Data scientist at Port Jackson Partners in Sydney, Australia. My PhD was in computational biology. In my spare time I write about medical research at BioSky.co.CVAbout
I’d like to address one point of contention some people have had with my previous article about finishing projects – what to do when it isn’t worth finishing. I think the important thing to focus on here is the minimal viable product (MVP) that you can tangibly show as a result of having worked on something.
Sometimes we reach a stage where we realise that a project is no longer a logical use of our time. Maybe you’re working on a startup and realise there is no market for your product, or maybe new opportunities arise and it makes sense to redirect your time to a different project.
In cases like these, I think it’s fair to re-evaluate what ‘finished’ looks like. For some of your projects it may mean releasing an MVP and open sourcing some code, for others it may mean pivoting and salvaging some of your work in the new project.
Let me give you an example. I have dedicated a significant amount of my spare time to building an app. I started this project with lofty goals, but over time I’ve come to question whether this app may be worth the time. Yet, if I give up now with 80% of the product done, then I might as well have never worked on the project.
I am now much more focussed on the objective of developing a tangible MVP that I can release. Maybe it will take off, maybe it will not, but I will treat the MVP as an experiment, and while I will give it a shot, I will also re-evaluate my goals for the project based on what the data is telling me is logical.
Lessons from Research
In research, we conduct experiments to learn new things and answer questions. If we find something new or interesting, that will frequently lead to new questions and new experiments. However, if an experiment does not find anything novel, at some stage the scientist will say:
Ok, we did the experiment and have the results, but it does not make sense to continue despite our original hope.
That is fine. In fact, through my years in research, I have come to realise just how often an experiment doesn’t achieve the “desired” result.
But what I am against is spending months setting up an experiment, getting 80% of the way there, and then moving on to a new experiment before you actually find anything out! Yes, you probably gained some new skills along the way as you set up the experiment, but if you never actually run the experiment, what was the point?
On the other hand, if you’ve run the experiment and gotten the results, then the data you’ve gathered might actually tell you something useful in setting up your next experiment. Even if it’s not your desired outcome, at this point you could use this MVP to re-evaluate where to go next.