There are simple project management tools that allow us to quickly and accurately estimate projects (or tasks) even when we don’t have a lot of time to make the estimate nor a lot of details on the effort.
Ron Rosenhead posed an interesting dilemma in his article Project estimating; avoid the must answer now syndrome. It concerned what to do when put on the spot to immediately commit to certain actions by certain dates. Ron’s guidance was to have a diplomatic way of saying something like “I will need 24 hours to look at what is fully involved from our end and from yours and the resources we have to deliver” and then to provide an answer in that time. My goal, instead, is always to know enough about how we’ve performed in the past to be able to provide a good estimate when asked.
When I was in the military and asking contractors for quick estimates in meetings, the folks that said “we can do that in X weeks” generally got the work over those that said “we’ll get back to you on that.” Neither type of contractor was very good at hitting their estimates however, but the ones who were willing to say “we can do that” were also generally the ones who were more responsive during the life of the project. Similarly, we want to be able to say “we can do it” and be reasonably accurate at the same time. (See related The Leap From Mediocrity To The Exceptional Is Shorter Than We Think.)
Knowing enough ahead of time so we can quickly estimate and give a quick, high probability of correctness, answer is where I generally try to be. If I do not know how my team is currently performing (and what resources are free, etc.) then another 24 hours is generally not going to improve my estimate.
Achieving this level of estimating speed and accuracy is not as hard as one would expect. It essentially requires we know — using hard data — how we have performed in the recent past. For many organizations, recent past performance is the key to accurate and rapid estimates. Three examples of very successful rapid estimations that I’ve used include:
- “When can you ship this new product we just outlined to you.” I can ship it in eighteen months. I know this because every new product we’ve shipped in the last three years took an average of eighteen months with little variation from the average. If this is the only opportunity I get to make my offer, I know that the odds are in my favor that I’ll ship on time. (For more details, see Get The Project Management Schedule Right!)
- “When will that defect get fixed?” I know the current average and variation on how long it takes to fix all our defects. I can say with great confidence that it is going to take right about ten days (i.e., the current average). I can say this without knowing a thing about the defect. (For more on this technique, see Your Project Management Average Is Powerful.)
- “When can my customer get that new feature?” They can get it the third quarter of next year if we get the requirements nailed down over the next 30 days. We knew it took us nine months on average to deliver a new feature. We knew immediately where (which quarterly release) to plug in new requirements when a customer (or account executive) came calling.
For this to work we need to regularly track how we perform when doing the tasks we are regularly asked to do. This trains our gut to be able to estimate, on a moment’s notice, how long something will most probably take. Yes, no two projects, products or requests are exactly the same. Yes, when using averages we are often overestimating and hence might not be competitive. The more practiced we are and the more we’ve done this the better we get at adjusting the estimate higher or lower based upon what we’ve heard. But I’ve rarely, almost never, used something other than the average.
Many people will actually not want a quick answer. Their confidence in our estimate will be higher if we take another 24-48 hours or even a week to provide an estimate to their request. However, if we are walking away from such a meeting and we have no idea what it will generally take, then the additional time might not help us improve our estimate. The goal is to know how our team has been performing and be prepared to provide an estimate or even a commitment if needed.
Being able to quickly and accurately estimate what we can commit to is the simple fruit of knowing how well our organization performs. Understanding this performance, using even simple averages, will help anyone become much more comfortable and consistently successful when being put on the spot for a commitment.
Do you know how long it takes, on average, for your project teams to do the most common and critical tasks?