When you have a plan to accomplish something, how to do know if that plan is a GOOD or BAD plan?
Is the plan good if it presented in an aesthetically pleasing manner? Is a plan good if people understand it? Is a plan good, perhaps, if some people do NOT understand it? What do you employ as a plan measuring stick?
In fairness, I would say a plan is typically considered to be good when reviewers of the plan feel that it addresses all necessary aspects of the project or effort and has tasks scheduled in such a way such that it seems the steps of the plan will reasonably get completed as specified. In other words, it holds up to pressure testing by stakeholders.
This said, however, this measurement would rely on many assumptions.The most important assumption is that you and the stakeholders understand all of the “necessary aspects” of the project. One way to foster understanding is to compare your project tasks to the tasks of previous projects. This, of course, is only possible if these artifacts for comparison exist and you know the degree of similarity.
When it comes to knowledge projects, such as software development, I assert that nobody really knows if a plan is good or bad. You only know if something is done and works once it is complete.
Completing work comes through solid execution. In fact, this is partially supported by 3 of the 4 phrases in the agile manifesto (and remember while we value the left part of these phrases more, there is still value on the right as well):
Also note the seventh principle behind the agile manifesto: Working software is the primary measure of progress. In other words, pointing to where a team may be on a plan is often an arbitrary measure of progress. Working software that results from execution, on the other hand, does not lie!
Additionally, SAFe Principles support focusing on execution as well.
Setting aside all of the observations above, how do you feel when you have to give a “percent complete” measurement based on a work in progress?
I personally do not like that. If the work is broken down into small pieces which can each be complete or incomplete and a percentage is derived from this, I feel much more confident in giving a percentage value measurement. This, again, is accomplished through executing completion of the fully functional artifacts.
With that in mind, check back for part two of this post, where I will give some steps you can follow to focus on execution.