Work, Job and Workforce Design

Introduction

Work design concerns the content, structure, relationships and context through which work is performed. Job design is the more specific arrangement of tasks and responsibilities within a role, while workforce design concerns the composition, deployment and coordination of the people and capabilities required by an organisation. In the age of artificial intelligence (AI), these three levels are increasingly inseparable. Automation can remove some tasks, augment others and create new coordination, monitoring and judgment responsibilities. Remote work changes the social and physical context of work. Platform work and contingent labour alter the employment relationship. Skills-based organisations challenge the assumption that work must be organised primarily around permanent jobs.

For human resource professionals, the central question is not simply whether technology can perform a task. It is whether work should be redesigned, by whom, for what purpose and with what consequences for performance, wellbeing, fairness and organisational capability. Effective design aligns tasks, technology, people, structures and employment arrangements.

This chapter applies major work-design theories to task decomposition, job redesign, skills-based work, remote and hybrid work, agile teams, platform work and contingent workforce models.

Learning Objectives

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

1. Distinguish work, tasks, jobs, roles and workforce systems.
2. Apply the Job Characteristics Model and Job Demands–Resources theory to diagnose work design.
3. Decompose jobs into tasks and identify opportunities for automation, augmentation and human judgment.
4. Design skills-based, remote, hybrid, agile and contingent work arrangements.
5. Evaluate work-design choices using performance, employee experience, inclusion, legal and ethical criteria.
6. Recommend an evidence-based redesign for a changing work system.

1. Levels and Elements of Work Design

A useful hierarchy is:

• Work system: The complete arrangement of people, technology, processes, information and governance used to produce value.
• Job: A bundle of related tasks and responsibilities assigned to a worker or position.
• Role: The expected pattern of behaviour and contribution associated with a job in a social or organisational context.
• Task: A coherent unit of activity with an identifiable purpose, inputs, actions and outputs.
• Activity: A smaller action or behaviour performed in completing a task.
• Workforce: The total population through which work is accomplished, including employees, contractors, agency workers, freelancers, partners and, increasingly, AI-enabled systems.

Work design has at least four dimensions:

A job description is not a complete work design. It may list duties but fail to specify decision rights, workflow dependencies, technology interfaces, learning requirements or psychosocial risks. Contemporary HR practice therefore treats job analysis as an investigation of actual work rather than a clerical exercise.

2. Theoretical Foundations

2.1 The Job Characteristics Model

Hackman and Oldham’s Job Characteristics Model proposes that five core job characteristics influence three critical psychological states and, in turn, important outcomes:

1. Skill variety: The extent to which a job requires different skills and activities.
2. Task identity: The extent to which a person completes a whole, identifiable piece of work.
3. Task significance: The perceived impact of the job on other people.
4. Autonomy: The freedom to schedule work and select procedures.
5. Feedback: The extent to which the work itself provides information about performance.

Skill variety, task identity and task significance contribute to experienced meaningfulness. Autonomy contributes to experienced responsibility, while feedback contributes to knowledge of results. These psychological states are associated with internal motivation, satisfaction, performance and lower withdrawal, although individual growth-need strength and contextual factors moderate the relationships.

The model is especially useful when redesigning repetitive or fragmented work. However, increasing variety is not automatically beneficial. Excessive variety can create cognitive overload, dilute expertise or obscure accountability. Similarly, autonomy without resources, competence or clear boundaries may produce anxiety rather than motivation.

2.2 Job Demands–Resources Theory

The Job Demands–Resources (JD-R) theory classifies work characteristics into two broad categories:

• Job demands: Physical, cognitive, emotional or organisational aspects requiring sustained effort and therefore creating potential cost. Examples include workload, time pressure, emotional labour, role conflict and technological complexity.
• Job resources: Aspects that help achieve goals, reduce demands or stimulate growth. Examples include autonomy, supervisor support, adequate staffing, learning opportunities, useful technology and role clarity.

