What is IT project valuation (estimation) and what are the popular methods?

Definition of IT Project Valuation (Estimation)

IT project estimation is the process of predicting the amount of work, time, and/or cost required to complete a specific scope of an IT project or part of it (for example, a particular feature, module, or user story). The purpose of estimation is to provide the information needed for planning, decision-making (such as project viability assessment and task prioritization), resource allocation, and managing stakeholder expectations.

It must be emphasized that estimation is inherently uncertain and represents a forecast, not a guarantee. The quality of an estimate depends on the available information, the experience of the estimating team, and the complexity of the project. Good estimates express this uncertainty explicitly, for example through confidence intervals, ranges, or probabilistic forecasts rather than single-point numbers.

The Importance of Estimation in Project Management

Despite its inherent uncertainty, estimation serves critical functions in IT project management:

• Planning and scheduling: Defining the timeframe of the project and its various phases or iterations. Without estimates, realistic project planning is impossible.
• Scope management: A better understanding of the complexity and effort required for individual scope elements enables informed decisions about what to include, defer, or exclude from the project.
• Resource allocation: Planning the availability of team members and other resources needed to complete tasks efficiently.
• Decision-making: Evaluating the feasibility of a project, comparing different solution options, and prioritizing tasks with the highest value-to-effort ratio.
• Stakeholder communication: Managing expectations regarding project timelines and costs. Unrealistic expectations are among the most common causes of project dissatisfaction.
• Budget planning: The financial governance of IT projects relies heavily on effort estimates translated into cost projections.
• Risk identification: Estimates help identify areas of high uncertainty that require special attention and risk mitigation measures.

Challenges of Estimation

Estimating work in IT projects is difficult for several interconnected reasons:

• Incomplete or changing requirements: In early project phases, requirements are often vague or ambiguous. Even well-defined requirements can change as the project progresses and stakeholders gain clarity.
• Technical complexity: New technologies, unfamiliar integrations, or unpredictable technical challenges make effort assessment inherently difficult.
• Human factors: Differences in individual team member productivity, optimism bias (the tendency to underestimate effort), anchoring effects, and planning fallacy all influence estimates.
• External dependencies: Dependencies on third-party vendors, other teams, or external technologies are difficult to quantify and control.
• Project uniqueness: Every IT project has unique aspects that limit the applicability of historical data from previous projects.
• Scope creep: Uncontrolled growth of project scope can render even the most careful estimates obsolete.

Therefore, it is essential to approach estimates with humility, treat them as forecasts, and update them regularly as new knowledge is gained.

Popular Estimation Methods in Agile Approaches

Agile methodologies, where the emphasis is on adaptability and working in short iterations, use specific estimation techniques often based on relative complexity assessment rather than precise time prediction:

Story Points

Story Points are an abstract unit of measurement used to estimate the relative complexity, effort, and uncertainty involved in completing a given backlog item (such as a user story). Instead of estimating time in hours, the team compares tasks to each other and assigns points using a modified Fibonacci sequence: 1, 2, 3, 5, 8, 13, 21.

Advantages of Story Points:

• Abstract away from individual productivity differences
• Enable measurement of team velocity (the number of points completed per sprint)
• Focus the team on complexity rather than duration
• Are comparable across sprints for the same team

Limitation: Story Points are team-specific and cannot be meaningfully compared between teams. A 5-point story in Team A does not equate to a 5-point story in Team B.

Planning Poker

Planning Poker is a team estimation technique using Story Points. Each team member independently selects a card with a point value corresponding to their assessment of the task’s complexity. Cards are revealed simultaneously, and differences in ratings are discussed until the team reaches consensus.

This method promotes productive discussion, ensures that all perspectives are heard, and prevents anchoring bias since estimates are initially submitted privately. It is particularly effective for teams of 3 to 9 members working on well-defined user stories.

T-Shirt Sizing

T-shirt Sizing is a relational estimation technique where tasks are assigned to size categories (for example, XS, S, M, L, XL) instead of numeric points. This method is particularly useful for quick, high-level evaluation of large backlog items or for the initial assessment of a new project backlog. T-shirt sizes can later be mapped to Story Point ranges for more precise sprint planning.

Estimation by Analogy

Estimation by analogy involves comparing a new task to similar tasks completed in the past and assigning it a similar estimate. This method works best in teams with extensive experience and well-documented project history. It is fast and intuitive but relies heavily on the quality and relevance of the historical reference points.

Affinity Estimation

In affinity estimation, the team arranges a large number of backlog items on a relative scale by grouping items according to their perceived complexity. This technique is particularly efficient when many items need to be estimated simultaneously, such as during initial product backlog refinement.

