In order to take corrective action for our project, to keep it on schedule, on budget, and producing the expected outcomes (more or less within out variance bounds) we need indicators of performance. Cost, schedule, and techncial performance.
In order to do that, we need a baseline to which to compare our measurements. Without the baseline - the target - we have no way to measure our variance. Without a variance, we have no way to know if we are off track.
So What Does This Mean?
- Know what capabilities the project is supposed to produce. Keep those in mind at all times.
- Plan far enough in the future to see trouble coming in time to take corrective action
- Define the edges of the road. Keep the project inside the white lines. Staying in your lane is the best, but at least stay out of the ditch.
- When estimating the future, don't guess, estimate. Anyone claiming estimating is guessing has not been on a project where estimating was done right. Ignore them. Look for reference classes. Look for past performance. Apply the 20 questions process. But don't listen to those who say estimating can't be done.
- Update your estimate all the time from new information. The purpose of the windshield on the car is to look out and see where you are going. Use the feedbacl from you project performance measures for that.