Learn as you go - A scientific method

Are you stuck in defined industrial processes, lined up in dependencies while working on an undetermined goal? Do you need to try out what works and what does not because your goals are too fuzzy? Do you need to learn as you go to prove your theory with data?

Then you might try a scientific method.

Observations in science. By Marretao22 (Own work) [Public domain], via Wikimedia Commons

In science you do experiments. This means you observe something, then you make a guess about it and formulate a hypothesis. To prove your ideas you experiment and see if you can validate your ideas - you gather data: hits and no-hits. If it sticks, you keep it - if not you formulate another hypothesis, experiment and observe the result again. It is like: "Oh, the apple falls to the ground. If I toss a stone, it will fall to the ground. Toss the stone. Yes, stone falls to ground."

So a scientific method works like this: Observation, hypothesis, experiment and back to observation - closing the cycle in iterations. 

Observation

Observation is the start and endpoint of any experiment. You gather insights toward your work and the decisions which were made from the data you have collected. You measure, interview and study the work you are doing or the experiment you have done.

Hypothesis

Once you observed the environment you want to improve you assume what might be the next right step to follow. Formulate it and set it as a goal. This kind of goals are rather fuzzy than predictable. Your goal needs to be proved. So you stop talking and start doing.

Experiment

Experiments are about exploring your ideas. Nothing is perfect from the start, so you have to gather insights. You start with an idea of what the world could be and check out  how wrong and right you are. Important insights are discovered after the start. Exploring is like: Build, break it, build it again and tweak it - that's the way solutions are discovered.

Iterations

Use small iterations to experiment different hypothesis. Adjust your hypothesis as soon as you got real data and reiterate to your goal. Go for fast feedback from your customers and users to generate insights. 

If you are looking for this kind of approach: Agile, e. g. Scrum uses this scientific approach.

SVNWNK