Estimation, Scheduling, Cost, Resources and Procurement

Introduction

Hybrid project management integrates predictive planning with adaptive delivery. It is particularly useful when some project elements are sufficiently understood to justify baselines, contracts, and dependency-driven schedules, while other elements require experimentation, customer feedback, and iterative reprioritisation. Estimation, scheduling, cost, resources, and procurement therefore cannot be managed as isolated disciplines. A forecast of scope affects the work breakdown structure (WBS); the WBS affects resource demand; resource availability affects schedule feasibility; schedule choices affect cost; and procurement strategy affects both delivery cadence and risk.

The central managerial challenge is not to produce a deceptively precise plan. It is to create an integrated, revisable delivery model that is sufficiently reliable for governance and sufficiently flexible for learning. A hybrid plan commonly contains a predictive layer—business case, milestones, release constraints, budget envelopes, contractual commitments, and governance gates—and an adaptive layer—product backlog, relative estimates, iteration plans, capacity forecasts, and frequent replanning.

Learning objectives

By the end of this chapter, the reader should be able to:

1. Decompose scope through a WBS and connect it to a product backlog.
2. Select and apply analogous, parametric, bottom-up, three-point, and relative estimation techniques.
3. Distinguish effort, duration, cost, capacity, velocity, and contingency.
4. Construct and interpret a critical-path schedule while using rolling-wave planning for uncertain work.
5. Develop integrated resource and cost plans.
6. Select procurement and contract arrangements appropriate to uncertainty and risk.
7. Govern changes to estimates, schedules, budgets, and supplier commitments.

1. The integrated planning architecture

A hybrid delivery plan is best understood as a set of linked views rather than one master document. The principal views are:

The plan should preserve traceability. A major milestone should map to deliverables, deliverables to work packages or backlog items, work packages to resources and cost accounts, and supplier outputs to acceptance criteria. In adaptive delivery, traceability does not mean freezing every detail. It means preserving a visible relationship between strategic outcomes, near-term work, evidence of progress, and authorised expenditure.

A useful distinction is between a commitment forecast and a planning forecast. A commitment forecast is sufficiently reliable to support a contractual or executive commitment. A planning forecast is a current best estimate that may change as knowledge improves. Treating all forecasts as commitments creates either excessive buffers or repeated governance failure.

2. Decomposing scope: WBS and backlog integration

The WBS is a hierarchical decomposition of the total project scope into deliverables and manageable work packages. It is deliverable-oriented: it describes what the project must produce, not merely the departments or activities involved. A work package should be small enough to estimate, assign, schedule, monitor, and accept, while remaining large enough to avoid unnecessary administrative overhead.

In a hybrid project, the WBS and product backlog serve different but complementary purposes. The WBS provides a stable view of contractual and governance scope. The backlog provides a change-tolerant view of product functionality and learning. A typical relationship is:

• project outcome → major deliverable or release;
• deliverable → product feature or technical capability;
• feature → epics and user stories;
• enabling work → architecture, testing, security, data migration, training, and deployment tasks;
• work package → activities, estimates, resources, cost accounts, and acceptance evidence.

The relationship should not be forced into a one-to-one mapping. One feature may require work across several WBS branches, and one work package may enable several features. A traceability matrix or digital work-management system can preserve these links.

A strong WBS includes project management, quality, integration, transition, and closure work. Omitting these items produces an optimistic estimate even when the product scope appears complete. The 100 percent rule states that the WBS should include 100 percent of the work necessary to accomplish the approved scope, including project-management work, but should not include work outside that scope.

Decomposition should stop when further breakdown no longer improves control. Indicators that a work package is appropriately defined include clear completion criteria, an identifiable owner, a credible estimate, a manageable duration, and a measurable output. Excessive decomposition encourages local optimisation and administrative effort; insufficient decomposition conceals risk and creates weak estimates.

3. Estimation as decision-making under uncertainty

An estimate is not a promise and not a single objective fact. It is a reasoned forecast based on information, assumptions, models, and uncertainty. At postgraduate level, estimation should be evaluated by its basis, calibration, range, and use. An estimate suitable for feasibility analysis may be inappropriate for a supplier commitment.

