My Projects

December 31, 2006 1881 comments so far

How to Accurately Estimate Effort

One of the most controversial topics in project leadership is how to accurately estimate planned activities. The major reason for this is simple: You are dealing with human beings and not with machines. And if you didn't know it yet: Humans are not always predictable ;-).

The knowledge worker is a human being, not a machine: People can have good days, they can have bad days, they can get sick, etc. The very same person can be totally effective on Monday and may not be able to concentrate on Tuesday - for whatever reason. A good project leader knows his team members, and - depending on the situation and the type of project or company - also beyond work.

Of course it depends on which type of project leader you want to be. I personally, like to do project leadership on a friendly basis, meaning that I also try to spend some time together with my project members, e.g., having lunch with the team, or going out for a beer in the evening where we do not only talk about work, but also try to get to know each other a little bit better. At the same time I always make it clear that there is only one project leader and this is me, meaning also that if we are not of the same opinion, it is my opinion which counts.

Accurately estimating effort is generally quite difficult and I do not want to go into details about estimation methods at this time. Instead, I want to provide some general rules I believe are important to obey, regardless on what estimation technique you use:

  • Let your team members estimate their own work, rather than telling them how long it should take them to finish the task. Top down micro management does not work very well in most of today's modern work environments
  • Double check your team members estimations for plausibility (if you are not deep enough into the technology, ask someone else to do it for you): Trust is important, but control is better
  • Learn who in your team is how good in estimating effort: You will see over time that not everyone is good in self-estimating work (BTW, from my experience, most people who are good in self-estimating will become quite good project managers)
  • Finally, use a sensible time unit for all estimates: If you are dealing with real planned activities and not ad-hoc tasks you are most probably better off with days rather than hours

The last two points are also very important when scheduling estimated activities, especially in conjunction with so called "buffers": We will take a closer look at this topic in the next article. Until then, maybe you want to tell us what kind of project leader you want to be and what you are doing to achieve this? In any way, I wish all of us a Happy and successful New Year!