISP-DM, which has allowed progress in the extraction of value from the data, using “agile” Kanban-type tools for the assignment and monitoring of tasks.
Understand the business
Understand the data collection process.
Understand the content of the data.
Assess the quality of the data.
Observe the context of operation.
Clean and standardize
-Data enrichment.
-Externalities search and data acquisition for geo-positioned visualization.
DataWarehouse modeling and construction
-Elaboration of the relational model.
-Definition of fact and dimension tables.
-Construction of relationships and attributes.
Scorecard Construction
-Definition of indicators.
-Definition of alert systems.
-Definition of presentation graphics.
-Structuring display screens.
-Definition of analysis flow.
Prediction
Identification of predictors.
Multi-variate analysis.
Visualization.
Technology
The technology used for the project is as follows:
Results
Geo-positioned visualization of the employees’ situation.
Evolution of employee clocking in/out.
Analysis of the situation of a specific location/store.
Hiring and settlements per year and per month.
Average contract duration by location and store.
Employees by department and by job position.
Type of contract termination (Dismissal, Retirement and Resignation).
Total and average age of employees by gender, status, city and store.
Geolocation of stores.
We learn from the past by observing how data relates.
We analyze data, obtain data of value and provide answers to business questions.
We offer technology solutions that support our customers’ transformation.
