Quantifying My Phd: Pomodoros And Productivity
How much work does it take to get a PhD? How do marathon work sessions affect future productivity? Maggie Delano will answer these questions and more using data from tracking over 5000 pomodoros over the course of earning her PhD.
Quantifying My PhD Pomodoros and Productivity
I’m Maggie and I’m going to be talking about my quantified PhD.
So, some of you may know a PhD involves a lot of unstructured time, so you’re hey okay, five or so years produce an original piece of research, go. And that’s kind of a lot of the direction you have, so I decided to track how much I was working to help keep myself motivated and making progress towards the eventual PhD.
So, I tracked my work sessions over the course of about five years, so I tracked 5,000 or so 25-minute work sessions of my coincidences using Beeminder which you already heard a little bit before. So, this is my Beeminder draft from about the past five years of tracking Pomodoro’s, so I thought a little bit about how I actually did that and what I learned.
So, how I did it. I found this Chrome extension called Strict Workflow that allows you to implement the Pomodoro method. So, 25 minutes work, five minutes break and it blocks distracting websites for you during that time.
I modified that extension so that I could actually have a little box where I could type in what I was doing during that time and automatically log the data to Beeminder for myself. And then I also used Beeminder to keep myself on track. So make sure I was doing a certain number of Pomodoro’s on a given day, in a given week.
Here are some of the things that I found. So I broke up my Pomodoros by category and I was doing my research PhD in electrical engineering building medical devices. So I broke that down into categories like research was acquiring my data. Hardware was building the devices I needed and programming for example was programming stuff like the MATLAB.
One of the things I was surprised by that I actually spent a lot of time on writing, so it was the largest individual category, even although only a fraction of that writing was actually ultimately writing the thesis, which is the main thing you have to produce at the end. So, it turned out for me, communicating my results, even to just my supervisors was challenging and also really important and time-consuming.
One of the things I prided myself on, and one of the reasons why I wanted to use the Pomodoro technique was because I wanted to work less and accomplish more in the same amount of time. So I try really hard to never work on a weekend.
And what I did find is that I did occasionally work on the weekends, but I worked way less on the weekends than I worked during the week. And I worked only Sundays than Saturdays. I kind of called Saturday as a day where I always tried to avoid doing work.
One of the things that did surprise me is that when you look at the mean number of Pomodoros that I did for each given day that when I did work on a Saturday like shit must have hit the fan. Because I worked a lot more and I worked almost as much as I did during the weeks from different Pomodoros, although there was a huge amount of variation for each individual day.
So, here we can see the number of Pomodoros per week that I did over the course of this entire experiment. And so, at the beginning there are a number of different breaks, so I was not as reliable about tracking right at the beginning of the experiment and I’m not really sure exactly why that was.
With one of them I think I had a major research deadline, where I was working way too much. So I think the most Pomodoros during that time was 26, so that’s 13 hours’ worth of very focused work and I do not recommend that. And so, after one of the major research deadlines was, I saw that I had a break after that, so I think that was probably burnout.
One of the later breaks was actually I had a concussion, and I talked about recovering from that concussion at the QS conference three years ago. Then after I was recovering from the concussion I started tracking much more consistently and that when I actually formally integrated strict workflow with Beeminder. So then for the last three years of my PhD, I was tracking very consistently, so I have data from all that time and then at the end. So in January I actually defended my thesis.
One of the things that you can kind of see from this graph and that I had noticed even before I analyzed this data was that I felt like I would work a lot and then kind of burnout, and then work a lot and burnout.
So, if we look at the four-week moving average, we can see that there is a lot of these like peaks and valleys as I’m working, but they are all around like the ultimate mean amount that I pledged to work on Beeminder, which was about 25 Pomodoros per week.
And one of the things that I was interested in is okay, these times when I was working a lot more, how much did that affect how much I was working the subsequent week?
So, I calculated the standard deviation, and so if for example you look at the weeks where I worked more than two standard deviations of the mean amount of work that I pledged to work there was a huge drop-off. So, it was really not a sustainable amount for me to work. So like the major research deadline in 2013, and then the thesis defense partway at the end it was very peaky, so I couldn’t really maintain that amount or work. Or just I wasn’t motivated to be if Beeminder didn’t tell me to.
And one of the last things that I was kind of interested in was you know, what is the ideal rate to say that I’m going to commit to working, because a PhD is there’s no one really telling you how much that you have to work.
And so, in the end I did work the amount that I said I would work, which is what we would expect to see, but deciding how much that should be was kind of a challenge.
With Beeminder if you are on track, you are following this little yellow brick road, and you can’t even see the road for most of my goal, except at the very end when you can see I went way above it because I was working so much to try to finish.
And so, this kind of suggests to me that my goal was probably around the right number of Pomodoro’s that I could reasonably sustainably do over an extended period of time. Although I might have been able to push myself slightly but in the end no matter what my goal was I was just working as hard as possible to actually finish at the end.
I also wanted to talk a little bit about some qualitative observations. So for me, Pomodoros plus Beeminder were a very good combination because it provided an accommodation of structure, efficiency and accountability, which for an unstructured period like the PhD process was very useful for me.
I also found that the frequent breaks were very helpful, so and maybe also the walking like we heard about earlier. So, during my Pomodoro breaks I often got up and walked around, and maybe go to the bathroom and I would always have my best ideas like back from the bathroom on my Pomodoro breaks. So, I found that being broken up was really helpful for me.
I also found that it was really hard for me to write efficiently for more than four to six Pomodoros per day, and a lot of really intense knowledge tasks I felt it was hard to do for more than that period of time. So, a lot of people talk about being in our lab for 80 hours a week, and I found for me that wasn’t really sustainable or even necessary really desirable. And that even working about five Pomodoros a day, like the hardest things that I needed to get done was enough to get me through to the end.
I also felt definitely that certain types of work lended themselves better to using Pomodoros, but ultimately, I found it to be so useful to have that structure, that even when I wasn’t actually formally doing a Pomodoro I still logged it in terms of the number of Pomodoros, because I found keeping track of it in that way just worked way better for me.
So, just to summarize, quantified PhD for me was five and a half years, 4840 Pomodoros that I tracked. I did derail. I paid Beeminder $165, and just found that goal any way it was worth it. And then in the end I did get one PhD.