“One manifestation of the overconfidence effect is the tendency to overestimate one’s standing on a dimension of judgment or performance.” Wikipedia
We know! And clearly we are good at knowing. We know how to drive a car! We know how to code in Cobol, in C# or whichever language you can dream of. We know how to paint a wall. We know how long it takes to mowe the lawn. And if somebody asks us how long it takes to do it, we are very keen in giving a well defined estimate.
Unfortunately, in 75% of the times, these estimates are just purely wrong. We will paint a wall in 3 hours and practically, 6 hours later we are still fighting with it, probably insulting the wall, the brush, the paint and anyone else showing up in the room.
Imagine now that you need to prepare a project schedule and you need estimates from your team. “How long will that take you?” “3 days” “Are you sure?” “Of course, I know what I am doing.”
Yeah right!…
And 6 days later, the guy is still busy with the same job.
And this happens every time!
And you cannot even blame him for this because he was really honest when he gave you his estimate.
Ask an expert for the time he will need to perform a job, and he will be wrong 75% of the time.
Isn’t that just awful? That’s the overconfidence effect. We are certain that we are better at doing something than in reality. Considering that we work in teams, this means that EVERYONE is potentially giving you an very incorrect estimate.
How can we prepare a realistic schedule when nobody is able to provide you a realistic estimate?
A solution
The best solution for a project manager is to systematically take a contingency in the estimate provided. Never trust anyone! They all are wrong. And you are wrong too when you talk about your own job.
Ask a peer who has some knowledge but is not responsible, how long it will take. Multiply the estimate by 1.25 or even by 1.5. In the worst case scenario, you will be only 25% too short and in the best case, you will be 5% faster than planned.
Never trust anyone!
Recent Comments