Your manager probably has asked you come up with some estimates. Without well-defined requirements and comparable historical information, project managers often resort to randomly pulling a number from thin air.
Such approach is so problematic because there is no basis for the estimates. Oftentimes, the assumptions are also not documented. Moreover, you will be held accountable for the poor estimates that you provided and you’ll have no way to defend yourself if questions arise.
To avoid this problem, clearly define the scope of the work that you need to estimate. Create a mini-WBS as appropriate. Look for comparable work from which you can derive educated estimates and make sure to document all of your assumptions!
Apply three-point estimating to address the uncertainties. Do not just blindly add contingencies—use standard deviation instead. Adjust the estimates as your assumptions get validated.
Connect with Dr. John A. Estrella via Facebook, LinkedIn and Twitter.
Tags: estimate, estimating
As a project manager of some decades experience, I have never in my career done anything like “randomly pulling a number from thin air.” In my career, I have known and worked with hundreds of project managers… None of whom have done anything like that either to my knowledge.
However, this charge is often heard from dysfunctional teams whose estimates were, shall we say, less real than they could have been. While dysfunction among the team is certainly something that can be laid at the feet of the project manager, the estimates from thin air charge is not.
The amount of time it takes and the amount of money spent to complete a project is not known until completion. Estimates are not actuals, otherwise we would call them “exactimates.” In addition to the good advice given regarding assumptions, scope, three-point estimates, and standard deviations, I would add defining, communicating, and gaining agreement on what estimation approach is used and what the final estimate means.
I hate when managers as you to build a project plan without well-defined requirements or comparable historical information. They can’t even describe what the end product is going to be.
If you resist or try to find out more information, you get labeled as difficult, not good at your job and/or not a team player. It’s as if they get perverse joy in setting you up to fail.
Woody, you made two good points. When I said, “Without well-defined requirements and comparable historical information, project managers often resort to randomly pulling a number from thin air.”, I could have said “some project managers sometime”—which is actually what I meant to say.
I also like your second point about agreeing on the approach—well said.
devans00, I guess it boils down to how you communicate your need for well-defined requirements AND the maturity of the organization to define/follow proper estimating processes. Without these, it is indeed an uphill and frustrating battle.
The key is in the requirements gathering, if it’s done right, then the estimates can be realistic. Btw, an estimate, by definition, is “an approximate calculation of quantity or degree or worth” (Princeton), so it doesn’t have to be 100% accurate. Most executives fail to understand the real definition of estimates.
Oh, and btw, here’s a decent article on requirements gathering in software projects, hope you’ll have the chance to read it and leave your feedback.