This is the second in a series of articles on estimating effort. In the first, I focused on definitions since I find that many people use terms such as estimate and budget as synonyms when, in fact, they are very different. Here is a brief recap of the key definitions that I will use:
Estimate. An informed assessment of an uncertain event. Informed means that you have an identified basis for the estimate. Uncertain recognizes that different outcomes are possible.
Effort. Effort is an expenditure of physical or mental effort on the part of a project team member. Effort is normally measured in terms of staff hours.
Budget. A management metric that uses the effort estimate as one of the key inputs.
Baseline. A time-phased budget that has received all necessary approvals.
What are you estimating??Let?s start with a fairly simple example to build a conceptual foundation.?It may seem obvious, but the first requirement for developing an estimate is to know what you are estimating. For now, let?s assume that you have been asked to estimate how much effort (how much of your time) is likely to be required to paint your bedroom. Although this is a fairly small activity, it is still one with a significant amount of uncertainty:
- Do you have to paint the ceiling and the woodwork, or just the walls?
- Are there windows to be painted? Do they have mullions?
- Is there furniture in the room? Can it be moved? If it can be moved, is moving it out and back in part of ?painting the room??
- Is choosing the color part of this activity? What about going to the store to buy the paint?
- How many coats are needed? Is there any special finishing involved (like marbleizing)?
- Will anyone be helping you? Or will you be doing all of the work yourself?
If you don?t know the answers to these questions, you have a couple of choices:
- You can decline to prepare an estimate at all.
- You can prepare an estimate that allows for a large amount of uncertainty.
- You can make some assumptions about the answers to these questions and estimate based on those assumptions.
- You can try to reduce the uncertainty by getting answers to the questions before preparing an estimate.
Any of these approaches is acceptable as long as your stakeholders know which approach you are using!?Most of the time, you will use some combination of approaches. You will make some reasonable assumptions (and you will always document all of your assumptions!). You will ask some questions. Then you will apply your best judgment.
Range estimates.?You can improve the usefulness of your effort estimates by making range estimates. How does a range estimate work? In its simplest form, you estimate a low value and a high value for the amount of effort that you think you are likely to need. For example, in the case of painting your bedroom, and assuming that the work is limited to applying paint to the walls, you might estimate 2-3 hours.
The fundamental concept behind range estimating is that you don?t need to know the exact amount of effort that will be required. You do need to know that it isn?t going to take a full day out of your schedule. You do need to know that this isn?t a trivial activity that can be completed in a few minutes. As long as the actual result is somewhere between 2 and 3 hours, you have made a good estimate.
Even if the actual result falls outside of that range, you still made a good estimate since it was the best estimate that you could make given what you knew at the time. In addition, you can learn from any?variance (an actual result that falls outside the predicted range) and use that information to try to improve your future estimates.
Using range estimates takes much of the pain out of the estimating process. If you are accustomed to single point estimates, it can be very difficult to choose between 2 hours and 3 hours. Even proposing a single-point estimate of 2.5 hours can be difficult because you know that it might take more or less than that amount. Using range estimates allows you to estimate even in the face of uncertainty.?If your boss insists on a single point estimate, he or she is most likely looking for a budget, not an estimate. There?s no problem with giving them a single number as a budget, but you really should share your estimate with them as well so that they will understand how much or how little uncertainty is involved.
Three-point range estimates.?You can improve the?usefulness?of your effort estimates still further by making three-point range estimates. In addition to the high and low values of a simple range estimate, you add an assessment of the most likely in-between result.
Let?s say, for example, that you are living in the world of Groundhog Day such that you can paint your room 100 times and keep records of how much effort it actually took. You discover that you never spent less than 2 hours and never more than 5. You also discover that your actual results produce a triangular distribution with a peak at 3 hours as shown below:
Based on the availability of this actual, historical information, you would estimate the amount of effort required to paint a similarly sized bedroom as follows:
- Most likely: 3 hours
- Optimistic: 2 hours
- Pessimistic: 5 hours
I understand that you will seldom if ever have such wonderful, perfect historical information to base your estimates on, but the great thing about range estimating is that you don?t need perfect data! As long as the three-point range estimate is reasonable, errors in the effort estimates for individual activities are likely to balance out over the course of the project.
Even in this case, even with perfect data, you still don?t know how exactly how much effort will be required for the next bedroom that you paint. It might be 2.5 hours; it might be 3.75 hours. Doesn?t matter. What is important is that you have effort estimates that are reasonable; that you have effort estimates that will allow you to manage your project more effectively.
Next: preparing three-point range estimates