Brooks' Law states that it is almost always true that "adding more resources to a late software project only makes it later". The law is named after Frederick Brooks, a PM who documented it over many years ago in his classic collection of program management essays entitled The Mythical Man-Month.
You've seen this in action. When deadlines approach, the projects scream for more people. The more crucial the deadline, the more likely the request will be granted. A downward spiral just spirals downward faster. The failing project that inevitably ends (but may or may not be complete), now has more people to share the blame with. What crucial, late project has the time to hire quality members or TRAIN them. Throwing bodies at the problem is like kindling in a fire.
EVERY project that I have been on has gone through this experience. In each case, manager's are firefighters that promote heroic efforts.
Readers of this blog entry have the desire to break this cycle and focus on the big picture and the life-cycle planning. What can we do?
1. network with like-minded process improvement friends, wherever in the world they are at the moment, they are still a sounding board for you to examine the root cause, key stakeholders, and roadmap to recovery.
2. create and track a performance metric to be an early indicator of project problems to the triple constraints (budget, schedule, or quality).
3. Project Management Institute (PMI) says to plan project scope (and resources) to ensure a 18 month schedule. Expect a commitment from teammembers to complete the project, provide golden handcuffs if need be. Turnover kills. Don't lose project momemtum because a teammember leaves for a 20% increase in pay. Society of Human Resources Management (SHRM) says the employee turnover costs 100-200% of an employees pay to replace (besides recruiting and training, think of opportunity cost!).
4. if you add teammembers late, hire known, trusted, proven team players. Prima donnas will isolate the team; consultants will baffle them with BS and not produce anything. Get team buy-in early on that a threshhold metric requires a senior teammember addition.
5. don't hire entry level support in a crisis. They add no value and learn the wrong way to react. A project in chaos will feel natural to him/her. Likewise, inject apprentices on a schedule, don't overwhelm the team resources and don't expect any return from the new hire; they are an investment to learn the correct processes and the culture to become contributing members on the next project.
Following any of these points will help your project avoid the common pitfalls that lead to mass hysteria and rapid degradation.
No comments:
Post a Comment