Three point estimating is a technique which utilizes an optimistic and pessimistic estimate to determine the ideal estimate value for a project task. It is a shoe-in for PMP exam questions, which is not likely to change anytime soon.
It allows known risks to be quantified and built in to the project budget.
Formula
There are two generally accepted formulas for three point estimates:
- Triangular Distribution
- Beta Distribution
Triangular Distribution
The simplest three point estimate is the simple average of the three values (known as the triangular distribution):
Where:
- E = estimated cost
- a = optimistic value
- m = most likely value
- b = pessimistic value
Beta Distribution
The beta distribution places the final estimate closer to the most likely value:
Where:
- E = estimated cost
- a = optimistic value
- m = most likely value
- b = pessimistic value
The triangular distribution is the default and should be used if there is no reason to use anything else. The beta distribution should be chosen when there is more confidence in the most likely value, that is, where the final estimate should be tighter to the mean.
Three Point Estimating and Risk Analysis
The key benefit of three point estimating is the way it takes into account project risk. The following procedure is how it works, PMBOK-style. From the very beginning:
- The project success factors are determined
- Risks which might affect the project success factors are identified.
- The risks are analyzed and prioritized to determine which are the most important.
- The project is divided into tasks (a work breakdown structure).
- Each task receives three estimates:
- Optimistic. Assuming the risks do not occur.
- Most likely. Assuming the risk occurrence level that is most likely to occur. Similar to other estimate types.
- Pessimistic. Assuming that the risks occur.
- The Three Point Estimating procedure is used to determine the final estimate for each task.
There are many situations where a risk results in a skewed distribution. For example, a risk might result in a cost escalation but no corresponding cost reduction if it doesn’t occur. This would make the Pessimistic estimate farther away from the Most Likely value than the Optimistic one (that is, the Most Likely value is not directly in the middle) and result in a skewed distribution.
Other Task Level Estimating Techniques
Three point estimating is one out of three estimating methods which are applied at the individual task level. The other two are:
- Parametric Estimating: Estimation of a task using a unit rate (parametric rate). For example, house construction costs $120 per square foot.
- Analogous Estimating: Estimation of a task using a previous, similar project. For example, the flooring in the previous house cost $10,000 therefore this one should cost $12,000.
These methods apply to individual tasks and are thus contained within the PMBOK‘s Estimate Costs process.
Overall Project Level Estimating Techniques
Do not confuse the three task-level estimation techniques, above, with the following two project-level estimation techniques. The PMBOK contains a separate process for rolling up the costs into the overall project, called Determine Budget, and it contains the following two procedures:
- Top Down Estimating: This refers to determining the overall project estimate first, and then apportioning it into individual tasks. For example, the previous house cost $240,000 to build (analogous), therefore this one should cost $260,000, which will be apportioned among the tasks. Usually, this technique requires task level estimation as well, to determine the correct ratios to use for apportioning.
- Bottom Up Estimating: Opposite of Top Down. This refers to determining the individual task estimates first, then rolling them up into an overall project estimate. Bottom Up is the norm in most project management applications. The project manager must estimate on an individual task level and determine the project budget by summing (aggregating) the individual tasks.
Good luck with Three Point Estimating on the PMP exam, and let us know in the comments if you have anything to add!