Improving Project Estimation Accuracy

Accurate project estimation is one of the most challenging aspects of a project. Project estimation becomes increasingly difficult as the project’s complexity and uncertainty increases. However, project estimates can be accurate.

The powerful project estimation techniques that are explained in this article will quickly improve the accuracy of project estimates, even if you have no project estimation experience.

Before Estimating Project Costs

Before you begin project estimation, there needs to be an understanding of the scope of the project. If you don’t know what the project is trying to achieve, then there is little chance of being able to accurate estimate the effort required.

The more detailed the scope of the project, the more detailed and accurate the project estimate will be. The size, complexity and stage of the project will impact greatly on the level of accuracy required, the amount of cost and time the business can commit to project estimation and the level of understanding and clarity of the scope of the project.

It is therefore important to align the project estimation requirements to the required accuracy and stage of the project.

Project Estimation Accuracy

When estimating a project it is important to understand what stage the project is in as this will determine the level of project estimation accuracy required. For example, if the project is in the initiation stage, the project estimate may have an accuracy of ±50%.

This means that if  the project estimate is $2,000,000 with an estimating accuracy of ±50%, the business needs to allow for $1,000,000 at the lower end of the estimate and $3,000,000 as the upper limit of the project estimate.

Project estimation is a process and it is likely that the estimate will undergo several iterations and development. As the project scope is understood and refined, the project estimation accuracy should also improve.

A project estimate in the planning phase may reduce form ±50% to ±30%. Prior to executing the project and providing sufficient project planning was conducted, the project estimate may be as accurate as ±10%.

Project Estimation Techniques

It can be extremely challenging to estimate a project, as a project by definition is unique in nature, often a new product, service or business change. But there are several techniques available that can greatly improve the project estimation process and accuracy which include:

Expert Judgment

Project estimation requires a level of expert judgement, not only in the project itself but more importantly expert judgement of the business environment in which the project is being delivered.

It is typically the role of the project manager to provide this expert judgement in order to decide what business risks need to be managed and which project estimating technique is likely to achieve the best results. Some areas of consideration are:

• Impact of inflation over the project lifecycle
• Foreign exchange risk when using international suppliers
• Can hedging or insurance reduce the exposure to the business
• Required labour and labour rates
• Material costs and assumptions
• Costs of quality
• Availability of historical information
• Level of project estimation accuracy required

Analogous Estimating

Analogous estimates are a great technique for early project estimation providing there is sufficient historical data for which assumptions of similarity in size and complexity can be derived.

For example, the project may be to implement a CRM system into an 80 person national company. Knowledge of past projects of similar complexity have taken 5 months to complete for a 40 person company.

With this information you might infer that the complexity is similar but the scale of the project is double and an Analogous project estimation would suggest this could take 10 months.

As with most project estimation techniques, each have their own advantages and limitations, some key considerations for Analogous project estimating are:

• How does the project differ from that of the historical data
• What adjustment can be made to the formula for those differences
• How accurate is the historical data
• Is the data source internal or external to the organisation
• What external factors have not been accounted for
• Can the estimate be verified by an external expert
• Analogous estimates offer a low level of accuracy at a low cost

Parametric Estimating

Parametric project estimation is similar to analogous estimating but provides an increased level of accuracy due to the statistical nature of the estimating technique. This estimating technique is often based on average known rates, such as square meterage for construction or software lines of code for software projects etc.

For example if a typical residential house usually costs around $1,000 per square meter, a 400 square meter house can be estimated to cost around $400,000.

Estimating a software project that is expected to have around 20,000 lines of code and the cost per line of code is $3. We can estimate that the project is likely to cost $60,000.

This project estimation technique is highly dependent on the quality of the data source and the knowledge of the project specifics, some key considerations are:

• How aligned is the quality requirement to historical data
• Age of historical data
• Understanding of the project specification
• Accuracy of historical data
• Datasource (internal or external)

Bottom-Up Estimating

Bottom-up project estimation breaks down and estimates each component of a project. Each key portion or work package is split and broken down to greater levels of detail. The individual costs of each work package is then totalled to form the full bottom-up project estimate.

