This idea always struck me as fairly misguided. It of course makes sense to try to muster evidence and data when making decisions but it's not some sort of panacea. The book The Tyranny of Metrics (https://www.amazon.com/Tyranny-Metrics-Jerry-Z-Muller/dp/069...) goes into the successes and failures of various attempts to use data for decision making in some detail. Found it to be an interesting read.

In terms of public policy you have to decide what you're optimizing for and that decision can't be made with data alone because it does not help resolve questions of value and fairness.

This topic was discussed in the last EconTalk [1] podcast featuring the author of a recent book “The Tyranny of Metrics” [2]. I think one of the author’s interest insights is the distinction between metrics for diagnostics and those for incentives.

1. http://www.econtalk.org/archives/2018/04/_what_is_the_ap.htm...

2. https://www.amazon.com/Tyranny-Metrics-Jerry-Z-Muller-ebook/...