Views: 17
Summary – When the scope remains vague, a ROM estimate structures assumptions and scope, defining a budget range (−25/+75 %) with confidence levels (P50, P80) to inform the go/no-go decision. It combines top-down analogies, parametric models, three-point estimation and Monte Carlo simulations to map risks and safety margins. Versioned at each milestone, it aligns CAPEX/OPEX and supports governance up to the EVM baseline.
Solution: deploy an iterative, traceable ROM to steer your investments with transparency and agility.
The Rough Order of Magnitude (ROM) estimation provides, from the earliest phases of an IT project, a budgetary and time range that is sufficiently reliable to make a go/no-go decision. It does not claim to deliver a final cost estimate but offers a strategic overview with typical bounds of –25%/+75%, accompanied by confidence levels (P50, P80) and a register of assumptions. In a context where the scope remains partial, this approach highlights key cost drivers, inclusions and exclusions, and plans the refinement toward a more precise estimate.
This article details how to combine three estimation methods, version your results, and integrate ROM into your project portfolio governance.
Understanding ROM Estimation: Objectives and Principles
ROM estimation provides a preliminary budgetary range to inform the Go/No-Go decision. It structures assumptions, defines inclusions and exclusions, and associates a confidence level.
Definition and Purpose
ROM estimation aims to produce an initial view of costs and timelines without a fully defined scope. It sits upstream of a detailed costing exercise and addresses the need to secure a provisional budget, guide portfolio prioritization, and prepare a coherent IT business case.
This approach emphasizes transparency: each numeric value stems from a documented assumption, an analogy, or a parametric model. The lack of extreme precision is compensated by a wide range that reflects the intrinsic uncertainty of a scope yet to be defined.
Beyond simple cost, ROM guides IT governance and facilitates communication with the executive committee or CFO by providing a basis to discuss financial trade-offs and business priorities.
Budget Range and Confidence Level
The ROM range is often defined between –25% and +75%, but can be adjusted according to project maturity: ERP project, IT modernization, cloud migration, or bespoke application development.
The confidence level (P50, P80, or P90) indicates the probability that actual costs will fall within the estimated range. A P80 means that 80% of the modeled scenarios fall within this range; the higher the uncertainty, the more the upper bound includes a safety margin.
Clearly defining these indicators builds stakeholder confidence and frames future refinement, avoiding later debates on the validity of the initial estimate.
Assumptions, Inclusions, and Exclusions
A ROM estimation relies on an assumptions register: available resources, daily rates, technological maturity, external factors. Each assumption must be traced to justify the estimate’s scope.
Identifying inclusions and exclusions stabilizes the baseline: cloud infrastructure, licenses, maintenance, training, support, bespoke development, third-party integrations. The excluded scope (e.g., O365 licenses, third-party managed services, migration of specific legacy modules) must be explicit.
This level of detail prevents misunderstandings and eases the transition to a detailed budget estimate by listing what remains to be explored. It also sets the stage for a high-level work breakdown structure (WBS) and cost-schedule baseline.
For example, a manufacturing group requested a ROM for redesigning its internal portal. The initial exclusions of document management modules reduced the lower bound by 30% and enabled the investment committee to commit based on these estimates rather than conducting an in-depth audit.
Combined Methodologies for a Defensible ROM Estimation
Combining top-down analogies, parametric models, and a three-point estimate strengthens ROM’s robustness. Each method offers a complementary perspective and limits biases.
Analogous Estimation (Top-Down)
The top-down approach relies on similar past projects, adjusting costs based on complexity, functional size, or duration. It provides a quick overview without detailing every component and suits early phases with limited information.
Analogies require a reliable reference database from internal or industry experiences. Selecting comparable projects must consider organizational context, technological maturity, and security or compliance requirements.
The limitations of this method lie in project variability and the difficulty of finding perfectly aligned references. That’s why it’s always supplemented by other techniques.
Parametric Estimation (CER and Unit Rates)
The parametric model uses Cost Estimating Relationships (CER), linking cost to factors such as the number of features, story points, or KLOC. Each parameter is assigned a unit rate (cost per function point, cost per story point) based on benchmarks.
These formulas allow a quick recalculation of a range by adjusting key metrics: number of modules, interfaces to develop, test scenarios. They often rely on open-source or industry reference databases, ensuring a solid comparison base.
By combining CER and unit rates, the estimate incorporates volume considerations, offsetting the sometimes approximate nature of the top-down analogy.
For example, an SME in the financial sector applied a parametric model based on cost per function point to estimate a client portal implementation. This calculation revealed a 20% underestimation compared to the initial analogy. The discrepancy highlighted a regulatory complexity risk factor and allowed adjusting the ROM before validation committees.
Three-Point Estimation and PERT Analysis
The PERT or three-point approach uses optimistic, pessimistic, and most likely scenarios to calculate a weighted expectation. It formally incorporates parameter variability and generates a probability distribution.
The PERT formula (optimistic + 4× most likely + pessimistic) / 6 provides a central value, while the distribution can be simulated via Monte Carlo to estimate P50, P80, or P90 levels. This method ensures risks are neither underestimated nor the upper bound overloaded.
