The Baconian method… What’s that got to do with projects?

Remember back to high school science for a moment, remember breaking an experiment report down into sections? That was primarily based on the Baconian scientific method! Lord Francis Bacon was an early thinker in the need of more evidence to support our understanding of the world and ensuring our inherent biases were not overly flavouring our thinking. He was like a modern Bent Flyvberg (Nobel winner for his work on risk in economics and financial estimation).

A hypothesis, a method, then carefully doing the experiment (or making many mistakes trying), taking fastidious notes (or not if you were me) and trying to turn them into cohesive results, drawing conclusions (likely that you needed to be more careful or take better notes… or is that just me again).

This kind of validated learning is exactly how we should start and deliver our change initiatives. As many experiments, feedback loops, learning moments, etc., as we can responsibly fit into the time and therefore budget available to us.

Hypothesis: the thing we think the customer wants based on previous experiments, personas, mandates, or whatever other data we can bring to the table. It should be based on empirical estimates, with a desired set of results communicated and understood by relevant stakeholders. Add a budget and some benefit realisation and voila! Business case!

Method: the plan. Be it to scrumban our way through one or more iterations or to outsource key components and let the suppliers pick their own delivery method, right through to (dare I say it) the rarely needed (overly?) detailed planning and highly (over?) controlled waterfall (eg. of a large procurement). A conscious choice about delivery method is required to ensure the method is appropriate to the hypothesis.

Conduct the experiment: deliver against your plan… sometimes you’ll need to swap out equipment/resources, sometimes you’ll change the planned sequence a little, these minutiae are inevitable. One thing is certain though, not having a plan IS planning to fail, without a baseline change is just chaos.

Results: It’s critical to science for your results to be validated, or at the very least validate-able. You’ve got to be able to trace where the money went, how much time was spent, whether the anticipated results were achieved (delivery of your business case) in order to learn how to do it even better next time.

Conclusions: This is the building block of the next experiment, the legacy of empiricism, the lessons you’re paying forward…. You have an agile responsibility to do it. It’s in the manifesto, it’s in lean startup, it’s in scrum… it’s also in PRINCE2. It’s best practice for a reason.

Cliché as it may be, I want to finish on a challenge… apply this thinking to your current project/change initiative. You’re “halfway through” now and it looks like you won’t be able to successfully complete the experiment on time/to budget/pleasing all your stakeholders, what do you do next? Rush your way through it and hope to not contaminate the outputs? Throw more learners at it in the hope they won’t trip over each other learning the method and hypothesis before getting into the experiment? Ask the teacher for more time and finish over lunch? Many times the best thing to do is save your remaining resources, reset the experiment, learn early rather than throw it all in on a failure, and try again. 

Image source: https://en.wikipedia.org/wiki/Margaret_Hamilton_(scientist)