3.1 Estimation dimensions

Several quantities are routinely confused:

• Effort: labour input, usually measured in person-hours or person-days.
• Duration: elapsed calendar time between start and finish.
• Cost: monetary expenditure associated with labour, materials, services, overhead, and reserves.
• Capacity: the amount of work a team can realistically accept in a period.
• Velocity: an empirical measure of completed relative work per iteration, often expressed in story points.
• Productivity: output per unit of input; it must be interpreted carefully because output definitions vary.

Effort does not convert mechanically into duration. A ten-person-day task may take ten working days for one specialist, but it will not necessarily take one day for ten people because of coordination, scarce skills, communication, setup, and dependencies. A simple staffing relationship is valid only under restrictive assumptions of divisible work, sufficient skills, no coordination loss, and no resource contention.

3.2 Bottom-up and analogous estimation

Bottom-up estimation estimates lower-level activities or work packages and aggregates them. It is useful when scope and methods are well understood. Its weaknesses are omission, false precision, and local optimism. Reviewers should ask whether integration, rework, testing, management, and learning have been included.

Analogous estimation uses actual results from a comparable project and adjusts for meaningful differences. It is fast and valuable early, but its reliability depends on the quality of the analogy. Similar project titles are not enough; comparison should consider scale, technology maturity, complexity, team capability, regulatory burden, and delivery model.

3.3 Parametric estimation

Parametric estimation uses a statistical or operational relationship between cost or effort and measurable drivers. A general linear form is:

where E is estimated effort, a is a fixed component, b is effort per unit, and X is the quantity of work units. For multiple drivers:

The model is applicable only when the drivers are defined consistently, historical data are relevant, and the relationship is sufficiently stable. Units must be consistent: if b is person-days per interface and X is interfaces, E is person-days. Regression fit does not establish causation, and extrapolation beyond the historical range is risky.

3.4 Three-point estimation

When uncertainty is material, three-point estimates represent optimistic, most likely, and pessimistic outcomes. A commonly used weighted mean is the PERT-style estimate:

where O is the optimistic estimate, M is the most likely estimate, P is the pessimistic estimate, and TE is the expected estimate. All three values must use the same unit and refer to the same quantity—for example, elapsed days or person-days. This formula assumes a beta-like distribution and gives greatest weight to the most likely case; it is not a guarantee of the actual result.

An associated approximate standard deviation is:

This approximation is useful for risk analysis when the three-point assumptions are credible. It should not be interpreted as a measured standard deviation unless the underlying distributional assumptions are justified.

Worked example. A data migration work package has an optimistic duration of 10 days, most likely duration of 16 days, and pessimistic duration of 28 days. The expected duration is:

TE = (10 + 4×16 + 28) ÷ 6 = 102 ÷ 6 = 17 days

The approximate standard deviation is σ = (28 − 10) ÷ 6 = 3 days. The estimate should be reported as approximately 17 days with assumptions and risks, not as a certain 17-day commitment.

3.5 Relative estimation and story points

Adaptive teams often estimate product backlog items using story points. A story point is a team-specific relative measure combining dimensions such as effort, complexity, uncertainty, and dependencies. It is not a standard unit of time and should not be converted into person-hours across teams.

Teams may use planning poker, affinity grouping, or a modified Fibonacci scale such as 1, 2, 3, 5, 8, 13, and 21. Larger values signal greater uncertainty and may indicate a need to split the item. A reference item establishes a shared meaning for the scale.

Velocity is calculated from completed, accepted work, not work started. If a team completes 24, 27, 25, and 28 points in four comparable iterations, its arithmetic mean velocity is:

The average is descriptive, not a promise. Forecasts should consider variation, holidays, onboarding, defects, operational work, and changes in team composition. A forecast of 130 remaining points at 26 points per iteration suggests five iterations only if scope, team capacity, definition of done, and velocity conditions remain sufficiently comparable.

4. Scheduling in a hybrid environment

Predictive scheduling represents activities, dependencies, durations, constraints, milestones, and calendars. Adaptive scheduling represents a prioritised queue of valuable work, a timebox, a definition of done, and a mechanism for selecting work based on capacity. Hybrid scheduling uses both: fixed milestones and external dependencies may be planned predictively, while detailed solution work is planned progressively.

4.1 Network logic and the critical path

The critical path is the longest-duration path through a logically connected network of activities. Activities on it have zero total float under the chosen network and calendar assumptions. Delaying a critical activity delays the calculated project finish unless schedule recovery or logic changes are introduced.