Estimation Methods in Traditional Approaches

More traditional project management approaches (for example, Waterfall) employ methods that focus more directly on time and cost estimation:

Expert Estimation

Expert estimation relies on the judgment and experience of specialists in the relevant domain. This method is the quickest to apply but carries the risk of individual bias. Combining multiple expert opinions through the Delphi method can improve accuracy by iterating through rounds of anonymous estimation and discussion.

Parametric Estimation

Parametric estimation uses mathematical models and historical data to predict time or cost based on specific project parameters. For example, the number of screens to develop, database tables to create, or API endpoints to implement can serve as input parameters for a cost model. This method requires a reliable historical dataset and works best for projects similar to past ones.

Bottom-Up Estimation

Bottom-up estimation divides the project into small tasks, estimates each task separately, and then sums the individual estimates to produce an overall project estimate. This method is the most accurate but also the most time-consuming, requiring a detailed Work Breakdown Structure (WBS). It works best when the project scope is well defined and decomposable.

Three-Point Estimation (PERT)

For each task, three values are determined:

• Optimistic (O): Best-case scenario assuming everything goes smoothly
• Pessimistic (P): Worst-case scenario accounting for significant problems
• Most Likely (M): Most realistic scenario based on normal conditions

The weighted estimate is calculated using the PERT formula: (O + 4M + P) / 6

This method explicitly accounts for uncertainty and produces more realistic results than single-value estimates. The standard deviation ((P - O) / 6) provides a measure of estimate confidence.

Function Point Analysis

Function Point Analysis measures the scope of software based on the functionality delivered to the user. It quantifies inputs, outputs, queries, internal logical files, and external interface files, deriving an effort estimate from the resulting function point count. This method is particularly useful for comparing productivity across projects and organizations.

How ARDURA Consulting Supports IT Project Estimation

The quality of project estimates depends significantly on the experience and competence of the people involved. ARDURA Consulting provides experienced IT project managers, Scrum Masters, technical leads, and business analysts through its staff augmentation model who bring extensive estimation experience from diverse project contexts. With a network of over 500 senior IT professionals and a deployment time of two weeks, organizations can quickly access seasoned estimation experts. Their cross-project experience improves estimate accuracy and helps avoid common pitfalls such as optimism bias and anchoring effects. ARDURA Consulting’s specialists bring practical knowledge of which estimation methods work best for different project types and organizational contexts.

Estimation as a Continuous Process

Regardless of the method used, estimation should be treated as an ongoing process rather than a one-time activity. As the project progresses, new knowledge is gained, requirements change, and actual effort data provides valuable input for refining future estimates.

Proven practices for continuous estimation include:

• Regular re-estimation: Estimates should be updated in each iteration or when significant changes occur. Earlier estimates were based on less information and should be replaced by more informed assessments.
• Velocity tracking: In agile projects, measuring team velocity (Story Points completed per sprint) enables increasingly accurate predictions of delivery capacity over time.
• Retrospectives: Regular reflection on the accuracy of past estimates and the reasons for deviations continuously improves the team’s estimation capability.
• Transparent communication: The assumptions and uncertainties behind estimates must be openly communicated to stakeholders. Confidence intervals (“we estimate 4 to 6 weeks with 80 percent probability”) are more informative than single-point values.
• Historical data collection: Systematic recording of actual effort and comparing it to estimates builds a valuable data foundation for future projects.
• Cone of Uncertainty: Recognize that estimates become more accurate as the project progresses. Early-stage estimates may have uncertainty ranges of plus or minus 60 percent, while mid-project estimates narrow to plus or minus 15 percent.

Common Estimation Anti-Patterns

Organizations should be aware of and avoid these common estimation mistakes:

• Treating estimates as commitments: Estimates are forecasts, not promises. When estimates become targets that teams are held accountable for, they inflate estimates defensively, reducing their usefulness as planning tools.
• Estimating under pressure: Stakeholders who push for lower estimates do not change reality, only the accuracy of the forecast. Honest estimates serve the organization better than optimistic ones.
• Ignoring uncertainty: Single-point estimates create a false sense of precision. Ranges and confidence levels communicate reality more effectively.
• Not updating estimates: An estimate made three months ago with limited information should not govern decisions today when much more is known.
• Estimating in isolation: Estimates produced by individuals are generally less accurate than those produced by the team that will actually do the work.

Summary

IT project estimation is an indispensable, though challenging, component of project management. It supports planning, decision-making, resource allocation, and expectation management throughout the project lifecycle. In agile approaches, relative complexity estimation techniques such as Story Points and Planning Poker are widely adopted, while traditional approaches more commonly use time- and cost-based methods such as bottom-up estimation, PERT, and parametric modeling. The key to effective estimation lies in treating estimates as forecasts, communicating their uncertainty transparently, and continuously refining them based on accumulated experience. Organizations that approach estimation professionally, staff the process with experienced practitioners, and learn systematically from past projects make better decisions and deliver projects more reliably.