I had lunch the other day with Johanna Rothman and the topic of planning research work came up. It is difficult to plan research work because the very nature of research is one of iteration and uncertainty. You don't know if your experiment is going to work, so how can you build a formal plan of everything you plan to do?
Johanna pointed me to her article on inch-pebbles, How to Use Inch-Pebbles When You Think You Can't, and specifically highlighted this paragraph on using inch-pebbles in research:
When you do research, as my colleague Sally does, it's sometimes difficult to know when you're researching or going off into never-never land. Sally plans her work, by using a set of questions to know when she's discovered what she needs to know. Her questions are specific and detailed, so in a sense they work like inch-pebbles, but because she doesn't know where she's going, she does not create a specific plan to get there. When it's time to transition the research out of her lab, she creates a project plan and works with a project manager to create inch-pebbles for a successful transition.
What kinds of questions are useful in this context? I am not completely sure, but I have some inklings:
- What is the goal of this phase of research? (How will I know when the research is considered done?)
- What information (experiments, outside research, etc) will tell me to stop? (Stop pursuing this line of research)
- What information do I need to make it through the current phase?
- What experiments do I need to conduct?
- What are the key variables (today) that I need to explore? What experiments will give me the data I need?
The other element that I haven't fully wrapped my head around is associated with a Theory of Constraints view on the research work. I think the concept of "planned load" that primarily comes from discussions of S-DBR (such as that in Supply Chain Management at Warp Speed, my review). If I know the key constraint in my research organization (the lab, equipment, person that limits the flow of work through research), is it possible to look at the planned load on that resource and work out the flow of work that comes across it?
I think it is possible. Planned load requires knowing the capacity of that resource AND the work that I want to send across the resource. This means that I know, in general, what types of experiments I will do in the next time horizon and how much capacity those experiments require. The time horizon is generally the lead time of experiments (how much time from initiation to completion of the experiments). The time horizon does not need to be months or years (unless you have very long experiments), just enough to look forward to see when there will be capacity to run your experiments.
I'd love to hear other thoughts on this topic.