This exercise helps you choose experiments that create fast learning.
Duration: Long exercise
With a long exercise you can dive deep into a topic. They typically take 30-90 minutes. We highly recommend to prepare these exercises well upfront.
When to use this exercise?
For innovation, experiments are often evaluated based on impact and effort. This can help you make choices, but small ideas at an early stage are undervalued or large projects that can hardly be called experiments get a high impact, but with the under recognized disadvantage that they take a lot of time. To counter these 2 flaws, in this tool, we will discuss experiments differently. If you want to learn quickly from experiments, you should first evaluate all your ideas (including the raw ones) based on the time it takes to learn from them.
How does it work?
Step 1: Brainstorm all the ideas you have.
Set a topic for the experiments you want to run. Is it just about something your team is concerned about or is it about a specific topic or direction? Then make a long list of ideas you have on this topic. The success of this step is to be specific in your scope.
Step 2: Define an experiment for each idea.
For each of your ideas, what is a small experiment you could do to test your idea and find out if your idea really works or is valuable? Assess how long it will take you to get these insights and assess how much effort (in time and/or money) you need to put into it. With that assessment, place your ideas on the experimentation matrix (see image and attached template).
Step 3: Make experiments smaller.
You can skip this step. But it can also be the most valuable step. Use all your brainpower, ingenuity and creativity to make the experiments smaller. What’s the simplest thing you can do to find out if your idea is valuable? Of course, there is a trade-off when you reduce effort and increase speed that can also reduce learning insights. So, this is the most important trade-off you need to make. Adjust the position of the ideas on the matrix with these smaller experiments.
Step 4: Choose the most impactful ideas and start experiments.
Now you have a good overview of experiments that you could start and that ensures fast learning. We recommend starting small. It’s better to pick a few experiments and really learn from those, than to pick too many and still get overwhelmed by the amount of extra work you’ve initiated (and can’t finish, let alone learn from). Based on the horizontal axis of your matrix, you also have a good insight into when you can expect results. Schedule an immediate team meeting to review your lessons.
Have fun running these experiments!