The theory proposes two interacting processes. High demands can contribute to exhaustion and health impairment; resources can support motivation, engagement and performance. Resources may also buffer the effects of demands, although buffering is not guaranteed. For example, autonomy may help an experienced employee manage workload but may not compensate for impossible deadlines or inadequate staffing.

A practical diagnosis should therefore ask both “What makes this work difficult?” and “What enables people to perform it sustainably?” AI implementation often changes both sides: it may reduce routine processing but increase monitoring, exception handling, accountability and the need to learn new systems.

2.3 Sociotechnical and Work-Design Perspectives

Sociotechnical systems theory holds that performance depends on the joint optimisation of social and technical systems. A technically efficient workflow may fail if it undermines cooperation, local knowledge or employee control. Conversely, a supportive team cannot compensate indefinitely for unreliable technology or poor process architecture.

Contemporary work design also draws on relational, proactive and team perspectives. Relational design examines contact with beneficiaries, customers and colleagues. Proactive design recognises that employees often craft tasks, relationships and interpretations of work. Team design considers interdependence, coordination, boundary spanning and shared mental models. These perspectives are essential where work is distributed across people and AI systems.

3. From Jobs to Tasks: Decomposition for AI and Redesign

Task decomposition breaks a job into observable units and identifies the conditions under which each unit is performed. A robust decomposition records:

1. Trigger: What initiates the task?
2. Inputs: What data, materials or requests are needed?
3. Activities: What decisions and actions occur?
4. Output: What is produced or changed?
5. Quality criteria: What counts as acceptable performance?
6. Dependencies: Who or what must provide or receive information?
7. Exceptions: What unusual cases require judgment?
8. Risks: What errors, harms or compliance failures are possible?
9. Capabilities: What knowledge, skill and experience are required?
10. Technology interface: Which tools support, automate or constrain the task?

Tasks should not be classified solely by whether they are “routine” or “non-routine.” A more useful classification considers variability, consequence, explainability and the availability of reliable data.

A task-level analysis avoids the misleading claim that an entire occupation will simply disappear. Occupations are bundles of tasks, and technological change usually redistributes tasks within and across jobs. The resulting job may require fewer routine activities but more exception management, customer communication, system oversight and continuous learning.

A Practical Task-Redesign Sequence

A redesign team can follow six stages:

1. Observe actual work. Combine interviews, workflow records, diaries, system logs and direct observation. Do not rely only on formal job descriptions.
2. Map task flows. Identify handoffs, delays, rework and decision points.
3. Assess task characteristics. Evaluate variability, criticality, data quality, interdependence and emotional demands.
4. Choose the intervention. Eliminate, simplify, automate, augment, delegate, rotate or deepen the task.
5. Redesign the job and controls. Specify decision rights, escalation, training, performance measures and accountability.
6. Pilot and evaluate. Compare quality, cycle time, workload, error rates, employee experience and equity before scaling.

The sequence reflects a critical principle: automation without redesign can create “thin” jobs in which people remain responsible for difficult exceptions while losing control over routine work and opportunities for mastery.

4. Job Redesign and Skills-Based Work

Traditional workforce planning begins with positions and headcount. A skills-based approach begins with capabilities and the work that must be accomplished. It may use skills taxonomies, proficiency levels, capability assessments, internal talent marketplaces and project-based deployment.

A skills-based architecture commonly includes:

• a validated skills taxonomy;
• definitions of proficiency and evidence of competence;
• inventories of employee capabilities and interests;
• transparent mechanisms for matching people to work;
• learning pathways linked to business priorities; and
• governance for data quality, privacy and fairness.

Skills-based design is valuable when work changes rapidly or when employees possess capabilities not visible in job titles. Nevertheless, skills should not be treated as context-free commodities. Effective performance also depends on experience, relationships, professional identity, motivation, resources and knowledge of organisational context.

Job enlargement adds tasks at a similar level; job enrichment increases responsibility, discretion or ownership; job rotation moves people among tasks or roles; and job crafting allows employees some discretion to alter task, relational or cognitive boundaries. Each intervention has different implications. Rotation can broaden capability but may reduce short-term efficiency. Enrichment can increase meaningfulness but also increase accountability and stress if resources are not added.

