This is just some announcement which people will want to pay attention to! Learn More
This is just some announcement which people will want to pay attention to! Learn More

A nationwide chain of American grocery stores uses advanced AI analytics to predict workforce attrition rates

Employee turnover comes at a cost

Employee turnover in retail, especially among hourly workers is on average more than 60% and can sometimes be as high as 90%. These turnover rates can directly impact bottom line and according to our estimates, the cost of replacing just one distribution center worker can be over $7,000 if you consider 1) Termination, severance pay, etc.; 2) Supplemental cost of overtime paid to other employees until a replacement is found; 3) Additional advertisement, recruitment costs;

4) Drug Testing, background checks, relocation costs; 5) Orientation and retraining costs. The success of efforts to improve attrition usually hinge on a business’ ability to predict and manage the reasons why employees depart from the organization.

Seeking predictability of turnover at an individual employee level

Attrition was an especially big problem for one large grocery store chain in the United States. With their many locations serving a variety of purposes, from selling directly to consumers to producing and packaging store-brand products, the company wanted to more effectively predict attrition rates, as well as analyze and monitor reasons for employee loss so that they could more expertly address them.

The company was able to compile data on employee attrition, including the average number of weeks employees stayed with the retailer, and the most common reasons for employee departure, which included job abandonment and dissatisfaction with the type of work.

But to make accurate predictions, and therefore effective changes, based on that data, the business turned to artificial intelligence. They explored several packaged solutions for predicting employee churn but none of them were tuned to uniquely account for challenges being faced by the retailer like considering the impact of the first 24 hours of an employee with the company, which seemed to determine a large percentage of exits. Also, the tools needed the retailer to first store their employee data in a certain format, which would require significant changes to the way data was collected.

The Grocer needed a solution that can not only be customized to their business but also gave them insights on the top factors impacting the turnover. This is when they turned to CrowdANALYTIX.

Our approach

CrowdANALYTIX used existing employee data from 2500+ employees across three facilities and built feature sets by combining those data sets with external data from competitors and credible public sources. This was used to create models predicting employee attrition and identify key factors in examining why workers left the company.

A usable dashboard was created that showed managers which of their employees were particularly at risk and why. The dashboard could also accept new and ongoing data so that new employee information could be used to track progress.

Some of the top predictors of attrition included factors that required more systemic long-term changes, such as employee commute time and medical problems. However, other factors gave the company the power to make positive changes that might improve their attrition rate, such as focusing on the work/life balance in particular roles and closely managing workers on certain shifts. Also, since almost 88% of the turnover happened in the first year, a system that constantly monitors the risk of attrition allowed them to intervene before things got bad.

Impact

Provided a customized solution for predicting employee attrition. In limited field trials in 3 distribution centers, the initial results were very encouraging, helping the client reduce attrition by almost 34% resulting in an ROI of almost 480% on the investment.

Over time, CrowdANALYTIX will add more features to help recruit the right talent and integrate the solution into a more complete AI-driven human resource management system across distribution centers and retail stores. 

Data used: Workforce data

Share This Post

Share on facebook
Share on linkedin
Share on twitter
Share on email

More To Explore

Thanks to Sergei Akulich for sharing their work on Unsplash.

Monitoring Air Quality with AI

It’s likely that when you heard the phrase “air quality” a year ago, the first thing you’d think about would be air pollution: smog, ozone,