M3.1 — Cost Estimating & Basis of Estimate

M3.1 — Cost Estimating & Basis of Estimate

1. Practical outcome

By the end of this section, the learner should be able to turn authorised scope and available project information into a traceable cost estimate whose quantities, rates, productivity assumptions, methods, exclusions and uncertainty can be understood, challenged and reproduced. They should be able to select estimating methods that are proportionate to the maturity of the scope, build and reconcile estimate components through the project control structure, use quotations, historical data and benchmarks intelligently, test the estimate for reasonableness, and document the Basis of Estimate (BoE) so that another competent practitioner can understand not only the number but how and why it was produced.

The objective is not to produce the most detailed spreadsheet or the most precise-looking total. It is to create a reasoned prediction of cost that is appropriate to the decision being made and transparent about what is known, what is assumed and what remains uncertain. A cost estimate should be capable of changing as project definition improves without losing the history of its basis.

A useful question throughout this Knowledge Area is: “If another competent estimator received only the scope, source data and Basis of Estimate, could they reconstruct why this estimate is credible?”

2. Core concepts

An estimate is a prediction; a budget is an authorisation

A cost estimate is a reasoned prediction of the cost required to deliver defined scope under stated assumptions and conditions. A budget is the authorised financial control position created from, or informed by, an approved estimate. They are related but not interchangeable. M3.1 focuses on estimating the expected cost of scope. M3.2 addresses how approved cost is converted into a controlled budget and cost baseline.

This distinction matters because an estimate can legitimately change as definition, market information or delivery strategy improves, while an approved budget is governed through authorisation and change control. Treating the budget as the answer the estimate must reproduce encourages target-driven estimating rather than evidence-based estimating.

Estimate maturity should reflect scope maturity

Early in a project, definition may be limited and an estimate may rely heavily on analogous projects, parametric relationships, broad quantities and allowances. As scope becomes better defined, the estimate should progressively use more specific quantities, resource plans, supplier information and bottom-up build-ups. The appropriate estimating method therefore depends on the maturity of the decision and the information available.

False precision is a warning sign. A highly detailed estimate built on immature scope can look more reliable than the underlying evidence justifies. Precision in the arithmetic should not be confused with confidence in the outcome.

Different estimating methods answer different needs

Most real project estimates are hybrids. Different cost elements can use different methods provided the choice is explicit and the parts reconcile to the same scope.

Quantities, productivity and rates must tell one coherent story

A transparent estimate separates the principal drivers. For quantity-based work, cost commonly starts with quantity multiplied by unit rate. For resource-based work, the practitioner may first convert quantity into effort using an expected productivity rate and then multiply the required effort by the relevant resource rate. The estimate should make these relationships visible rather than burying them inside one composite number.

For example: labour effort = quantity ÷ productivity; labour cost = labour effort × crew-day or person-hour rate. If productivity changes, the estimate should respond logically. This is one reason the resource and productivity assumptions in M2.3 need to reconcile with the cost estimate.

Cost = Quantity × Unit Rate is useful only when the quantity, rate and scope basis are all defined consistently.

Direct cost and indirect cost should remain distinguishable

Direct costs can be associated reasonably with a specific deliverable or work package, such as equipment, installation labour or specialist testing. Indirect or project-wide costs support delivery but are not naturally attributable to a single installed item, such as some site management, temporary facilities, project support or shared logistics. Organisations classify these differently, so the important point is to define the rule and avoid omission or double counting.

An estimator should also distinguish the cost of defined scope from contingency, management reserve or other uncertainty provisions governed elsewhere. M3.1 should expose uncertainty; it should not hide it by quietly increasing rates or durations.

The price basis matters

An estimate is meaningful only when its price basis is understood. The BoE should state the estimate date or price level and identify material treatment of escalation, inflation, currency, location, taxes, duties or other factors where relevant. Two estimates for identical physical scope can differ materially because they use different price dates, market conditions or currency assumptions.

Label constant-price estimates, nominal/outturn estimates and discounted present values distinctly. Constant-price values use a stated price date; nominal values include expected price movement over delivery; present values additionally reflect a discounting convention for appraisal. Reconcile to one basis before comparing totals or using them as a performance reference.

Escalation should reflect when cost is expected to be incurred, not simply be applied as an unexplained percentage to the total. Where the schedule changes materially, the estimator should consider whether the price basis or escalation treatment also needs to change.

