Q&A with John Willis on Burnout in the Software Industry
"Did anyone notice that PHD in psychology I had up there?" John Willis, Director of Ecosystem Development for Docker, asked at QCon London 2016. "No. That’s because I don’t have one." Willis preceded to give a highly personal, and at times emotional presentation on burnout.
Burnout was coined by German-American psychologist Herbert Freudenberger in 1974. The Japanese have a word for burnout that leads to death - "karoshi." When suicide is involved they call it "karojisatsu".
Whilst it’s tempting to think of this as a largely Japanese problem over the past few years, there seems to have been an increase in suicides in the software side of the IT industry. With complexity and responsibility ever increasing in the creation and support of digital services industries, we have seen a correlated increase in stress levels that seems a plausible underling cause.
After the presentation InfoQ set down with John to talk more about the issue.
InfoQ: What made you decide to give a talk about burnout?
Willis: In early 2015 I learned at a conference about a colleague and friend, in the DevOps community, who committed suicide. It was pretty clear from his close friends and his twitter feed that he was experiencing clinical symptoms of burnout leading up to his death. I had dealt with other people in the past who had issues with suicide and this event created a watershed of emotions for me. So much so that I wrote the blog Karōjisatsu as a cathartic response. However, in the end it was the response to the blog post that made me realize what a serious issue this was in IT. There were over 1000 comments to the blog in less that 48 hours. I received over 200 emails of personal testaments of issues people in our industry where dealing with. This prompt me to do a Devopsdays NYC Keynote on burnout and I have done a fair amount of research on the issue since.
InfoQ: What are the possible effects of a burnout for a person?
Willis: Some researchers put burnout in the same clinical category of diseases like PTSD and Depression. In some cases the ultimate effect can be death; however, more common results are health related stress symptoms (stroke, heart attack). In Japan in the late 1980’s the government found work related burnout to be so prevalent that they classified it with it’s on unique name (Karōshi - death from overwork). Other work-life balance issues can arrive from unresolved burnout symptoms. For example, divorce, disconnectedness from family and friends. One of the more interesting finding about burnout is that on average it takes 6 months to get to a clinical burnout state; however it can take on average 2 years to recover. In many cases the issues never get resolved in that a some burnout victims wind up just leaving their industry all together.
InfoQ: Can you give some examples how burnouts are impacting the software development industry?
Willis: An industry leading test for burnout is called the MBI (Maslach Burnout Inventory). In 2013 the IT Security industry did a MBI of their community due to a high rate of suicides. The three primary indicators from the MBI are increased exhaustion, increased cynicism and decreased efficacy. The IT Sec Burnout survey was the first time, to my knowledge, that the IT industry had ever been surveyed for burnout. Although it was a small survey set the findings where that IT workers ranked very high (negative) in all three. Other research indicates that there are severe opportunity costs loses due to burnout due to decayed productivity. One of the most interesting research I found, was that typically the people most susceptible to burnout are the highest performers.
In general any organization that doesn’t deal with burnout can be impacted by the following:
- Increased Healthcare Costs
- Possible Lawsuits
- Turnover (losing key employees)
- Optics (related to the perception of future employees)
Hidden costs can also impact an organization:
- Missed Deadlines
- Missed Opportunities
- Missed Threats
- Decreased Innovation
InfoQ: What are the leading indicators of a potential burnout?
Willis: Christina Maslach, author of the MBI, is one of the leading researchers on occupational burnout. In her original research she codified the three aforementioned areas as burnout indicators (Exhaustion, Cynicism and Efficacy). Exhaustion is typically more obvious than Cynicism and Efficacy. Indications of clinical related burnout/cynicism typically look more like withdrawal (for example, I just want to do my job so please leave me alone). Efficacy is a little harder to identify. Efficacy typically is related to self worth or a perception of an organization not recognizing an employees worth. In all cases these are the indicators that should be dealt with with professional help. However, there are few well known MBI SaaS services that are reasonably low cost. I have actually run an MBI for myself and actually published the results in my Qcon presentation.
InfoQ: Do you have suggestions how such indicators can be used to prevent burnouts?
Willis: In general burnout is a clinical physiological disease and in all cases the end game should include professional advice. I think the MBI is a great hybrid tool that can help with self awareness.
InfoQ: You talked about possible mismatches between employees and the organization that they are working with or would like to work that can cause a burnout. Which are these mismatches?
Willis: In some of Christina Maslach’s later research she identified what she calls the Six Mismatches. What is interesting about this research is that it is not just a lens that focuses on the employee. This model uses these Six to look at the mismatches between the employee and organization. In other words, in some cases it might not be the employee or the origination. What she found is that it is the mismatch. For example, an organization by choice might be Machiavellian in nature; where as an employee might be opposed to this kind of ideal. In this case there might not be a bad guy or the bad guy is actually the system (i.e., the mismatch).
Here is a list of the Six mismatches:
- Work Overload
- Lack of Control
- Insufficient reward
- Breakdown of Community
- Absence of Fairness
- Conflicting Values
InfoQ: What’s your suggestion when there are one or more mismatches?
Willis: Some of the MBI SaaS services also offer what is called the Areas of Worklife Survey (AWS). This is a statistic survey that includes psychometric data based on the Six mismatches. Also Christinana Maslach, as you might imagine, has written a few books on the subject of burnout. However, in my humble opinion two books that have been very helpful for me over the years (although not directly related) are Feeling Good: The New Mood Therapy by David D Burns and Happiness by Matthieu Ricard.
InfoQ: If people want to learn more about burnout and how to prevent it, where can they go?
Almost all of my research links can be found in my presentation on burnout.