A skills-based system should therefore distinguish between:

• core skills, required across many roles;
• technical skills, tied to a domain or technology;
• transferable skills, useful across contexts;
• adjacent skills, close to existing capabilities and suitable for development; and
• emerging skills, whose value is rising but whose definitions may be unstable.

5. Remote, Hybrid and Distributed Work

Remote and hybrid work are work-design choices, not merely location policies. They alter communication, visibility, access to information, coordination rhythms, social connection, learning and boundary management.

A sound hybrid design specifies the purpose of presence. Office attendance may be most valuable for team formation, complex collaboration, mentoring, creative work and conflict resolution; individual focused work may be performed effectively elsewhere. However, these assumptions should be tested rather than universalised. Employees differ in home circumstances, disability, caregiving responsibilities, commuting burdens, digital access and preferences.

Important design variables include:

• Temporal flexibility: When work may be performed.
• Spatial flexibility: Where work may be performed.
• Synchrony: The extent to which people must work at the same time.
• Coordination infrastructure: Shared documentation, task boards, decision logs and communication norms.
• Inclusion mechanisms: Equal access to information, promotion, voice and development.
• Boundary supports: Norms concerning availability, recovery time and after-hours communication.

Hybrid work can create proximity bias if managers unconsciously favour people who are physically present. It can also produce meeting overload, fragmented attention and weaker informal learning. Performance management should therefore rely on clear outcomes, quality standards and regular developmental conversations rather than online presence or response speed alone.

6. Agile Teams and Team-Based Work Design

Agile teams are typically small, multidisciplinary and oriented toward iterative delivery and customer or stakeholder feedback. Agile design can improve responsiveness by bringing relevant expertise closer to the work. It may also increase role ambiguity, coordination demands and emotional pressure if priorities change continuously.

Effective agile work design requires clarity about:

• the team’s purpose and boundaries;
• authority to make operational decisions;
• product or service ownership;
• responsibilities for risk, compliance and quality;
• interfaces with other teams;
• mechanisms for prioritisation and escalation; and
• learning through retrospectives and review cycles.

Agility is not equivalent to the absence of structure. Teams need stable goals, decision rules, adequate resources and psychologically safe communication. HR professionals should evaluate not only delivery speed but also workload sustainability, inclusion, skill development and the distribution of invisible coordination work.

7. Platform and Contingent Work

Platform work matches labour supply and demand through digital systems. It may involve transportation, delivery, professional services, online freelancing or microtasks. Contingent work includes temporary agency work, fixed-term contracts, independent contracting, seasonal work, on-call arrangements and statement-of-work engagements.

These models can provide flexibility, specialist expertise and variable capacity. They may also transfer risk from organisations to workers through uncertain income, limited benefits, algorithmic management and weak voice. Classification is not determined solely by a contract label. Relevant legal tests often examine control, economic dependence, integration into the business and the practical reality of the relationship; the applicable rules differ across jurisdictions.

Responsible workforce design should address:

• fair and transparent pay mechanisms;
• safe working conditions and access to remedy;
• data protection and surveillance limits;
• non-discrimination in allocation and rating systems;
• predictable scheduling where feasible;
• training and access to relevant information; and
• clear accountability across the organisation, platform and client.

A mixed workforce also creates coordination and fairness challenges. Permanent employees may possess institutional knowledge while contractors perform critical work without equivalent access to systems or development. HR must design onboarding, security, knowledge transfer and inclusion processes that reflect different employment relationships without creating unlawful or unjustified disparities.

8. AI, Algorithmic Management and Governance

AI can support recruitment, scheduling, workforce analytics, training recommendations, service delivery and performance management. Its use changes work design through both substitution and complementarity. AI may execute tasks, recommend actions, monitor patterns, generate content or mediate interactions.

