The Data Scientist: The Afternoon Slump, Part 2

The Data Scientist: The Afternoon Slump, Part 2

Cha Li LUMO Data Scientist

Cha Li
LUMO Data Scientist


This is a follow-up to my earlier blog post examining the “afternoon slump.”

I know that it has been a while since my last big data post, but we’ve been hard at work here at Lumo Headquarters digging into our user data and closely examining posture patterns.

This time around I’m going to focus on average posture during the workday (defined as 7am to 8pm) across the 4 months from June to September of this year, drawn from several hundred active users.

The following graph displays what the average posture score is on an hourly basis during a typical work day of a given month. There’s a lot going on in the graph so I’ll try to explain each piece concisely.


As a refresher, the “posture score” is the time spent in good posture divided by time spent in good and bad posture. Time is measured using a 24-hour clock: 7:00 to 20:00 for 7am to 8pm. Each line represents the typical workday for a single month, and the numbers above the September line show how much the average posture score for a given hour of the workday in September have improved compared to that same hour in June.

I’ll begin with the most noticeable change: posture score, when aggregated across users, increases steadily month to month across all workday hours. The core work hours (9am to 3pm) show, on average, a 16% overall improvement in sitting posture. The second change I want to point out is that the downward trend of the dreaded “afternoon slump”, which begins around noon, becomes less severe with each passing month.

I have one more graph to show you, which simply includes standing posture as well as sitting posture. The two above trends I discussed can also be seen in the following combined posture graph.


One big difference is that the overall posture score is noticeably more consistent during the early morning hours compared to the sitting posture score. My hypothesis is that people tend to stand and walk around more during these hours as they prepare for their day and make their way to work; many people tend to have better standing posture than seated posture.

In conclusion, these two trends show amazing improvement among our active users and demonstrate how regular use of LumoBack can help you become more aware of your posture and allow you to make substantial improvements. I’m excited to have discovered these trends while playing with our data, and I want to congratulate everyone on their improvements!

For our long-time users, do you think that your posture has improved over the past few weeks? Months?  What about other metrics? Are you walking more? Sitting less? Let us know in the comments below!

Share on Pinterest
There are no images.
Share with your friends


About Tansy

Tansy wears research, public relations and partnership hats at Lumo BodyTech. She is passionate about harnessing technology to support personal behavior change. Tansy’s background is in wearable tech and social entrepreneurship. She graduated from the University of Colorado with a BA in Social Marketing.

1 Comment

  • Pete Olds
    Pete Olds

    Cha, fwiw, I am noticing a direct correlation between local weather and posture. Are you seeing this kind of correlation in the larger data set?


Leave a Comment