For an activity, the basic forward-pass relationships are:

where ES is early start, EF is early finish, and d is activity duration. For a finish-to-start relationship with zero lag, the successor’s early start is the predecessor’s early finish. The backward pass uses:

where LS is late start and LF is late finish. Total float is:

These equations assume a defined network, consistent calendars, and the specified dependency relationship. In real schedules, resource constraints, calendars, lags, and multiple critical paths complicate interpretation. Criticality should therefore be monitored, not treated as permanent.

Worked example. Consider four activities:

Path A–B–D equals 3 + 4 + 2 = 9 days. Path A–C–D equals 3 + 6 + 2 = 11 days. Therefore A–C–D is the critical path and the earliest calculated completion is day 11, assuming the activities are sequential as shown and resources are available.

4.2 Rolling-wave planning

Rolling-wave planning develops near-term work in detail and keeps later work at an appropriate level of abstraction. The near-term horizon may include task-level estimates, assigned resources, acceptance criteria, and dependency resolution. Later work may remain at feature, milestone, or capability level until discovery reduces uncertainty.

Rolling-wave planning is not uncontrolled postponement. It requires planning horizons, decision dates, entry criteria for elaboration, and a record of assumptions. A release plan may identify a regulatory milestone six months ahead while detailing only the next two iterations. As each iteration produces evidence, the plan is updated and the effect on cost, capacity, and milestone confidence is assessed.

4.3 Schedule compression and trade-offs

Schedule compression techniques include adding capacity, overlapping work, reducing scope, simplifying quality or functionality, and changing sequencing. Fast tracking overlaps activities that were previously sequential and increases coordination and rework risk. Crashing adds resources or expenditure to shorten duration and is effective only where additional resources can reduce the controlling work. Both techniques should be evaluated against quality, safety, regulatory, and sustainability constraints.

5. Capacity and resource planning

Capacity planning asks what the team can realistically deliver, whereas demand planning asks what work is desired. A simple capacity calculation is:

where C is available productive capacity in hours for a period, N is the number of people with the relevant capability, H is nominal working hours per person, and U is the usable-capacity factor between 0 and 1. U accounts for leave, meetings, support, administration, learning, and expected interruption. The formula assumes comparable skills and a common time period; it must be adjusted for part-time work and specialist bottlenecks.

Example. Six team members have 80 nominal hours in a two-week iteration. If the planning factor is 0.75, capacity is C = 6 × 80 × 0.75 = 360 hours. This is not necessarily 360 hours of one interchangeable skill. A team may have 240 development hours but only 40 testing hours, making testing the constraint.

Resource planning should identify:

• required roles and specialist skills;
• availability and time-zone constraints;
• internal, external, and shared resources;
• onboarding and knowledge-transfer time;
• single points of failure;
• decision authority and escalation paths;
• sustainable workload and psychological safety.

A resource histogram can reveal peaks and valleys, but averaging demand can hide bottlenecks. Skills matrices, scenario analysis, and constraint-based planning are often more useful than simple headcount ratios. In adaptive teams, stable membership supports empirical forecasting. Frequent movement of specialists between teams can increase apparent utilisation while reducing flow and accountability.

6. Cost estimating, budgeting, and control

A cost estimate predicts the resources required to perform authorised work. A budget is an authorised time-phased allocation of funds. The cost baseline is the approved, time-phased budget for measuring performance; management reserve is typically held outside the performance measurement baseline for unforeseen work within project scope, subject to governance. Contingency reserve is generally associated with identified risks and is included in the cost baseline when organisational policy defines it that way. Terminology varies by standard, so the project governance framework should define it explicitly.

A basic cost model is:

where each component is expressed in the same currency and R represents authorised reserves. Labour cost may be estimated as:

where hᵢ is hours for resource category i and rᵢ is the corresponding fully loaded hourly rate. The rate definition should state whether it includes salary, benefits, contractor margin, equipment, and overhead. Double counting must be avoided.

Earned value in hybrid projects

Earned value management can be applied to predictive work packages and selected release-level outcomes, but story points are not automatically monetary earned value. Common measures are:

Here, PV is planned value, EV is earned value, and AC is actual cost, all measured in the same currency and time-phased basis. Positive cost variance and schedule variance are favourable under these definitions. A CPI below 1 indicates less value earned per unit of cost; an SPI below 1 indicates less value earned than planned at the measurement date. In adaptive delivery, teams may use completed acceptance-based increments to define earned value, but the valuation rule must be established before performance is measured.

