Mastering the Art of Project Estimating: A Guide to Key Techniques for Accurate Forecasting

Estimating is a critical component of project management. Accurate estimates of cost, schedule, resource needs, and other factors are essential for properly planning and managing projects. There are several techniques commonly used by project managers to estimate various aspects of projects. Selecting the right techniques and using them effectively is key to developing realistic plans and managing projects successfully.

What is Project Estimation?

Project estimation is the process of approximating or calculating the time, effort, cost, and resources needed to complete project activities and deliverables. Good estimates provide the foundation for effective project planning and execution. Estimates help project managers determine required budgets, develop reliable schedules, and plan the work realistically.

The goal of project estimating is to allocate the right amounts of time, human resources, materials, and budgets to planned project tasks. Accurate estimates support the project team in meeting objectives on time and within budget. Estimating, however, is not an exact science. There are always unknowns and uncertainties to account for. Project managers use different techniques to develop estimates that consider risks and include contingencies.

Why Estimate on Projects?

There are several key reasons that estimation is so important in project management:

• Support planning and scheduling - Accurate estimates of task durations, resource needs, and costs provide the basis for creating realistic project schedules and budgets.
• Identify required resources - Good estimates help determine staffing, equipment, materials, and other resources required to complete project work. Resource allocation and acquisition can then be proactively addressed.
• Manage stakeholder expectations - Data-driven estimates help set realistic expectations with project sponsors and stakeholders about budgets, timelines, deliverables.
• Reduce uncertainty and risks - Estimating prompts detailed analysis of project requirements and evaluation of uncertainties that could become risks. Contingencies can address these.
• Enable effective monitoring and control - Comparing actual performance data to estimates helps identify variances from plans requiring corrective actions to keep projects on track.

While absolute precision is usually impossible on initial estimates, project teams can employ different useful techniques to develop estimates within an acceptable margin of error for planning purposes. As more information becomes available, estimates can be refined and updated. Now let’s examine popular estimation approaches.

There are many different types of project estimation techniques used in Project Management with various streams like Engineering, IT, Construction, Agriculture, Accounting, etc. A Project manager is often challenged to align mainly six project constraints - Scope, Time, Cost, Quality, Resources, and Risk in order to accurately estimate the project. The common questions that come into the mind of a project manager at the start of the project are–

• How much work is to be estimated (scope).
• How to estimate the project (techniques).
• How much time it will require to complete the project (Schedule).
• Who will be doing the project (resources)?
• What is the budget required to deliver the project (cost)?
• Any intermediary dependencies that may delay or impact the project (Risks).

The 3 Major Parts to Project Estimation

• Effort estimation
• Cost estimation
• Resource estimate

While accurate estimates are the basis of sound project planning, there are many techniques used as project management best practices in estimation as - Analogous estimation, Parametric estimation, Delphi method, 3 Point Estimate, Expert Judgment, Published Data Estimates, Vendor Bid Analysis, Reserve Analysis, Bottom-Up Analysis, and Simulation. Usually, during the early stages of a project life cycle, the project requirements are feebly known and less information is available to estimate the project. The initial estimate is drawn merely by assumptions knowing the scope at a high level, this is known as ‘Ball-park estimates’, a term very often used by project managers.

Types of Project Estimation Techniques

There are two broad categories of estimating methods used in project management:

1. Qualitative Techniques - These methods apply expert judgment, prior experience, or historical data to evaluate a project and approximate estimates. Qualitative approaches rely more on objective information than measurable data.
2. Quantitative Techniques - These methods calculate cost, schedule, and resource estimates mathematically. They rely on numeric analysis of measurable project data rather than subjective expertise alone.

While qualitative methods are faster and simpler, quantitative techniques provide more objectivity, accuracy, consistency. Project managers may first develop quick estimates using an informal approach. Detailed estimating later applied mathematical modeling to refine projections for plans. Commonly-used techniques from both categories are described next with examples.

Qualitative Estimation Techniques

1. Expert Judgment

The expert judgment technique leverages the knowledge and expertise of individuals based on their prior experience with similar projects. Subject matter experts, senior technical resources, and other seasoned specialists are consulted to provide estimation input.

Example: The project manager assembles a group of engineers who have worked on comparable projects to estimate the time and resources required to design and develop a new prototype product.

2. Delphi Technique

The Delphi technique collects independent estimates from multiple subject matter experts through an anonymous survey process. Estimates and assumptions from each respondent are shared with the group for additional feedback until a consensus is reached. Anonymity helps avoid biasing effects.

Example: A Delphi exercise is conducted asking ten IT managers to independently estimate the hours required to implement a system upgrade. After three survey rounds, the group reaches consensus on the most likely estimate.

3. Analogy Estimating

Analogy estimating draws upon actual costs and durations from previous, similar projects as the basis for estimating the current project. This method is most reliable when past projects closely align in size, scope, complexity, staffing, and other parameters.

Example: Historical actual costs and resource hours are used to estimate a new website development effort of comparable scale and technologies.

4. Three-Point Estimating

The three-point estimating technique establishes estimate ranges using three data points: most likely, optimistic, and pessimistic projections. The typical value is used for baseline planning. This technique accounts for uncertainty.

Example: An engineer estimates that designing a prototype robot will take 3 weeks optimistically, 6 weeks most likely, and 9 weeks if obstacles arise, for planning purposes using 6 weeks.

