Project Management 101: Estimation Techniques

Estimating is one of the most fundamental core skills of project managers. It lays a solid foundation for the effective planning and scheduling of projects. So, here are some reasons why to use estimation techniques in project management:

• To plan and schedule projects effectively: Estimates can help project managers to plan and organize projects by providing a rough idea of how long each task will take and how much it will cost. This information can then be used to create a project schedule and budget.
• To communicate with stakeholders: Estimates can help project managers communicate with stakeholders by giving them a realistic idea of when the project will be completed and how much it will cost. This helps build trust and confidence between the project manager and the stakeholders.
• To identify and manage risks: Estimates can help project managers identify and manage risks by providing a baseline against which to measure actual performance.

Remember that the same estimation methods can be used for schedules and costs. It means that by mastering one, or preferably a few estimation techniques, you can re-use them to cover two out of three points of the Project management triangle - Scope\Cost\Duration. So what are the most commonly used methods?

Analogous Estimation

Analogous/top-down estimating is a project management technique that utilizes historical data from similar projects to estimate the cost and duration of a new project or task. It is a top-down method since the project managers start with a high-level estimate and then break it down into individual tasks.

It is also a quick and easy way to estimate a new project. Still, it needs to be more accurate than other estimating techniques. The best time to use this method is during the early planning stages when little information is available about the project. Then, based on the business case, the scope of work, and maybe even a project charter, the project manager identifies a similar project within the organization that has already been completed. Then the project manager needs to look for the factors that are similar or comparable between the projects. It could be a certain number of elements, deliverables, similar technologies and several people involved, or anything related to the scope of work, size (scalability) and project complexity. By utilizing these factors, the project manager can estimate the cost and duration of the new project.

To summarize, here is a list of these method benefits:
- It is the quickest and easiest estimation technique
- It does not require a lot of data as long as projects are at least somewhat comparable
- If projects are scalable, there is no limit to the size of projects and their estimations.

Some drawbacks of analogous analysis:
- It is less accurate than the majority of other techniques. The use depends on the acceptability of RoM.
- It could be challenging to find similar projects within the organization.

Parametric Estimation

The parametric estimating technique uses historical data and statistical relationships to estimate a project's cost, duration, and other parameters. This method is significantly more accurate than the analogous estimating technique but, at the same time, is much more complex. To use this technique, the project manager must analyze historical data relevant to the new project. This could include the cost, duration, or any other parameters of similar projects or resources used at the project.

Then the project manager uses statistical methods and mathematical relationships to create a model to estimate the required parameters and unknowns of the new project. This is a common method in construction or any other industry where costs are calculated per unit. For example, you can think of a specific dollar value per square meter, meters of cables, etc. As you can see, this method requires more data and effort than the analogous technique.

Quick summary of benefits:
- It is more accurate than analogous estimating.
- There is no restriction on the project size.

In terms of drawbacks:
- It requires more data and effort than analogous estimating.
- Finding an organization's historical data regarding completed projects can be challenging.
- Challenges associated with the creation of an accurate model.

Expert judgement

Use your subject matter experts (SMEs) to get your needed estimates. Often, project managers don't know how SMEs come up with their estimates, but they don't really need to know. You need to ensure that you have trustworthy professionals and contractors working on your projects.

Bottom-up Estimation

Bottom-up estimating is the most common estimation technique used outside of project management. This method breaks down a project into smaller tasks and then estimates the cost and duration of every task. The project manager estimates what resources are needed t complete a task and uses them for the estimations. By summarizing these individual task estimates, the project manager can evaluate the total cost or duration of the whole project, thus, bottom-up (task-project) estimation.

Based on the description of this method, you have already noticed that it should be more accurate than the analogous technique that uses historical data for its estimates.

On the other hand, it is challenging and time-consuming to estimate every single task on a project. Especially if it involves multiple departments and people, this exercise could take a long time to complete.

Three-point and PERT estimation

Three-point or triangular distribution is an estimation method that can be used as a stand-alone technique to improve other estimation techniques' results. This one uses three values:
- Optimistic/The best case scenario (O). Imagine that everything goes flawlessly - which would be your optimistic cost or time.
- Pessimistic/The worst case scenario (P). And now imagine the opposite, everything goes wrong; what would be your time or cost forecasts?
- Most Likely (M). What are your realistic estimates?

By using this formula for the triangular distribution E = (O + M + P) / 3
you can calculate your average values. It should be noted that the weights of all three numbers are the same, meaning neither affects the final result more than the others.

Conversely, PERT transforms the triangular distribution into the bell curve displaying a normally distributed curve. This method considers that the most likely estimate has more weight than your calculations' best and worst-case scenarios.

Its formula is E = (O +4xM+ P) / 6, and the PERT technique's standard deviation is Standard Deviation = (P – O) / 6.

The approximate resulting probabilities are:
- 68.3% for one standard deviation,
- 95.5% for two standard deviations,
- 99.7% for three standard deviations.

This is the empirical rule in statistics (68%-95%-99.7%) to show the percentage of values within an interval estimate in a normal distribution. These numbers correspond to one, two and three standard deviations of the mean value.

As an example, let's say that our subject matter experts got back to you as the project manager with the following numbers regarding one task duration:
- Optimistic - 15 days
- Pessimistic - 60 days
- Most likely - 30 days

According to the PERT Final estimate = (15+4x30+60)/6 = 32.5 days, and the triangular estimate = 15+30+60/3 = 35 days.

PERT standard deviation is (60-15)/6=7.5 days, meaning that we can determine the probabilities:

The most significant difference and advantage of the PERT estimate are that the "Most Likely" scenario has the highest weight, unlike in the Triangular Distribution, where all scenarios have the same weight. Also, using the PERT method allows us to consider probabilities of value ranges.

Summary

Estimating the time and cost of activities is critical for planning and scheduling projects. However, in many cases, more accurate estimates, such as parametric ones, are unavailable. The three-point estimation technique can process and balance top-down or subject-matter expert estimates.

To improve the results even further, the PERT distribution is a probability distribution that can combine the three-point estimates into a single estimate or a range of estimates. In addition, the PERT distribution considers the uncertainty of the forecast and the potential for variation in the calculations.

The PERT distribution is a valuable tool for project managers because it allows them to make more informed decisions about planning and scheduling. In addition, by considering the uncertainty of the estimates, project managers can better manage risks and ensure that projects are completed on time and within budget.