Historical data and benchmarks are challenge tools, not automatic answers

Historical actuals can be powerful because they show what comparable work really cost. They need normalisation for differences in scope, quantity, productivity, location, market conditions, price date, contract strategy and delivery environment. A benchmark that has not been normalised can create more confidence than information.

Use benchmarks to ask “Why is this estimate different?” rather than “How do we force this estimate to match the benchmark?” A justified difference is acceptable; an unexplained difference deserves investigation.

Supplier quotations need to be read as scope documents as well as prices

A supplier quotation can provide a strong market signal, but the quoted number is only useful when the practitioner understands what is included, excluded, optional, conditional or time-limited. Freight, installation, testing, warranties, training, taxes, design support, currency exposure or interface work may sit outside the headline price. Commercial terms can also affect the timing of cost without changing the physical scope.

The Basis of Estimate is part of the estimate, not a cover note

A Basis of Estimate records the logic behind the number. It should allow a reviewer to understand the scope, estimating purpose, maturity, methods, quantities, rate sources, productivity, price basis, inclusions, exclusions, assumptions, interfaces, uncertainty, validation checks and review status. It should be concise enough to use but detailed enough to reconstruct the important reasoning.

A strong BoE normally records at least:

·       the estimate purpose, decision and effective date;

·       scope included, scope excluded and important interfaces;

·       WBS/CBS mapping and estimate ownership;

·       definition maturity and estimating methods used by major cost element;

·       quantity sources, rate sources and quotation references;

·       productivity, resource and schedule assumptions;

·       price basis, escalation, currency and location treatment where relevant;

·       allowances, uncertainty and risk treatment, including what is deliberately not included;

·       benchmark, reconciliation and sensitivity checks;

·       review, approval, version and change history.

Uncertainty should be visible rather than disguised

Every estimate contains uncertainty because quantities, rates, productivity, market conditions and scope maturity are imperfectly known. The correct response is not to make the estimate artificially exact or to bury contingency inside every rate. Identify the major uncertainty drivers, test their sensitivity and connect material risks to the risk process. Quantitative risk analysis and contingency are developed more fully in M4.3 and M4.5.

Where the organisation uses estimate ranges or maturity classifications, those should be defined consistently. The Open Standard does not require one proprietary classification system; it requires the practitioner to communicate the relationship between scope maturity, method and confidence honestly.

Where an estimate classification is adopted, name its source, edition and industry application. Do not treat a class label as a guaranteed accuracy range or apply another sector’s ranges automatically. State the actual scope maturity, estimating basis and uncertainty evidence.

Estimate quality is more than arithmetic accuracy

An estimate can add perfectly and still be wrong for the project. Quality requires scope completeness, appropriate methods, credible quantities and rates, consistent assumptions, transparent uncertainty, traceable data and independent challenge. Arithmetic checking is necessary, but it is only one part of estimate assurance.

3. Step-by-step method — how to develop a credible cost estimate

1. Define the estimating purpose and decision. Clarify whether the estimate supports option selection, business case, funding, procurement, baseline approval or change. State the lifecycle boundary: delivery to completion, or a wider period including operation, maintenance and disposal. Record inclusions, exclusions, estimate date, detail and decision timescale. A delivery-cost estimate should not be presented as whole-life cost.

2. Confirm the authorised or assumed scope boundary. Start from the WBS and scope information established in M1.1. Identify inclusions, exclusions, interfaces, owner-supplied items, third-party work and any scope that remains provisional. Do not estimate an undefined boundary without making the assumption visible.

3. Assess definition maturity. Determine what is actually known about quantities, design, delivery method, procurement route, schedule, locations and interfaces. Use that maturity assessment to decide how much detail the estimate can credibly support.

4. Structure the estimate for traceability. Map estimate lines to the WBS and relevant CBS/control structures from M1.2. The structure should allow the estimate to reconcile to scope and later support budget, commitments, actuals, forecasting and change.

5. Select an estimating method for each material cost element. Choose analogous, parametric, unit-rate, resource-based, quotation, bottom-up or another defensible method according to the available evidence. A hybrid estimate is normal; document why each method is appropriate.

6. Establish the quantity basis. Obtain quantities from controlled drawings, models, schedules, equipment lists, take-offs or defined assumptions. Record source, revision, unit of measure and any allowance for incomplete definition. Reconcile major quantities to the scope boundary.

