Steve McConnell wrote a great article called Building a Fort: Lessons in Software Estimation where he tells us the story how he built a fort for his children and what classic mistakes he did during this adventure. It is a brilliant lesson of how many mistakes we can make when we are put in a situation where we are not too familiar with the nature of the problem even if we are very experienced in the area of project management. Steve provides a thorough analysis of what went wrong and what were his mistakes in planning and estimating. Of course, it would be more useful for him if he had done it before he started his endeavor.
Although it’s not a software project, it’s a great example for all of us how many things we should think about and should take in consideration when making our plans and estimates. And it is always better to do this analysis in the beginning of our project instead of doing it at the end.
Jeff Atwood at Coding Horror wrote an article called Steve McConnell in the Doghouse where he shares his thoughts on McConnell’s fort-building experiment, which you can also find interesting. His main point is: One size doesn’t fit all and No two projects are the same. So you should always consider the uniqueness of each project you undertake and make your estimates very carefully.