How Do Professors Spend Their Time? A Personal Perspective

A Ph.D. student in our program recently conveyed a common impression of academic life:

The parameter “required effort to be above average in my profession” has annoyingly high values for tenure-track jobs. Committing to this level of effort for decades seems to be a major deterrent [for students going into academia].

This is definitely a concern for me, too. Maintaining a healthy work-life balance is important to me, most critically because I have a family, but also for health (I’m a runner) and sanity (I have non-academic hobbies) reasons. Equally important, for me, is managing the time I spend at work, both so that I can be effective in what I do, and so I can spend most of my time doing the rewarding things I came to academia for: teaching, research, and working with students.

Fortunately, you don’t have to spend 60-80 hours a week to succeed as a professor; for example Radhika Nagpal at Harvard wrote a great article on the approach she takes to achieve a good work-life balance, spending 50 hours/week or less on the job. My goals are similar to hers (though my working style and “rules” aren’t exactly the same), and I’ve been tracking my time since March 2015 to help me manage the time I spend better. The resulting data might be interesting to anyone wondering how a professor’s 50-hour work week might be spent.

Here’s a summary of my time in the last academic year, August 2016-July 2017:

I spent about 2233 hours working, which is about 49 hours per nominal work week, leaving reasonable time for vacations (there are about 45.5 nominal work weeks in the year if you take 52 weeks and subtract the average American worker’s vacation, holiday, and sick days). I broke this down as follows:

26% Advising
15% Teaching
15% Raising money and managing funded projects
13% Professional (e.g. professional meetings, reading papers that I am not reviewing)
12% Service to Carnegie Mellon, my university
11% Service to the external community
6% Wyvern (my “keystone” project if you will - this is time not spent specifically advising)
2% Other (e.g. organization, non-advising/non-Wyvern research)

The above is reasonably close to “time on task” - e.g. if I spend 15 minutes reading a news article about some crazy politician’s antics, I don’t include that time in my records. If I have lunch with a colleague, I leave 30 minutes blank (eating and socializing) and count the rest as “professional” time (catching up on research, etc.) On the other hand, I don’t try for stopwatch-level precision; I fill in 15-minute boxes on an online app, Bubbletimer, based on what I am doing.

What do I personally conclude from this data?

I work hard, but it isn’t crazy. I have time for vacations and family. I would like to work a bit less in the average week, but in weeks when it doesn’t impinge on my personal life, I enjoy my job enough that I don’t mind working a bit beyond the “standard” 40 hours. Radhika comes across as a reasonably well-adjusted person in her article, and I’m close to where she is. Lots of people in the software industry work more.
The teaching number is low. This past year that is because I taught classes where it was possible to deliver a course in less time (seminar-like courses and courses I’ve taught before), but that won’t be the case every year. Thus, this percentage will likely be higher in other years. Even beyond that, I’d like to spend more time teaching so as to provide a better learning experience to students, and also to invest in some longer-term projects such as converting my program analysis course notes into a textbook.
I think a lot of faculty (maybe especially at places like CMU?) spend even more than 15% of their time chasing funding, but my current allocation (which works out to over 7 hours/week) is too much for me. The main reason is that I find many parts of raising money to be misery-inducing, and I need to reduce the time spent on it to maintain my sanity. Of course, I have to raise some money if I want to advise students, which is one of my favorite activities. I’m working on being more efficient with fundraising, choosing funding sources that trigger less misery, and moving gradually towards a modestly smaller research group in which I will invest more of my time per student (which I would like to do for other reasons anyway!)
The service numbers are perhaps a bit high, but not too far off. This year, I organized the OOPSLA PC, which was a big community service task. I probably won’t run another top-tier PC for a few years, but I will still need to do my bit for the community in other ways. Closer to home, I run the Software Engineering Ph.D. program at CMU, which is also a relatively large commitment. Despite the workload, I’m quite happy to keep doing this, as I find it highly personally rewarding–more than average for service tasks, and again, of course, we all need to do our part.

One of the lessons I’ve learned is to be strategic about my time, something I definitely wasn’t doing when I first started as a professor. Tracking my time helps with that, even though there is some overhead to it, because it reminds me to think about what I am doing. It gives me a good reason to say “no” to good (but optional) things in areas where I am already spending too much time. It also encourages me to prioritize: maximizing long-term impact and fun (for research, education, or service) or money (for grants) while minimizing time spent and “pain factor.” As a result, I think I am actually working a bit less than I did early in my career (though I wasn’t tracking my time then so it’s hard to be sure) but I think I’m having at least as much impact as I did then, mostly because I’m working smarter.

Overall, although it’s not perfect, I’m reasonably happy with the way I’m spending my time, which is actually a bit better than it was the year before (I can dig up those numbers if anyone’s interested). What I’d really like to do is divide a nominal 40-hour workweek into about 35% research/advising, 25% teaching, 20% service, 10% professional growth, and 10% funding–then in weeks where it’s convenient, spend a few additional hours each week on “extra” personally rewarding projects (generally in the “Professional” and “Wyvern” categories above). I’d also like to pack the hours I do spend more efficiently, to enable me to spend more time with my family. Of course it’s hard to truly optimize given all the constraints in a busy career. We’ll see how the coming year goes! :)

Getting back to the original issue raised by our graduate student…careers take lots of different forms, and your mileage may vary, but I think academia has a lot going for it if you manage things the right way. I haven’t completely nailed the “right way” for me, but I think I’m getting closer to it. The two best things about my job are working with students and getting to work on the problems I find most interesting. To get to do that and still have a reasonably scoped job is pretty cool!

Edited slightly 12/29, mostly to reflect that although I am relatively happy with my workload, I’d probably work a bit less in an ideal world. So still some things to manage better!