It is particularly useful when validated historical metrics are available, but even in highly uncertain contexts, it structures the analysis of variances and safety margins.
Risk Analysis and Monte Carlo Adjustment
Integrating a risk analysis allows adding targeted buffers for critical points (ERP integrations, compliance, data migration). Each risk can be assigned a probability and a business impact.
Monte Carlo simulation runs thousands of scenarios on identified parameters, generating cumulative curves that inform decision-making according to the desired confidence level. This avoids relying on a single midpoint and demonstrates the estimate’s resilience.
Combined with other methods, it delivers a quantified, traceable, and defensible ROM during investment committees, justifying each buffer with a documented risk.
Edana: strategic digital partner in Switzerland
We support companies and organizations in their digital transformation
Versioning and Refinement Plan from ROM to Budget Estimate
Managing ROM through versioning traces the evolution of assumptions and costs. A progressive refinement plan ensures a smooth transition to a detailed budget estimate.
Versioning and Tracking Assumptions
Each initial ROM should be versioned in an estimate register, including date, author, validated scope, and list of assumptions. Successive updates reflect the evolving scope definition and business feedback.
The assumptions log retains the history of changes: unit rate updates, integration of new modules, adjustments to internal or external resources. This traceability facilitates audits and bolsters credibility in committee.
An illustrative example: a public organization documented five versions of its initial ROM for an online service platform project, with each version specifying hosting, security, and support costs. This versioning demonstrated rigorous tracking to funders and secured progressive CAPEX funding.
Progressive Refinement Plan
Refinement schedules estimation milestones at each key project stage: functional specifications, technical specifications, prototype, testing. At each milestone, the ROM approaches a budgetary estimate and then a definitive estimate.
These milestones often align with PMO or PMBoK estimation reviews and are linked to specific deliverables (use cases, detailed WBS, test plan). They progressively reduce variance and ensure a seamless transition.
The success of this plan relies on the joint commitment of the IT department, PMO, and business owners to continuously validate adjustments and anticipate impacts on ROI.
Transition to Detailed Estimation
When the scope stabilizes, ROM gives way to detailed estimation by work packages, typically based on a fine-grained WBS, story points, and adjusted daily rates. This step incorporates the final architecture variants and definitive technology choices.
Detailed estimation consolidates CAPEX and OPEX, refines the business case, and prepares the cost-schedule baseline. It serves as a reference for Earned Value Management (PV, EV, AC) tracking.
At this stage, safety margins may decrease, test coverage is validated, and teams have a precise understanding of the remaining effort before deployment.
Integrating ROM into Governance Cycle
ROM becomes a Go/No-Go and prioritization tool, integrated into project portfolio management. It aligns CAPEX, OPEX, and key performance indicators.
Go/No-Go and Portfolio Prioritization
In the initial phase, ROM feeds steering committees to decide which projects to launch. Ranges are compared against available budgets and business objectives: expected ROI, time to market, compliance.
Prioritization relies on a cost/impact matrix where each ROM is weighed against functional gains, risks, and deployment time. This process guides the selection of flagship projects and quick wins.
It prevents pipeline overload and ensures alignment with overall strategy and IT department capacity.
Alignment with CAPEX and OPEX
ROM specifies the split between capital expenditures and operating expenses. License, development, and initial infrastructure costs are classified as CAPEX, while maintenance, support, updates, and hosting are allocated to OPEX.
This breakdown facilitates CFO approval in line with Swiss accounting rules and internal policies. It also prepares for periodic budget monitoring.
It ensures rigorous financial control, minimizes surprises, and supports multi-year investment planning.
Monitoring via Cost-Schedule Baselines
Once ROM is refined into a definitive estimate, the cost-schedule baseline becomes the reference for operational tracking. EVM dashboards compare Planned Value, Earned Value, and Actual Cost.
These metrics enable early detection of deviations, triggering corrective actions and measuring project performance. They can be enhanced with automated alert reporting.
Establishing a stable baseline ensures cost control and visibility into actual progress.
Lessons Learned and Continuous Improvement
After each project, the variance analysis between ROM and actual costs feeds a post-mortem. Deviations over 20% undergo thorough review: unmet assumptions, external factors, scope creep.
This process progressively improves CER accuracy, refines unit rates, and enriches the analogy database. Teams gain maturity and can reduce future ROM variances.
A large public services group implemented this mechanism and saw a 15% reduction in its ROM upper bound over two years, boosting the reliability of budgetary trade-offs and sponsor confidence.
From ROM to Agile, Transparent Budget Management
ROM estimation is the foundation of a progressive, traceable, and defensible costing approach. By combining analogies, parametric models, PERT, and risk analyses, then versioning each step, organizations secure go/no-go decisions and prepare a final budget aligned with business and financial realities.
Integrating ROM into governance, from the steering committee to EVM baselines, ensures transparency, agility, and resource optimization. Investment committees can make trade-offs with confidence, and the IT department gains a structured framework for portfolio management.