7. Establish productivity and resource assumptions. For labour- or plant-driven work, define crew composition, production rate, working calendar, access conditions, mobilisation and expected inefficiency. Reconcile these assumptions with M2.3 where resource and schedule feasibility have been assessed.

8. Establish the rate basis. Use current quotations, framework rates, payroll/resource rates, historical actuals, market intelligence or other credible sources. Define what each rate includes and excludes. Avoid combining rates from incompatible scope or price bases without adjustment.

9. Normalise historical and benchmark data. Adjust comparator data for relevant differences in scope, quantity, price date, location, productivity, market, procurement route and delivery conditions. Retain the normalisation logic so later reviewers can understand the comparison.

10. Incorporate supplier and commercial information carefully. Review quotation validity, optional items, exclusions, delivery terms, freight, taxes, currency, warranties and payment assumptions. Make clear whether the quotation is budgetary, firm, indicative or conditional. Do not treat a supplier total as a complete project estimate without checking interfaces.

11. Build the direct-cost estimate. Calculate the cost of defined deliverables using the selected methods. Keep quantities, rates, productivity and formulae visible enough that the estimate can be reproduced and sensitivity can be tested.

12. Include appropriate indirect and project-wide costs. Identify shared management, temporary facilities, logistics, project support, mobilisation or other indirect costs required by the delivery model. Apply the organisation’s classification rules and avoid double counting between packages.

13. Apply the price basis consistently. State the estimate currency and price date. Apply escalation, inflation, location or currency factors where relevant and on a basis consistent with when the cost is expected to occur. Document major assumptions and sources.

14. Identify uncertainty, allowances and risk treatment. Separate defined-scope cost from explicit allowances and uncertainty provisions. Identify the major drivers that could move the estimate. Connect material risks to the risk process and check that the same uncertainty is not included twice.

15. Reconcile the estimate to scope, schedule and control structures. Check that every material scope element has a cost home, no cost element is duplicated, and resource/productivity assumptions align with the schedule. Reconcile alternative WBS/CBS views to the same total where they should.

16. Validate and challenge the result. Perform arithmetic checks, independent quantity checks, benchmark comparisons, ratios, first-principles cross-checks and sensitivities to major assumptions. Investigate material differences rather than averaging them away.

17. Prepare the Basis of Estimate. Document the scope, methods, sources, assumptions, exclusions, price basis, uncertainty, interfaces, validation and review outcome. A reviewer should be able to understand why the estimate is credible without reverse-engineering the spreadsheet.

18. Review, approve, version and retain the estimate. Record who prepared, checked and approved the estimate, its effective date and the decision it supported. Preserve the model and source evidence so later budget development, forecast comparison and lessons learned can distinguish changed scope, changed assumptions and estimating error.

4. Worked Reference Project example — Community & Learning Centre

The Community & Learning Centre Reference Project has an approved project budget of 50 million CU, but M3.1 does not start by forcing package estimates to equal that budget. It asks what defined scope is expected to cost on the best available evidence. Consider WBS 1.4 Digital & Technology Systems, continuing the technology quantities and productivity logic introduced in M2.

Step A — define the estimating purpose and maturity

The project is preparing a package estimate before procurement approval. Technology design is sufficiently mature to identify the principal equipment, four installation zones and approximately 1,200 standard technology points, but final supplier selection and some room-level details remain open. The estimate is therefore detailed enough for unit-rate and resource-based methods in some areas, while equipment still relies heavily on a current budgetary supplier quotation.

The estimate date is Month 6 of the project and all values are stated in neutral Cost Units at the Month-6 price basis. The estimate is for WBS 1.4 only and is not yet the authorised M3.2 cost baseline.

Step B — define scope and interfaces

Included scope: technology hardware, local cabling/components assigned to WBS 1.4, installation labour, configuration, system testing, training/handover support, package logistics and project-specific indirect support.

Excluded scope: permanent building power, primary containment and building fabric works, which are provided under WBS 1.2 Site & Building. This exclusion matters because the supplier quotation refers to “installation-ready areas”; the BoE must state what that condition means so the same interface is not estimated twice or omitted entirely.

Step C — build the estimate from appropriate methods

The arithmetic is deliberately transparent. The important learning point is that each cost element has a different evidence basis and the BoE tells the reviewer which assumptions are strongest and which are still immature.

Step D — test productivity and schedule consistency

The installation estimate assumes 20 technology points per crew-day, consistent with the productivity basis used in M2.3. At 1,200 points, the estimate therefore contains 60 crew-days. With two qualified crews, that represents about 30 crew working days before allowing for mobilisation, area sequencing and access constraints. If the schedule assumed the same installation could be completed in 15 working days with one crew, the estimate and schedule would be telling different stories and one or both would require correction.