The principal governance risks include:

• Bias and disparate impact: Historical data or proxy variables may reproduce unequal outcomes.
• Opacity: Workers may not understand how decisions or ratings are produced.
• Automation bias: Users may accept system recommendations without adequate scrutiny.
• Privacy intrusion: Monitoring may exceed legitimate business necessity.
• Deskilling: Excessive dependence on automated systems can erode human expertise.
• Accountability gaps: Responsibility may become unclear when decisions involve vendors, managers and algorithms.
• Psychosocial harm: Continuous measurement can increase stress and reduce autonomy.

A responsible design establishes purpose limitation, data minimisation, access controls, documentation, impact assessment, human oversight, auditability and appeal. High-impact decisions affecting employment should receive heightened scrutiny, and employees should receive understandable information about relevant automated systems where required by law or organisational policy. Governance must be continuous because models, data, tasks and social effects change over time.

9. Measuring Work-Design Quality

Evaluation should combine operational, human and equity outcomes. Useful indicators include:

Metrics should be interpreted together. A reduction in average handling time may represent improvement, or it may indicate rushed service and hidden rework. A productivity increase accompanied by burnout is not a sustainable work-design success.

Where quantitative evaluation is appropriate, a simple percentage change can be calculated as:

The baseline must be non-zero, values must measure the same construct in the same units, and the interpretation should account for seasonality, selection effects and concurrent changes. For example, if average processing time falls from 10 minutes to 8 minutes, the change is ((8 − 10) ÷ 10) × 100 = −20%. This indicates a 20% reduction in time, not necessarily a 20% improvement in value or quality.

10. Illustrative Case: Redesigning Claims Work with AI

Background

A fictional insurance organisation, Northstar Mutual, employs claims specialists who receive documents, verify coverage, assess damage, communicate with customers and approve or refer claims. Rising volumes have created delays and overtime. The organisation proposes an AI system to classify documents, identify missing information and estimate low-complexity claims.

Problem

Initial implementation treats the job as a single unit and sets a target to automate as much processing as possible. Specialists report that the system produces incomplete summaries, while managers worry about inconsistent decisions and customer complaints. Employees increasingly handle unusual and emotionally difficult cases but have less time for customer communication and learning.

Analysis

A task decomposition identifies five clusters:

1. Document intake and classification.
2. Policy and coverage verification.
3. Damage and evidence assessment.
4. Customer explanation and support.
5. Approval, escalation and fraud-related judgment.

The first cluster is high-volume and relatively structured, making it suitable for automation with sampling and exception controls. The fourth and fifth clusters involve empathy, ambiguity, rights and consequential judgment; they require human involvement. The JD-R analysis finds that workload has shifted rather than disappeared: routine effort falls, but cognitive and emotional demands rise. The Job Characteristics Model indicates reduced task identity and feedback because specialists see only fragments of cases and receive opaque system scores.

Approach

Northstar redesigns the work as follows:

• AI classifies documents and flags missing information, but does not make final coverage decisions.
• Specialists receive confidence indicators and can inspect source documents.
• Cases are triaged by complexity, with mandatory human review for vulnerable customers, disputed coverage and high-value claims.
• Teams own complete case outcomes rather than isolated processing steps.
• A rotating quality role reviews errors and updates guidance.
• Performance measures include accuracy, customer understanding, fair treatment, rework and sustainable workload, not only speed.
• Employees receive training in exception handling, system limitations, communication and data protection.
• A formal appeal and incident-reporting process records algorithmic errors and customer harm.

Outcome

In the pilot, routine processing becomes faster, but the main improvement comes from reduced rework and clearer escalation. Specialists report greater responsibility for complete cases and improved learning from feedback. The organisation does not eliminate human review; it reallocates human effort toward complex judgment and customer support.

Learning Points

The case demonstrates that AI value depends on work-system redesign, not tool installation. Task decomposition reveals where automation is suitable, JD-R identifies changed demands and resources, and the Job Characteristics Model highlights the need to preserve meaningful ownership, autonomy and feedback. Governance is part of job design because accountability, review and appeal must be built into the workflow.

