Let’s start with some background. We were working with a client to help them understand how effectively they were using labor hours in their Distribution Center. This organization’s employees are paid hourly and considered full time.

Similar to most clients, they were tracking Regular, OT and Absence hours by employee. Looking at the data summarized by department across time in the chart below showed nothing unusual. In fact it looks like they have increasing amounts of overtime when they use more overall hours. This is a good method for flexing labor to meet demand. (For a great article on the effective use of OT, check out this The Overtime Lie by John Frehse at Core Practice Partners)

hours by week of year

It appears that they are scheduling effectively too. The chart below shows that in general they are working the hours that are scheduled by department.

scheduled and worked hours

At the summary level many of the details are hidden, It's only when we get down to the individual employee level that it becomes apparent that there is room for improvement. This is the hardest part of deriving insights from data...organizing the data to start at a summary level and then progressively drilling down towards the right amount of detail to discover something of value.

employee hours

Let me explain the chart above…the x axis shows the number of hours worked in one week.  The y axis shows how many of those hours are overtime hours. Each dot represents the hours worked by an individual employee. The color of the dot represents the range of hours that employee was absent during the week. For example a green dot means all the hours were worked and there was no absence. A yellow dot means that while a person was scheduled for say 30 hours. They were absent between five and ten hours and worked the rest. The legend on the right of the chart shows the complete gradient.

We saw two distinct situations in this data. The first is that for this particular week charted there is a higher rate of absenteeism compared to others. It wasn’t a holiday week or something else that would drive absenteeism. By looking at this chart week by week it became clear that on busy weeks there was higher absenteeism than in slower weeks.

Secondly, we immediately noticed that while everyone is considered full time, many people were not working 40 hours and yet there were many people working significant overtime. So while at a summary level, the OT percentage looked reasonable, the reality was that there was plenty of regular hours capacity to satisfy the demand without the use of overtime. This of course drives costs up, but additionally as we hear in the news frequently, many people are not getting as many hours as they would like and to them, this situation is likely frustrating.

Creative visualization can turn data that companies must collect into valuable insights for improved productivity and increased employee satisfaction. In this case eliminating half the overtime and replacing it with regular hours would have resulted in a ~$35k a week savings. Don’t let your data rest!

Gregg Gordon oversees the analytics, data science, and big data focused group at Kronos. He is the author of Lean Labor: A Survival Guide for Companies Facing Global Competition and Your Last Differentiator: Human Capital.

Published: Wednesday, September 5, 2018