From Chaos to Order: Using Scrum in my PhD Research


February 16, 2023

PhD is a four-year-long journey into the unknown. And it's a journey taken alone. Don't get me wrong: science needs collaboration to thrive. And I have amazing colleagues who have helped me so many times by brainstorming ideas, solving experimental difficulties, and giving life advice. But after all, my PhD research is my responsibility. This means that by the time I finish my PhD, I will become the person who knows the most about a couple of particular experimental assays to study synthetic cells' mechanics and genetic information processing.   

In other words, it's up to me to find the way to manage the multitude of tasks I face on the daily basis. These, among others include: doing multistep experiments, conducting biophysics seminars, presenting at international conferences, building a professional network in my field, writing Python scripts to process gigabytes of data and learning numerous safety protocols for working with hazardous substances. It did become overwhelming very quickly after I've started my PhD. Therefore, I had to come up with some way of tracking my progress. 
The most basic kanban board [1]
This necessity and my love for keeping things neatly organized led me to two agile frameworks. The first one is called kanban and relies on visualizing the tasks. The second one is scrum, which centers around workflow division into sprints of fixed length. And crucially, both require continuous communication and openness between the team members. If you work in a company, you likely use them to track progress and assign team tasks. 
Not so basic kanban board. The kanban framework is very flexible and can be adapted for complex projects with multiple stages and teams involved [2]
Unlike, let's say, in software development, agile-based methods such as kanban or scrum are not commonly used in academia. The main reasons for that are the solitary nature of many tasks, much less interdependency between people's projects, and the longer duration of the scientific projects. Here came the major obstacle. All the factors I've listed above are also valid for me. Fortunately, the kanban framework is very flexible. Thus, I have adapted it for my PhD and incorporated the sprint-centric aspects of scrum. How did I do that?

To begin with, I have borrowed the idea of breaking my PhD down into shorter intervals from scrum. Instead of planning for my whole four-year-long PhD (marathon), I have divided it into segments:
  • Two weeks (sprint) - what are the things to do now?
  • Quarters (middle-distance run) - where should I stir my research?
This way, I can finish the things I need to do now and keep the bigger picture of my PhD in mind. 
Me after running physically, though still in the process of running a metaphorical PhD marathon
I regularly discuss with my PhD supervisor the goals for the quarters. In the short two-week sprints, however, I organize and plan everything independently. This takes up around an hour a week and includes:
  • sprint planning  (15 min) at the beginning of it to lay down my aims and plans for the upcoming two weeks; 
  • daily sprint reviews (5 min daily) to see what my priorities are and which things I should focus on today ;  
  • mid-sprint review  (10 min) at the end of the first week/beginning of the second one to see how realistic my sprint plans were and how should I approach the second half of the sprint; and the 
  • sprint wrap-up (10 mins), where I look back at my progress and note down things to do during the next sprint
The kanban board in Trello I use to track my progress, projects and activities.
Furthermore, I have created a digital kanban board (Trello). This took some refining, but I settled up on the following structure/columns left-to-right:
  • Sprint - what am I planning to do in the current sprint?
  • Input needed - what do I need advice/external input on?
  • Waiting - what am I waiting for (delivery)?
  • In progress - what am I working on today?
  • Done - what have I already done during this sprint?
  • Previously done - what have I done before (archive of past cards)?
  • Backlog - what tasks should be done, but not urgently/necessarily?
For each task, I make a card. These vary from "Registering for an Annual Biophysical society meeting" to "Measure ... DNA with ... and ..." and from "Correct BSc exams" to "Do the lab cleaning duty". The tricky part is since I am a one-person band, these tasks have wildly different profiles. I use a long list of labels to distinguish and better prioritize them easily. 

These labels have different colors, depending on what they designate. Some colors are used for a large group of labels: such as the experimental setups I will need to use for a particular experiment (a set of pink labels) or the person I should discuss/do it with (a bunch of blue labels with names hidden for privacy reasons). Others are more unique, like the red one, which explicitly marks everything teaching-related, or the orange one, which concerns the Building a Synthetic Cell (BaSyC) consortium [3] and my project as a part of it.
Some of the color-coded labels I use to track my tasks.
Let's look at a few cards on my board. The first task below is an experiment I will do for my synthetic cell project (smIVTT) on a setup called F-Trap. The second card refers to the upcoming BaSyC consortium symposia I will attend at Radboud University (Nijmegen). Finally, I am one of the retreat organizers for my and a few other labs in the department [4]. My responsibility there is to find fun group activities to do on Texel.
A few examples of cards on my kanban board
It's been a year since I adapted this agile-inspired combination of short sprints and visualization of tasks for my PhD. And it has been incredibly helpful in tracking my research progress, prioritizing experiments to do next, and planning the upcoming weeks. Plus, seeing the board empty at the end of the sprint is a very satisfying feeling. PhDs often feel stuck in place even though we work very hard. Tracking progress like that gave me an accurate picture of my progress and helped defeat the imposter syndrome. 

Thank you for reading!
My kanban board at the end of a very productive sprint
References:
[1] - https://kanbanzone.com/
[2] - Boeg, Jasper (February 2012). "Priming Kanban". InfoQ. Retrieved 17 February 2014.
[3] - https://www.basyc.nl/
[4] - https://vu.nl/en/about-vu/more-about/physics-of-living-systems

Share

Tools
Translate to