11. Implementation Principles and Limitations

A professional work-design intervention should be participative. Employees possess situated knowledge about exceptions, workarounds and risks that may not appear in system data. Consultation is not merely a change-management technique; it improves the validity of the analysis.

Recommended practices include:

1. Define the value and problem before selecting technology.
2. Analyse actual tasks, including invisible coordination and emotional work.
3. Involve affected employees, managers, customers and relevant specialists.
4. Preserve human discretion where context, rights or relationships matter.
5. Build training, recovery time and learning into the design rather than adding them later.
6. Pilot with explicit success and harm indicators.
7. Segment outcomes to identify unequal effects.
8. Document decision rights and escalation routes.
9. Review employment classification and collective consultation obligations.
10. Reassess the design as technology, markets and regulation change.

Work-design theories have limitations. The Job Characteristics Model can understate collective and cultural influences. JD-R categories are broad and may vary across occupations. Skills taxonomies can simplify expertise and become outdated. Hybrid-work research may not generalise across countries, occupations or household conditions. AI impact assessments can become procedural exercises if leaders do not act on findings. These limitations reinforce the need for triangulation: combine theory, qualitative evidence, operational data and employee voice.

Chapter Summary

Work design operates across tasks, jobs, roles and the wider workforce system. The Job Characteristics Model explains how variety, identity, significance, autonomy and feedback influence meaningfulness, responsibility and knowledge of results. JD-R theory distinguishes demands that consume effort from resources that enable performance, motivation and sustainable work. Sociotechnical perspectives emphasise joint optimisation of people, technology and social arrangements.

Task decomposition is the foundation for responsible AI-enabled redesign. It identifies inputs, decisions, outputs, exceptions, risks and required capabilities, allowing HR professionals to decide what to automate, augment, delegate or retain as human judgment. Skills-based work shifts attention from fixed positions to capabilities, but skills must be interpreted alongside experience, context and resources.

Remote and hybrid work require deliberate coordination, inclusion and boundary design. Agile teams need autonomy together with clear purpose, interfaces and accountability. Platform and contingent work can provide flexibility but raise issues of security, classification, fairness and voice. AI and algorithmic management require transparency, human oversight, privacy protection, impact assessment and appeal. Finally, work-design success must be evaluated through a balanced set of performance, wellbeing, learning, equity and governance measures.

Key Terms

Work design: The arrangement of tasks, relationships, technology and conditions through which work is performed.

Job design: The structuring of tasks, responsibilities and decision authority within a job.

Task: A coherent unit of work with an identifiable purpose and output.

Role: Expected behaviours and contributions associated with a position in context.

Workforce design: Planning the composition, deployment and coordination of people and capabilities.

Job Characteristics Model: A theory linking five job characteristics to psychological states and work outcomes.

Skill variety: The range of skills and activities required by a job.

Task identity: The extent to which a person completes an identifiable whole piece of work.

Autonomy: Discretion over work scheduling and methods.

Job demand: A work aspect requiring sustained effort and creating potential cost.

Job resource: A work aspect that supports goals, reduces demands or promotes growth.

Task decomposition: Systematic separation of a job into inputs, activities, outputs, decisions and exceptions.

Job enrichment: Increasing responsibility, discretion or ownership in a job.

Skills-based organisation: An organisation that structures deployment and development substantially around capabilities rather than fixed positions.

Hybrid work: Work combining remote and in-person arrangements.

Agile team: A multidisciplinary team using iterative delivery, feedback and adaptive prioritisation.

Platform work: Work matched, allocated or managed through a digital platform.

Contingent worker: A worker engaged through a non-standard or limited employment arrangement.

Algorithmic management: The use of computational systems to allocate, monitor, evaluate or direct work.

Human oversight: Meaningful human review, intervention and accountability in technology-supported decisions.

