Why Are Software Development Time Estimates So Wrong?

Here’s a great answer in the form of a hiking analogy to the important question of why software development time estimates are routinely off by factors of 2 or more. (Warning, some PG13 language in this post.)


As I read this, I first thought, the same time applies to most book writing, especially computer books about computer programming. You rarely know all of the little twists and turns your book will take before you write it, especially for first time authors.

My second thought was way back to graduate school mathematics when I was fascinated by complexity and specifically the Mandelbrot set that was all the rage in the early 90’s. The hiking map, programming, and writing all resemble the infinite complexity of the Mandelbrot set. No matter how smooth they look from a distance, every step closer you get uncovers another layer of complexity which itself looks smooth but again, is hiding the next layer of complexity.

Programmers, authors, what do you think? Is writing code or a book like walking along a raged shoreline you’ve never explored?



2 responses to “Why Are Software Development Time Estimates So Wrong?”

  1. zoo says:

    Some times, the programmer drop into a problem and they cost too much time to make it out.

  2. AKS says:

    Because the PM puts pressure on the timelines and often the developer cannot correctly estimate and evaluate all dependencies and the complexity involved in coding the requirement. Cook a curry in a hurry and then keep applying patches. I am a developer and I want to know how a buider builds a high rise and it does not fall. There must be something wrong with the process or it is crowded with too many non IT people securing managerial IT roles. Maybe, few of them have managed a football team.

Leave a Reply to zoo Cancel reply

Your email address will not be published. Required fields are marked *