(Originally Published on Iowabiz.com in June 2007)
It's hurricane season, and here in Iowa I feel safe and snug dealing with the occasional tornado watch and the even more infrequent tornado warning. I'm always fascinated with the process of tracking hurricanes and, being a Weather Channel junkie from time to time (my wife teases because I know all of the meteoroligists names and that I consider Jim Cantore one of my heroes), I can glue myself to the television whenever a big storm is brewing. To me, that's real news.
Now, imagine if the folks at NOAA placed one of our "typical" corporate executives in charge of their hurricane center. I can just hear the conversation now:
Executive: Jenkins, I see there's a new hurricane forming.
Jenkins: Yes sir, it's just a tropical depression off the coast of Africa right now, but we'll see how it develops.
Executive: Screw that namby-pamby generality stuff, Jenkins! I want to know how strong it will be, exactly where it will hit, and when. Ever since Katrina, our stakeholders are demanding better information sooner, and you know it.
Jenkins (blinking in disbelief): But, sir... the fact is that we just don't know all of that information yet. There are too many variables affecting each storm. Jet streams. Air currents. Pressure fronts. There's no way...
Executive (interrupting): You will get me that information by close of business today, Jenkins. And you'd better be right. Your job is riding on it. Now go!!!
Sigh. We think it sounds ridiculous, right? After all, we all understand the "cone of uncertainty," and we allow the meteoroligists some degree of latitude over the course of a few days and several thousand miles to provide us with the best information known to date.
Estimating is one of those thorns in the side of project managers and project stakeholders alike. As a project manager, I understand that there are variables that can affect any given task. There are risks that can undermine even the best-laid plans. There are assumptions that need to be made about how long something will take.
A couple of months ago, James Sawyer, a Senior Analyst with TranSystems / Automation Associates, Inc. posted this about estimating in his blog. Here's an excerpt from his thoughts:
As we get more and more experienced, we're supposed to get better and better at the accuracy of estimating our tasks. But bad things can and do happen during the course of a project, and our estimates don't always line up with what it takes to do the work in reality. This can be frustrating for us and for the PM, especially when it starts blowing the budget. But why does it happen? Maybe it's not just that we're horrible estimators (though all of us need to continue to improve in this area), but that there's something inherently funky about the act of estimating itself.
I think Mr. Sawyer hit on the key issue: ESTIMATES ARE NOT REALITY. They should not be treated as such. They are important, and they do constitute the lifeblood of the project's cost and duration and resources needed. However, they should be drawn in sand rather than cast in cement. Based on my own experiences (as well as advice shared by trusted colleagues), here are my top 10 tips for estimating:
- Estimating is a team sport - do no rely on an estimate given by only one person. Do a Google search, talk to similar professionals, and ask intelligent questions.
- Estimates improve as time progresses - revisit your estimates on a regular basis. As you learn more, you may be able to provide better data to reset expectations.
- Do not punish bad estimating - instead, learn from inaccurate estimates. Why were we so far off? What didn't we know that we should have known?
- Document assumptions - undocumented assumptions resurrect later as excuses. If you know that an estimate is only true if certain conditions exist (e.g., on-time delivery), then write that down. It gives you more credibility if the assumption was violated than it does if your estimate was wrong.
- Identify risks - new technology, new processes, and inexperienced team members can all affect an estimate. Make sure you go into your estimates with your eyes wide open.
- Apply contingency according to risk - avoid the urge to add 50% across the board. Look at each task individually in light of #4 and #5, above. Those tasks or resources that have more assumptions and risks associated with them should be given more contingency time and cost.
- Sometimes a SWAG is OK - for those not familiar with the term, SWAG stands for Silly Wild-A**-Guess. And there will be times when that is the best answer until additional information becomes available. While there are tools such as PERT to help you, occasionally the best estimates come from the gut.
- Estimates are more about communication than math - in reality, setting and sticking to a budget is often-times subordinated to ensuring that expectations are set (and reset) appropriately and that key stakeholders are not blindsided.
- Obtain signoff on estimates from resources - make sure that the resource working on the task agrees with the estimate allotted for that task. If they don't agree, you're just setting them up for failure.
- Look at the big picture of the whole project - if your individual task estimates are off, but your project concluded within a reasonable time frame, celebrate the accomplishment. As discussed in #3, punishment does nothing to improve estimating skills. Use it as a learning tool for next time, but let the team know how proud you are that the project completed.
Estimating is only a fraction of the activity which a project manager performs. Don't misunderstand me; it's an important aspect of project management. It is not, however, the only aspect of project management. Following the tips above should make it a slightly easier task.