The learner should also test sensitivity. If productivity were only 17 points per crew-day, installation effort would rise to approximately 70.6 crew-days and labour cost to about 0.198m CU — roughly 0.030m CU above the current estimate. That movement is small relative to the equipment quotation but still important to understand because the same productivity assumption can affect both time and cost.

Step E — test the result against benchmarks and major assumptions

Assume a normalised internal benchmark for comparable technology packages is 3,000–3,250 CU per technology point on the same broad scope and price basis. The current 3.711m CU estimate equates to about 3,093 CU per point. That does not prove the estimate is correct, but it is a useful reasonableness check. If the estimate had been 1,500 or 5,000 CU per point, the practitioner would investigate the difference rather than simply accepting or forcing the result.

Three sensitivities illustrate where management attention should focus:

·       an 8% increase in the 2.200m CU hardware quotation would add 0.176m CU;

·       a reduction in installation productivity from 20 to 17 points per crew-day would add about 0.030m CU to installation labour and could also affect schedule duration;

·       a 10% increase in the 1,200-point quantity would add about 0.072m CU of components plus additional installation effort before considering any secondary effects.

These are sensitivity tests, not a substitute for risk analysis and not amounts to be added automatically to the estimate. Their purpose is to reveal which assumptions matter and where better information or mitigation would most improve confidence.

Step F — record the Basis of Estimate

The BoE records that the hardware quotation is budgetary and valid for 60 days; permanent power and primary containment are excluded to WBS 1.2; installation productivity assumes released, accessible areas and two qualified crews; quantities are based on the current technology-point schedule; escalation is applied only to identified future material/equipment exposure; and project-level contingency or management reserve is not included in the 3.711m CU total.

The estimate can now be reviewed, challenged and, if approved, used as an input to M3.2 Budgeting & Cost Baselines. If supplier quotations, design quantities or schedule assumptions change later, the project can explain why the estimate moved because the original basis has been preserved.

5. Interpretation and professional judgement

Cost estimating is not primarily a spreadsheet skill. The arithmetic is often straightforward; the professional challenge is deciding which evidence is relevant, what level of detail is justified, how uncertainty should be communicated and whether the estimate tells the same delivery story as scope, schedule, resources, risk and commercial information.

How much detail is enough?

Use enough detail to expose the material cost drivers, ownership and uncertainty that matter to the decision. Adding hundreds of lines does not improve an estimate if those lines all depend on the same uncertain quantity or rate. Conversely, a single lump sum can hide exclusions and interfaces that management needs to understand.

When is a supplier quotation stronger than an internal estimate?

A current, well-scoped quotation can be powerful evidence, but it should still be challenged for validity, completeness, commercial conditions and interfaces. A quote for equipment alone is not evidence for installation, integration or project support unless those items are explicitly included.

Should the estimate match the available budget?

No. The estimator should report the expected cost of the defined scope on the chosen basis. If the result exceeds the available budget, that is management information. Scope, delivery strategy, specification, risk response or funding may need to change; the estimate should not be manipulated to remove the problem.

How should incomplete design be treated?

Do not invent detailed quantities merely to make the estimate look mature. Use a proportionate method, identify the definition gap, use explicit allowances where governed, and state the assumption. As definition improves, replace assumptions with evidence and preserve the estimate history.

How should historical data be used?

Historical actuals are valuable when the practitioner understands what they represent. Normalise before comparing and retain the adjustments. A benchmark is evidence to challenge an estimate, not permission to ignore current project conditions.

How should productivity be challenged?

Productivity should reflect the expected working environment, not an ideal textbook rate. Consider crew composition, learning curve, access, congestion, shift patterns, rework, mobilisation and interfaces. Where current project performance later provides stronger evidence, forecasts should use that evidence rather than preserving the original estimate assumption.

What belongs in contingency?

The detailed governance of contingency sits with risk and budgeting disciplines, but the estimator must prevent uncertainty from disappearing inside the base cost. Identify what the base estimate includes, what explicit allowances exist and what is excluded for separate risk treatment. Double counting and hidden contingency both weaken control.

How precise should the estimate be?

