We Should Measure the Uncertainty in our Product Metrics

How do you know whether you’ve predicted an outcome correctly? In some cases – for example, a political race – the answer is straightforward. Either the outcome happened or it didn’t. In others, the answer is less straightforward. Often this happens because you can’t observe all of the data that you’d need in order to get a definitive answer. I currently work as a product manager for a tech organization, and in that capacity I ask myself this kind of question a lot.