A head of an innovation unit recently wrote to me, asking why do I include setting a time boundary as part of my lean experimentation guide.
Here's an excerpt from my answer, so everyone can benefit.
Time-boxing your lean experiments is beneficial for a multitude of reasons:
It gives you a specific date on which you have to stop, process the data, and make a decision.
You are almost always better off by running multiple small experiments, instead of a few big ones.
It challenges you to consider activities that will take less time and effort to generate the same learning (since you define the learning goal in the first step).
It gives you focus and cadence. Depending on the type of the experiment, you will either be fully focused on conducting it, or you will be waiting for results. In the former case you have a clear set of expectations, and in the latter you have peace of mind while doing daily tasks.
It provides a feeling of satisfaction, progress, and closure. You should always be learning, testing, and adjusting. But this one experiment is done and closed. Celebrate and move on.
It reduces the likelihood of confounding. The longer the experiment runs, the more likely are you to introduce factors that will make data less reliable and trustworthy.
Bottom-line, time-boxing allows you to achieve your learning goals faster.
Don't hesitate to reach out if you have any questions related to lean experimentation, innovation, or strategy.