Review Questions

Short Answer Questions

1. Distinguish between a task, a job and a role.
2. Identify the five core characteristics in the Job Characteristics Model.
3. What is the difference between a job demand and a job resource?
4. Why is task decomposition important in AI-enabled work redesign?
5. State two risks associated with hybrid work.

Long Answer Questions

1. Explain the Job Characteristics Model and evaluate its usefulness for redesigning digitally monitored work.
2. Discuss how JD-R theory can guide the implementation of AI in a customer-service function.
3. Compare skills-based work design with traditional position-based workforce planning.
4. Analyse the opportunities and risks of agile team design.
5. Explain the governance requirements for responsible algorithmic management.

Analytical / Application Questions

1. A call centre introduces an AI assistant that reduces routine queries but increases escalations and monitoring. Use JD-R theory and the Job Characteristics Model to diagnose the redesign and recommend changes.
2. Design a task-decomposition framework for a hybrid HR business-partner role. Identify which tasks could be automated, augmented or retained as human-led, and justify your choices.
3. An organisation plans to replace 20% of permanent administrative roles with platform contractors. Develop an evaluation framework covering performance, workforce capability, worker wellbeing, equity, legal risk and governance.

Multiple Choice Questions with Answers

Question 1

Which Job Characteristics Model dimension refers to freedom over scheduling and procedures?

A. Task identity

B. Autonomy

C. Feedback

D. Task significance

Correct Answer: B. Autonomy

Question 2

Under JD-R theory, time pressure and emotional labour are generally classified as what?

A. Job resources

B. Social characteristics

C. Job demands

D. Core skills

Correct Answer: C. Job demands

Question 3

Which task is generally the strongest candidate for straightforward automation?

A. Resolving an unprecedented ethical complaint

B. Providing emotional support to a distressed customer

C. Applying a stable rule to high-volume, low-consequence data

D. Negotiating a disputed employment settlement

Correct Answer: C. Applying a stable rule to high-volume, low-consequence data

Question 4

What is the primary purpose of task decomposition?

A. To eliminate all jobs containing routine activities

B. To separate work into activities, decisions, outputs and exceptions for analysis

C. To replace job analysis with employee preference surveys

D. To rank employees by personality type

Correct Answer: B. To separate work into activities, decisions, outputs and exceptions for analysis

Question 5

Which practice helps reduce proximity bias in hybrid work?

A. Rewarding physical presence regardless of outcomes

B. Using clear outcome standards and equal access to information

C. Requiring all decisions to occur informally in the office

D. Measuring commitment through online response speed

Correct Answer: B. Using clear outcome standards and equal access to information

Question 6

What is a central principle of sociotechnical systems theory?

A. Technical optimisation should always override social considerations

B. Social and technical systems should be jointly designed

C. Human judgment should be removed from complex work

D. Jobs should be designed independently of technology

Correct Answer: B. Social and technical systems should be jointly designed

Question 7

If processing time falls from 10 minutes to 8 minutes, what is the percentage change?

A. 2% increase

B. 20% decrease

C. 25% decrease

D. 80% increase

Correct Answer: B. 20% decrease

Question 8

Which feature is most characteristic of a skills-based workforce model?

A. Deployment based exclusively on permanent job titles

B. Capability inventories and matching people to changing work

C. Elimination of learning pathways

D. Use of one universal skill definition for every occupation

Correct Answer: B. Capability inventories and matching people to changing work

Question 9

Which is a principal risk of algorithmic management?

A. Guaranteed elimination of discrimination

B. Increased transparency in every decision

C. Opacity and unclear accountability for system-supported decisions

D. Automatic improvement in employee autonomy

Correct Answer: C. Opacity and unclear accountability for system-supported decisions

Question 10

Why should high-consequence AI-supported employment decisions receive heightened human oversight?

A. They have no effect on employee rights

B. They may affect people’s opportunities and require review, explanation or appeal

C. Human review always makes decisions slower without benefit

D. High-consequence decisions are always routine

Correct Answer: B. They may affect people’s opportunities and require review, explanation or appeal