Contents ContentsPrev PrevNext Next

Letters to a Young Manager

The End of Coding, #20
LTYM > Project Management

Dear Sophie,
Why are engineering estimates usually a bad predictor of when the product will launch? I'll write you later about the maxim that half of what you need to know in IT project Management, you will learn from the doing [1]. Users can't give you perfect spec's. But even if they could, the building of the system is only part of the work.

While at a subsidiary of Lotus in the 1980's, what I found about software projects, is that when the code was written and tested; when the legendary "gold disk" was produced, was the half-way point in the project life-cycle. All the marketing, sales planning and launch preparations followed. That involved departments outside of IT. So if you're only talking to your engineers, you only have part of the story. Ask and listen to all involved before you commit to a delivery date.

[1] See Story #170


Feature freeze, or the point at which all required features has been added to the system, is the mid-point of the development cycle, not the end-point

Discussion Questions:

1) What are the steps required to launch new software in your organization?
2) Are these points irrelevant in a "perpetual beta" approach to software in the web-world? Why and why not?
3) Does feature freeze make sense for current project management approaches? Why and why not?

For Further Reading:

© Copyright 2005, 2024, E. G. Happ, All Rights Reserved.