Metric obsession is just a special case of confirmation bias. It has the same antidote. Try to constantly and creatively disprove the hypothesis that the only

Metric obsession is just a special case of confirmation bias. It has the same antidote. Try to constantly and creatively disprove the hypothesis that the only thing that can move your metric is the thing you care about. Search for confounding variables, counter examples, etc. Every metric deck needs a healthy serving of humility and acknowledgement of uncertainty. It seems like the article was dancing around this idea without actually coming out and saying it. The suggestions are good, but I wish the authors had described the fundamental causal problem with metrics so that we can get a deeper intuition instead of a checklist.