The bottom-up estimate takes the longest amount of time to prepare, but providing sufficient component detail is calculated this project estimation technique can provide the highest level of accuracy. Some considerations are:

• Estimate accuracy relies on level of detail known
• Can take considerable amount of time for complex projects
• Can take considerable amount of money
• May not be ideal during early initiation due to the cost of this estimation technique.

Three-Point Estimates

The three-point estimate is one of my favourite project estimation techniques because it can not only greatly increase the project estimates, but it’s approach also makes it easier for other experts to provide input.

The three-point estimate, also known as the PERT technique, provides a range of project estimates and calculates the weighted average of that range. In order to use the PERT project estimation technique, we provide 3 data points, the “best case”, “most likely case” and the “worst case”:

• Best Case (O) – What is the best case scenario to build X
• Most Likely Case (L) – What is the best case scenario to build X
• Worst Case (P) – What is the worst case scenario to build X

This technique works great for several reasons.

• Psychologically it is easier to provide a number when you can provide a wide range
• Starting with the worst case often leads to less resistance
• Once worst case and best case is identified, it becomes easier to provide the most likely case
• Reduces the natural instinct to inflate estimates

I will generally ask experts to first provide their worst case estimate and then the best case estimate. Once these 2 points are agreed, It is easier for them to determine the most likely case knowing their upper and lower estimates.

Once these 3 points of data are captured, the formula (O+4L+P)/6 can be applied to calculate the weighted average estimate.

If we go back to our lines of code estimate above a three-point PERT estimate might look something like this:

We can see that buy using the PERT technique in our project estimation we now have a weighted average software lines of code, which should be more accurate than our initial estimate of 20,000 lines of code.

Depending on the level of certainty, the weightings of the average can be adjusted to suit.

Download the PERT estimating template below ↓

Vendor Bid Analysis

If there is a specific element of the project (or all of the project) to be outsourced and procuring an external vendor to deliver the project on your behalf, a simple and effective way in determining project estimates is to have these external suppliers conduct it on your behalf.

This will likely be through a competitive bidding process, where multiple vendors will provide a proposal including a fixed price and required time frames for completion.

Once all of the vendor bids have been returned, an evaluation can be made on each bid in order to determine the range of costs associated with each solution, taking into account any quality assumptions and documented exclusions.

Key aspects to remember in vendor bid analysis project estimation are:

• Bid quality relies on quality of scope documentation provided
• Requires knowledgable and capable vendors
• Price may be inflated due to unknown or risk transference
• May be difficult to determine individual component costs
• Does not improve internal knowledge for future project estimation

 Reserve Analysis

Finally, once the project estimation has been completed and the project has a defined budget, it is important to provide for the level of “uncertainty”.

As with any estimate, it is just that, an estimate. It should not be expected to be 100% accurate and only required to be as accurate as the rough order of magnitude that was used, i.e. ±15% for the known components.

An estimate will inevitably include some element of uncertainty and due to the uniqueness of a project, will carry certain risks. The project estimation process should account for this uncertainty.

Uncertainty is often managed with a “contingency” budget, contingency is used where there is 20% of uncertainty surrounding the scope, a reasonable estimate may be to allow 20% contingency for the grey areas.

Combine Multiple Project Estimation Techniques

No single project estimation technique will suit all projects and for best results I suggest using and combining all or many of these estimating techniques to provide increased accuracy to your project estimates.

I like to take it one step further and use a three-point (PERT) estimate for all project estimation techniques, applying a PERT weighted average for everything! This is another key reason the three-point is my favourite project estimation technique.

Have a project estimation technique you think should be in this list? Let me know in the comments below.

Don’t forget to follow us and share these articles using the social link below ↓