Improving estimates using past performance
In my previous post I discussed why we should act more like Superforecasters when estimating. I suggested that we should leverage our historical performance and communicate estimates as a range and confidence level instead of a simple delivery date. Here…