7. Procurement and hybrid contracting

Procurement strategy should allocate risk to the party best able to manage it without creating destructive incentives. The buyer should decide what must be specified, what can be discovered collaboratively, what market capability exists, and how acceptance will be evidenced.

Hybrid contracts commonly combine mechanisms. For example, discovery may be time and materials, a defined infrastructure component may be fixed price, and implementation may use a capped capacity-based arrangement with agreed quality and security criteria. A contract may also include a target cost, a ceiling, incentive measures, change thresholds, and termination or re-prioritisation rights.

A sound statement of work describes outcomes, interfaces, assumptions, dependencies, security and regulatory obligations, evidence of completion, and acceptance authority. For adaptive work, acceptance should focus on demonstrable increments and a transparent definition of done rather than prescribing every internal task. However, agility does not excuse ambiguity about intellectual property, data protection, confidentiality, accessibility, audit rights, subcontracting, service levels, or transition assistance.

Procurement governance should prevent conflicts of interest, favouritism, bid leakage, and inappropriate supplier dependence. Evaluation criteria should be published internally before selection, and commercial negotiations should preserve a defensible audit trail. AI-generated procurement content must be reviewed for confidentiality, bias, invented supplier claims, and unauthorised disclosure of commercially sensitive information.

8. Integrated planning and governance

An integrated baseline is useful only when it is coherent. The project manager should test at least four alignments:

1. Scope–estimate alignment: every authorised deliverable has an estimate and acceptance approach.
2. Estimate–resource alignment: estimates reflect actual skills, availability, calendars, and learning effects.
3. Schedule–cost alignment: time-phased resources and procurements match the budget profile.
4. Procurement–delivery alignment: supplier lead times, contract controls, and acceptance events fit the roadmap.

Change control in a hybrid project should be risk-based. A minor backlog reprioritisation within an authorised release may require product-owner approval only. A change affecting a regulatory milestone, contract ceiling, architecture baseline, or funding envelope requires broader governance. Decision thresholds should be explicit.

Useful indicators include forecast completion date, forecast cost, throughput, aging work, dependency health, defect escape, supplier milestone performance, resource utilisation, and risk exposure. Metrics should support decisions rather than reward local optimisation. For example, high utilisation can coexist with poor flow if teams are overloaded and work remains partially complete.

9. Illustrative case: Northstar public-service portal

Background

Northstar, a public agency, plans a digital portal for permit applications. Regulatory reporting and a fixed public launch date require predictive governance. The user experience, workflow design, and integration details are uncertain, so the delivery team proposes six two-week iterations within a four-month release window.

Problem

An initial estimate of 12,000 labour-hours appears affordable, but it excludes accessibility testing, data migration, supplier integration, and operational training. The proposed fixed-price contract also specifies detailed features before user research has occurred.

Analysis

The team creates a WBS with governance, discovery, portal capabilities, integration, migration, assurance, training, and transition branches. Features are represented in the backlog, while the WBS preserves accountability for non-feature work. The team estimates near-term stories using relative points and uses analogous data from two prior government portals for early release forecasting. A three-point estimate is used for data migration because legacy data quality is uncertain.

Capacity analysis identifies eight delivery members, but only one accessibility specialist is available half-time. Accessibility review is therefore a resource constraint and a schedule risk. The network schedule shows that security accreditation and migration rehearsal are on the controlling path to launch.

Approach and solution

The contract is divided into work streams. Discovery and iterative product development use a capped time-and-materials model with fortnightly acceptance of increments. A stable hosting component is procured through a fixed-price statement of work. The contract includes a ceiling, transparent backlog reporting, security obligations, intellectual-property provisions, and a right to reprioritise within the agreed outcome envelope.

The budget contains labour, supplier fees, testing, training, and an authorised risk reserve. The release forecast is updated after every iteration using completed accepted work and remaining scope. The steering group receives a decision report showing forecast date, cost range, top risks, and decisions required—not simply a colour-coded status.

Outcome

In the illustrative scenario, early accessibility involvement prevents late redesign. Migration rehearsal reveals data cleansing work, which is funded from the risk reserve and scheduled before final integration. Two low-value features are deferred without affecting the regulatory launch. The portal launches within the approved funding envelope, although one post-launch enhancement is moved to a later release.