Report enough precision to support the decision, but do not imply more certainty than the evidence provides. Rounding a 3.711m CU calculation may be appropriate in a working model while an executive paper could communicate 3.7m CU together with the key uncertainty drivers. The underlying calculation and the communicated confidence are different things.

Questions the practitioner should be able to answer

·       What decision is this estimate intended to support?

·       What exact scope is included and excluded?

·       How mature is the scope and is the estimating method appropriate to that maturity?

·       Which quantities and rates drive most of the total?

·       Where did those quantities and rates come from, and how current are they?

·       What productivity and resource assumptions sit behind labour or plant cost?

·       What price date, escalation, currency and location basis is being used?

·       Which supplier quotations are firm, budgetary, conditional or time-limited?

·       What historical data or benchmarks have been used and how were they normalised?

·       Which scope interfaces could create omission or double counting?

·       Which assumptions or uncertainties could materially move the result?

·       Does the estimate reconcile with the schedule and resource plan?

·       Can the total be reconciled through the WBS/CBS and to the underlying estimate lines?

·       What has been independently checked or challenged?

·       Could another competent practitioner reproduce the reasoning from the BoE?

A useful practitioner test is: “If the estimate changes next month, can I explain whether the movement came from scope, quantity, rate, productivity, market, schedule, uncertainty or correction — rather than simply saying the spreadsheet changed?”

6. Practice diagnostic

Use this table to challenge the quality of the approach or deliverable. The rows consolidate the most important recurring red flags and checks; they are prompts for professional judgement rather than a substitute for the method above.

7. Try-it-yourself practical exercise

Using the Community & Learning Centre Reference Project, develop a defensible estimate for one material package. The exercise should test estimating judgement, not merely arithmetic.

1. Choose either WBS 1.3 Furniture & Specialist Equipment, WBS 1.4 Digital & Technology Systems or WBS 1.5 External Areas & Access and define the specific package you are estimating.

2. Write a concise scope statement containing at least five inclusions, three exclusions and two interfaces with other WBS elements.

3. State the estimate purpose, estimate date/price basis and your assessment of scope maturity.

4. Create an estimate structure mapped to WBS/CBS with at least fifteen cost lines. Use at least three different estimating methods where appropriate rather than forcing one method onto the whole package.

5. For at least five lines, show the quantity source and calculation. For at least three labour- or plant-driven lines, show the productivity/resource logic.

6. Use at least two different rate sources, such as a quotation and a normalised historical rate. Record what each rate includes and excludes.

7. Introduce one supplier quotation and identify at least four questions you would ask before accepting it into the estimate.

8. Apply a clear price basis and, if relevant, show how escalation, currency or location effects have been handled.

9. Identify at least five material estimating assumptions and three uncertainties. Do not hide them inside arbitrary rate uplifts.

10. Perform at least three reasonableness checks: for example a benchmark ratio, a second-method cross-check, quantity reconciliation or comparison with the schedule/resource plan.

11. Perform sensitivities on two material assumptions and explain which one management should investigate first and why.

12. Introduce three deliberate defects: one duplicated scope item, one outdated/unadjusted historical rate and one productivity assumption inconsistent with the schedule. Find and correct them during an estimate review.

13. Prepare a one-to-two-page Basis of Estimate summarising scope, method, quantities, rates, productivity, price basis, assumptions, exclusions, interfaces, uncertainty and validation.

14. Produce a short estimate review note stating whether the estimate is suitable for its intended decision and what information would most improve confidence at the next estimate update.

What good performance looks like

The exercise is successful when the learner can defend the estimate rather than merely produce a total. They should be able to trace cost to scope, justify why different methods were chosen, explain the important quantity/rate/productivity assumptions, show how the estimate was checked, distinguish defined-scope cost from uncertainty, and explain what evidence would cause the estimate to move. A complex spreadsheet with an undocumented basis is not sufficient.

8. Bridge to capability

The How-to teaches how scope and evidence become a cost estimate and Basis of Estimate. At APPLIED and above, capability assessment should test whether the practitioner can select suitable methods, build or review a realistic estimate and defend its assumptions. Traceable calculations, source data, exclusions and uncertainty should substantiate the estimate. Supporting work products are illustrative and should be assessed with the practitioner’s reasoning. Sample learning questions illustrate the judgement at each recognition level; they are discussion prompts rather than validated assessment items.

Learning question: Do you understand how scope and evidence support a cost estimate?

Capability question for APPLIED and above: Can you demonstrate a traceable estimate, justify its methods and defend its assumptions and uncertainty?