Deriving Recommendation via Visualizations from Employee Churn Data
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
README.md
employee retention.ipynb

README.md

Problem: Employee Churn

This notebook details the approaches and the conclusions of how I approached discovering what leads to employee churn rates. The data is made out different employees from 12 different companies. The data we have includes the department they worked in, the date they worked there, whether they quit, years of seniority, and salary.

This notebook covers:

  1. Data Cleaning
  2. Data Exploration
  3. Creating a Decision Tree as a Statistical Tool
  4. Conclusion and Recommendations

Conclusions:

  1. Employees are most likely to leave a company after one year of working there and this peaks again every year after that.
  2. The top reason why people leave is salary. The highest churn rates appear in the middle range of salaries and churn rates are lowest in when the salaries high and low.
  3. Years of Seniority is the second highest reason for employees quitting, this could point to the salaries and seniority being correlated.

Recommendations:

  1. Companies should pay extra attention to employees after one year of working there, offering extra incentives at the end of one year.
  2. The biggest incentive they could give is higher salary, which is shown to be the most important reason that people stay or leave.
  3. Companies should create extra incentives for customer service employees that have the highest rate of churn

Aside from that, I hope you enjoy the notebook!