Project Manager: How's the project progressing to plan?
Developers: We're spending money, consuming resources, producing outputs that the customer likes.
Project Manager: I was more interested in what's our performance against our planned spend, planned resource consumption, and planned outputs of value to the customer?
Developers: What do you mean, we didn't estimate any of that, we're managing this project with #NoEstimates. You know that new alternative to estimates for making decisions in software development. That is ways to make decisions with "No Estimates." of the impacts on the future of those estimates or of our work on the future cost, schedule, or technical performance. You know where we can use Decision making frameworks for projects that do not require estimates, or apply Investment models for software projects that do not require estimates, and have our project management methods for risk management, scope management, progress reporting, not require any of those annoying estimates. 'Cause we kinda suck at them anyway, so we just decided instead of learning how to estimate, we'll just not estimate and get back to coding.
Project Manager: Oh, you mean that approach of managing other peoples money that violates the principles of software microeconomics with Open Loop Control - where our organization can make business decisions on the allocation of our limited resources, without examining how those decisions effect the supply and demand of our resources. You do know about those resources? Like money, people, and time?
Developers: Yea, we don't need any that mumbo jumbo microeconomics that we all learned in school, since we didn't pay attention in that boring the statistics and probability class that tried to teach us that all variables on a project are actually random variables and we should to know something about their behaviour in the future if we're going have a hope in hell of ever managing this project in the presence of uncertainty about those values.
Project Manager: What's that smell? Maybe we'd better start rearranging the deck chairs on our ship here real soon, cause I smell an Iceberg getting closer.
No project can be managed to successful closure in the absence of steering targets defined at periodic intervals for the expenditure of cost, schedule, and technical performance. Knowing what those steering targets should be requires estimating their values, then measures the actual values to develop the needed steering signal - the variance between plan and actual.
The only way out of the need to estimate those intermediate steering targets is to straight line the budget, schedule, and needed technical performance - from start to end, then measure the actual performance.
Like the intended route of the Titanic, our project does not proceed in a straight line, so that idea is a non-starter. And like the Titanic, our project cannot confuse the intended speed with the actual speed, just like we can't confuse the budget - the total planned crossing time - with the actual cost - the actual total crossing time.
Without those pesky intermediate targets to steer toward - those targets created by estimating the needed cost, needed scheduled arrival date, and, needed capabilities on the needed date for the needed cost. We're managing the project Open Loop, driving in a straight line. Never knowing what will pop up in front of our path.
Say good bye to Kate Leonardo, you're gonna get wet.
† Full attribution for the inspiration for this post comes from the very useful blog by Gene Hughson
