I’ve often wondered whether how hard people work impacts their overall performance. Do people who work longer hours do better? How about people who start work really early or work lots of overtime and weekends? It’s common for social pressure to incentivize this kind of behavior but does it pay off in results?
Worklytics Dataset Builder provides us with the opportunity to answer some of these questions. By analyzing data in the productivity tools people use, Worklytics is able to provide objective insights on how people work. These include:
- What time people tend to start and end work
- How long people tend to work
- How much overtime and weekend work people do
- How intensively people work
- How much people collaborate with others
- What type of work they do (meetings, documentation, code etc)
We compared this data with the results of quarterly performance reviews for 217 people, to look for any patterns. Here’s are some of the interesting relationships we found:
Working long hours does not lead to better performance –The total amount of time spent working seemed to have little impact on performance scores as did the amount of overtime worked.
Working on lots of weekends had a slightly negative impact on performance scores, as did regularly working late evenings.
Doing more with your time – Interestingly, one of the strongest predictors of performance was work intensity, measured as (amount of work done)/(time spent working). People who did more with their time at work tended to score better. We also found that people who were more active (email, documents, code etc) and produced more output overall tended to score slightly better on their performance reviews.
Scatter plot of daily hours worked vs performance ratings
Early to rise and early to bed… – A tendency to start work early in the morning also had a positive impact on performance scores whereas, as stated above, working later did not!
Wasting time – Some of the strongest predictors of low performance were large amounts of time spent in meetings and a high total number of meetings attended.
A word of warning! –As always, one needs to be aware of a significant risk of various confounding factors. For example, I did not take into consideration people’s roles or seniority within the organization. It may be that manager meet more often and tend to score lower on reviews overall! It’s always important to consider the full context when doing this type of analysis.
Another key point to consider is that performance review scores are often a somewhat subjective measure of performance. They can be heavily influenced by how managers and peers perceive a person has performed.
It was interesting to see some early indications that merely working longer hours does not mean you’ll make the top performer. What really matters is what people do with their time at work! I think this merits a more in-depth analysis with a larger sample set. We’ll working with a few volunteer organizations to gather the required (anonymized) data. Please contact us if you’re interested in participating.