Learning points

The case demonstrates that hybrid planning is not a compromise between two opposing methods. It is a deliberate allocation of planning precision: stable constraints receive firm control, while uncertain solution details remain empirically governed. The strongest controls are traceability, transparent assumptions, evidence-based acceptance, realistic capacity, and contract terms that preserve learning.

10. Challenges, limitations, and contemporary practice

Estimates can be biased by anchoring, optimism, strategic misrepresentation, pressure to meet a target, and the planning fallacy. Independent review, reference-class forecasting, historical calibration, explicit ranges, and separation of estimate from target can reduce these effects. A target is a desired outcome; an estimate is a prediction; a commitment is an authorised promise. They should not be conflated.

Hybrid environments also face measurement problems. Story points may be gamed if used as an individual productivity metric. Velocity comparisons across teams are invalid when scales, definitions of done, and work types differ. Earned value can become misleading if value is assigned to partially complete or low-value output. Supplier performance measures can encourage undesirable behaviour when they reward activity rather than usable outcomes.

Digital planning tools and AI can assist with dependency discovery, schedule scenario analysis, historical comparison, capacity visualisation, and anomaly detection. They do not eliminate the need for expert judgment. Data may be incomplete, historical projects may not be comparable, and automated recommendations may reproduce organisational bias. Human approval, explainable assumptions, access controls, data minimisation, and auditability are essential, especially where commercial or personal data are involved.

Best practice is therefore to:

• record the basis, range, confidence, and owner of every material estimate;
• maintain separate views for commitments and forecasts;
• use WBS and backlog traceability without pretending they are identical;
• plan near-term work in detail and later work progressively;
• model specialist capacity and dependencies, not merely headcount;
• time-phase budgets and reserves;
• select contracts according to uncertainty and controllable risk;
• define acceptance before work begins;
• review forecasts using evidence and calibrated historical data;
• govern changes according to impact, not bureaucracy alone.

Chapter Summary

Integrated planning in hybrid project management connects scope, estimation, schedule, cost, resources, procurement, and governance. The WBS provides a deliverable-oriented structure, while the backlog supports adaptive prioritisation and learning. Estimation methods should match the maturity and uncertainty of the work: analogous and parametric methods are useful early, bottom-up methods support defined scope, three-point estimates express uncertainty, and story points support team-relative forecasting.

Critical-path analysis identifies schedule sensitivity, while rolling-wave planning prevents premature detail. Capacity planning must account for usable time, specialist constraints, and sustainable workload. Cost plans should distinguish labour, procurement, overhead, and reserves, and performance measures must use consistent valuation rules. Procurement and contract models should allocate risk deliberately and preserve appropriate flexibility. The integrated delivery plan is strongest when it is transparent, traceable, empirically updated, and governed according to decision impact.

Key Terms

WBS: A hierarchical decomposition of total project scope into deliverables and work packages.

Work package: A manageable component of scope that can be estimated, assigned, scheduled, and accepted.

Backlog: A prioritised, evolving list of product and enabling work.

Analogous estimation: Forecasting based on comparable historical work.

Parametric estimation: Estimation using a relationship between cost or effort and measurable drivers.

Bottom-up estimation: Estimating detailed components and aggregating them.

Three-point estimation: Estimation using optimistic, most likely, and pessimistic values.

Story point: A team-relative measure of effort, complexity, uncertainty, and dependencies.

Velocity: Completed accepted relative work per iteration.

Capacity: Realistic work availability during a defined period.

Critical path: The longest logical path through a schedule network.

Float: Permissible schedule movement without affecting a defined constraint or finish date.

Rolling-wave planning: Detailed near-term planning with progressively elaborated future work.

Cost baseline: Approved, time-phased budget used for performance measurement.

Contingency reserve: Allowance associated with identified risks, subject to governance definitions.

Earned value: Budgeted value of work accepted or completed under an established valuation rule.

Fixed-price contract: An arrangement with an agreed price for defined scope and conditions.

Time-and-materials contract: An arrangement paying agreed rates for actual authorised effort or resources.

Acceptance criteria: Conditions that work must satisfy to be accepted.

Traceability: The maintained relationship between outcomes, scope, work, evidence, cost, and decisions.

