This post is going to talk about incremental learning and the process of setting hypotheses ahead of time to test subsequent outcomes.
The idea was sparked in a conversation with my buddy Rahul over coffee about learning. In particular:
How can we best learn from our experiences?
The scenario proceeds like this we’re two graduate students with limited time who are in the process of setting up a number of entrepreneurial ventures (exambootcamp.me and the GW hackathon). We want to optimize our time-investment in a natural way so as no to be disrupt our studies. Over the past couple of months we’ve tried a number of different options, including negative disincentives, prioritizing tasks and strict schedules. All contribute to our continued success yet one key element is (was) missing, learning. In particular learning from the implementations of these ideas. Rahul suggested that while we’re getting more things done we we’re not necessarily completing them efficiently. He developed a simple rule, borrowed from the “The Lean Start-up” by Eric Reis, that we’ve begun to apply:
It start’s with the premise that there is a desired end goal and an established way to proceed towards this goal, let’s call this our null action $N_0$. To possibly improve on $N_0$ we tweak our current method by hypothesizing the slightly modified alternative action $A$ may be more effective. Subsequent to implementation of $A$ we “test” against a defined measure to see whether the desired outcome was achieved more effectively. If so we have learned that action $A$ may possibly constitute a better method for approaching a given process. Continued implementation of this incremental tweaking ultimately leads to a more refined sequence for executing a process without disruptive changes that would result from implementing a large (rather than gradual) change.
The idea isn’t novel and has been successfully implemented in the car industry (see the Toyota Way be Jeffrey Liker). Where our approach differs is in it’s application to personal goals, social interactions and work which can all be tested. What remains to be seen is whether the integration of such a methodology will become a habit in our daily lives and whether realized gains offset initial time investments. Particularly in the social arena where setting hypotheses and learning from our mistakes don’t come as naturally (well at least compared with work).
For those interested a nice article which goes into more detail from a product development point of view is presented by Avinash Kaushik in his blog which can be found here.