This article is a very interesting read for everyone that works in a project. It deals with the perils of time estimation of a project and the underlying thesis is that the further you break things down to estimate the time of each piece of work to do, no matter if you are doing agile stories or some other way of organising your work, the longer the project estimate will be. This is beacause the more fine-grained you break things down the more factors like rounding errors, lead times to start each task and some general fear factors will be multiplied in and create a longer over all project time. As the granularity approaches zero the time to complete the project approaches infinity – we have in short a fractal time.

Everyone who has worked in the form of a project knows that most business organisations are top-down organisations that focus on mainly three things, money, time and quality. You are given a budget for a project based on an estimate that seems sound to the investors but when the statistics says that only 1/3 of what was originally described is ever delivered, another 1/3 is delivered but with severe changes and the last 1/3 is obsoleted, replaced and never delivered as orginally stated, then what bearing on the actual work has any ever so good project estimation?

Is this a sound structure for running projects?