Review Questions

Short Answer Questions

1. Why should effort and duration not be treated as interchangeable?
2. What is the purpose of the 100 percent rule in WBS development?
3. Define rolling-wave planning.
4. What does a velocity measure, and what does it not measure?
5. State two factors that should influence contract selection.

Long Answer Questions

1. Explain how a WBS and product backlog can be integrated without making them identical.
2. Compare analogous, parametric, bottom-up, three-point, and relative estimation.
3. Explain how critical-path analysis and rolling-wave planning complement one another.
4. Discuss the relationship among capacity, cost, schedule, and sustainable workload.
5. Evaluate the advantages and risks of hybrid contracting for uncertain digital products.

Analytical / Application Questions

1. A team has 5 members, 80 nominal hours each in a two-week iteration, and a usable-capacity factor of 0.70. Calculate capacity and explain why the result may still overstate delivery capability.
2. A project has two paths: A–B–D with durations 4, 5, and 3 days, and A–C–D with durations 4, 8, and 3 days. Identify the critical path and earliest completion, assuming finish-to-start logic and adequate resources.
3. Design an estimation, scheduling, resource, and procurement approach for a project with a fixed regulatory launch date but uncertain user requirements. Justify which elements should be predictive and which should remain adaptive.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes a WBS from a product backlog?

A. A WBS is deliverable-oriented, while a backlog is a prioritised and evolving list of work.

B. A WBS contains only technical tasks, while a backlog contains only business tasks.

C. A WBS changes every iteration, while a backlog is permanently fixed.

D. A WBS is used only for procurement, while a backlog is used only for budgeting.

Correct Answer: A. A WBS is deliverable-oriented, while a backlog is a prioritised and evolving list of work.

Question 2

A work package has optimistic duration 10 days, most likely duration 16 days, and pessimistic duration 28 days. Using TE = (O + 4M + P) ÷ 6, what is the expected duration?

A. 14 days

B. 16 days

C. 17 days

D. 19 days

Correct Answer: C. 17 days

Question 3

A team completes 24, 27, 25, and 28 story points in four comparable iterations. What is its arithmetic mean velocity?

A. 24 points

B. 25 points

C. 26 points

D. 28 points

Correct Answer: C. 26 points

Question 4

Which path is critical in a schedule network?

A. The path with the fewest activities

B. The longest-duration logical path through the network

C. The path with the lowest labour cost

D. The path containing the most specialised resources

Correct Answer: B. The longest-duration logical path through the network

Question 5

Six team members each have 80 nominal hours in an iteration, with a usable-capacity factor of 0.75. What is calculated capacity?

A. 240 hours

B. 360 hours

C. 420 hours

D. 480 hours

Correct Answer: B. 360 hours

Question 6

Which contract arrangement is generally most suitable when requirements are evolving and the buyer can actively govern priorities?

A. Time and materials

B. Unconditional fixed price

C. Fixed price with no change mechanism

D. Payment only after project closure

Correct Answer: A. Time and materials

Question 7

In earned value analysis, what does a CPI below 1 generally indicate?

A. More value is being earned per unit of cost than planned

B. Less value is being earned per unit of cost than planned

C. The project is ahead of schedule

D. Actual cost is zero

Correct Answer: B. Less value is being earned per unit of cost than planned

Question 8

What is the principal purpose of rolling-wave planning?

A. To eliminate all project uncertainty before execution begins

B. To plan near-term work in detail while progressively elaborating later work

C. To prevent any change to the approved scope

D. To replace the need for a schedule network

Correct Answer: B. To plan near-term work in detail while progressively elaborating later work

Question 9

Which item is most important when evaluating a parametric estimate?

A. Whether the model uses the largest possible coefficient

B. Whether drivers are consistently defined and historical data are relevant

C. Whether the estimate is expressed only as a single number

D. Whether all activities have identical durations

Correct Answer: B. Whether drivers are consistently defined and historical data are relevant

Question 10

In the Northstar case, why was accessibility capacity treated as a schedule constraint?

A. Accessibility work was legally prohibited from being planned early

B. Only one specialist was available half-time, limiting a required capability

C. Accessibility testing did not require acceptance criteria

D. The team used no backlog for accessibility work

Correct Answer: B. Only one specialist was available half-time, limiting a required capability