Quantitative Estimating Techniques

1. Bottom-up Estimating

This method involves estimating at lower work breakdown structure levels by having the engineers, programmers, technicians, or others who will perform the actual hands-on work estimate their own tasks. These detailed task-level estimates are then aggregated into a total project estimate.

Example: Software developers estimate the workdays required to complete individual code modules. The estimates for all modules are summed to provide the overall project timeline and budget.

2. Parametric Estimating

This mathematical technique uses historical data to identify relationships between program parameters (size, weight, duration) and cost or schedule drivers. Models are derived to calculate estimates by plugging in project-specific measurements.

Example: Historical construction data is used to develop a regression model correlating the square footage to cost per square foot, allowing estimates based on the size specifications.

3. Three-Point Estimating

As explained previously, the three-point estimation technique develops an approximate and contingency projection by leveraging the minimum, maximum, and most likely data points for a forecast. This method explicitly accounts for uncertainty.

Example: A contractor estimates best case, worst case, and most probable costs for a new building project, using the expected cost for budgeting and applying contingency funds based on the range between optimist and pessimist assumptions.

Let's check overview of the two subtypes of the Three Point Estimating technique with formulas and examples:

1. Triangular Distribution

The triangular distribution method uses the three estimation data points - minimum, maximum, likely (mode) - to calculate the average estimate. It assumes a triangular probability density function between the extremes and mode. This is a Linear Distribution and it not considers any risks. It gives the equal weight to the three-point estimates when you calculate the Expected Activity Duration or the costs.

Formula: Average Estimate = (Minimum Estimate + Maximum Estimate + Likely Estimate) / 3

Or

Expected Activity Duration (EAD) = (P + M + 0) / 3, here

• P = Pessimistic estimation?
• M = Most Likely estimation
• O = Optimistic estimation

Example:

Minimum estimate = 4 weeks

Maximum estimate = 10 weeks

Likely estimate = 7 weeks

Average estimate = (4 + 10 + 7) / 3 = 7 weeks

The triangular distribution better accounts for variability across the range than a single value. The actual duration may be anywhere between the defined minimum and maximum based on the assumed shape of the triangle.

2. Beta Distribution

The beta distribution method models the variability between estimates using a beta probability distribution. It employs the three data points to define the beta distribution shape parameters α and β. For the BETA Distribution, This terms are often associated : ‘Weight average‘ or ‘Historical datas‘, ‘Samples to work with‘.

This is a Non Linear Distribution and it considers the risks. It gives stronger consideration to the Most likely estimate (M) (also called PERT estimate) when you calculate the Expected Activity Duration or the costs.

PERT = Program Evaluation and Review Technique

Formulas:

Expected Activity Duration (EAD) = (P + 4M + O) / 6

Or

α = (Likely Estimate - Minimum Estimate)/(Maximum Estimate - Minimum Estimate)

β = α (Maximum Estimate - Likely Estimate)/(Likely Estimate - Minimum Estimate)

Average Estimate = (Minimum Estimate + 4 x Likely Estimate + Maximum Estimate)/6

Example:

Minimum estimate = $50,000

Maximum estimate = $100,000

Likely Estimate = $70,000

α = ($70K - $50K)/($100K - $50K) = 0.5

β = 0.5 x ($100K - $70K)/($70K - $50K) = 0.5

Average Estimate = ($50K + 4x$70K + $100K)/6 = $75,000

The beta distribution method is considered more robust than triangular since it uses a more precise statistical distribution rather than a basic triangular shape. But it involves more calculations.

What Is PERT?

PERT stands for Program Evaluation and Review Technique and was developed as an advanced project schedule planning and management system by the US navy in the 1950s

Another, not too serious, story of its origination was once published by an anonymous author in PMI’s journal

In PMI-style projects, PERT is primarily used as a supplemental technique to the Critical Path Method for scheduling activities. However, it can also be applied to stand-alone estimates of work items and activities.

The so-called PERT distribution leverages on the values determined with the three-point estimation technique. It can basically be used for all planning levels, ranging from activities to entire projects. However, finding the right granularity for meaningful estimating may require some critical and conceptual thinking.

The PERT method implies overweighting the ‘most likely’ estimate. It transforms the three-point estimate into a bell-shaped curve and allows to determine probabilities of ranges of expected values.

4. Reserve Analysis

Reserve analysis compares the amount of contingency funds or reserves set aside for cost overruns or schedule delays to the level of risk exposure on the project. More reserves may indicate excessive risk avoidance; inadequate reserves likely indicate understated risks.

Example: A project manager examines the 20% contingency budgeted for a software project relative to likely risks identified through detailed risk analysis to assess if reserves are sufficient or excessive.

5. What-If Analysis

This project estimation technique uses assumptions based on varying factors like scope, time, cost, resources, etc., to evaluate the possible outcomes of the project by doing impact analysis. In a usual scenario, the project estimate is done by conducting estimation workshops with the stakeholders of the project, senior team members who could give valuable inputs to the estimation exercise. The high-level scope is broken down into smaller work packages, components, and activities, each work package is estimated by effort and resources needed to complete the work package. The project may be detailed into the smallest chunk that can be measured.

Choosing Estimating Techniques

Project managers select estimation methods based on factors like project type, the phase of the product lifecycle, data availability, time constraints, and desired accuracy. A combination of techniques is often used to develop an appropriate, well-rounded forecast and contingency plan through triangulation. As actual performance data becomes available, successive re-estimating refines projections. By applying skillful and deliberate estimation practices, project managers deliver better plans, decisions, and outcomes.