Managing by objectives has been around for a long time. It is true that if people are given objectives they usually try to meet them, they want to do a good job. But let’s look at the situation in project management (a large proportion of business activity is projects) – a function that is famous for completion delays, budget overruns and project promises not met. Let us look at how managing by objectives delays projects!
Imagine you are one of the people doing one or several of the tasks in a project (Use the drawing to follow). You have many years of experience and you know very well that Murphy’s Law applies to tasks in projects. You also want to do a good job. So, you estimate the time of your task so that you have something like a 90% certainty of completing it on time. You provide your 90% estimate (plus a little bit more for management to cut). In the end you are committed to a task time that gives you, what you hope, is enough safety time. Moreover, everyone else in the project has done the same thing.
What have we done? Every task has an enormous amount of safety in it. This is because the probability distribution of a tasks completion time is skewed – with a long tail to the right. In many cases the time between the median (the time where you have a 50:50 chance of being early or late) and the 90% of finishing on time can be double and triple an estimated task time.
The picture above describes the situation. This means that there is a lot of safety in a project. In fact knowing that the sum of a series of events has a tighter variation than any individual event means that for the project we actually have much more safety than just 90%. So why are projects so often late?
Usually most people, being very busy and knowing they have a lot of safety in their task time estimate, will not start work straight away. When they do finally start they have frittered away a lot of their original safety, they have wasted it. Most of the time they finish on or near their original estimate – but those with bad luck finish late – sometimes very late. So with most people finishing on time (meeting their objectives) and only a few that are very late the project inevitably is late. Managing individuals by objectives is not a good idea in projects – what we want is everyone working to one objective, the project due date – task due dates are in fact meaningless.
Maybe it would be better to cut task time estimates in half and, since statistics help us, put only half of what we have cut into a project buffer. This should eliminate ‘student syndrome’ (starting at the last minute) and with better communication task hand-offs will happen with minimum delay – especially if everyone is now measured on the project performance. Will it work? Probably, but I recommend the reader studies the book ‘Critical Chain’ before attempting to make such a change.
