Customer, Stakeholder and Market-Driven Quality

Introduction

Quality is not determined solely by conformance to an internal specification. In contemporary organizations, quality is judged by customers, users, regulators, employees, suppliers, owners, communities and other stakeholders. A product may meet its engineering tolerances and still fail because it is difficult to use, delivered late, poorly supported, inaccessible, unsafe or inconsistent with customer expectations.

Market-driven quality therefore begins with understanding value from the perspective of those affected by an organization’s products and processes. The central managerial challenge is translation: vague statements such as “the service should be fast” or “the product must be reliable” must become measurable, prioritized and technically feasible requirements.

This chapter examines the principal methods used for that translation: voice-of-customer research, stakeholder analysis, satisfaction measurement, service-quality models, Kano analysis, complaint management and Quality Function Deployment (QFD). These methods are most effective when used as an integrated system rather than as isolated tools.

Learning Objectives

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

1. Explain customer-, stakeholder- and market-driven quality.
2. Distinguish needs, wants, expectations, requirements and measurable specifications.
3. Select appropriate methods for collecting and analysing the voice of the customer.
4. Apply satisfaction analysis and service-quality models critically.
5. Classify requirements using the Kano model.
6. Design an effective complaint-management process.
7. Construct and interpret the main elements of a QFD House of Quality.
8. Recognize ethical, technological and managerial limitations in customer-based quality decisions.

1. From Internal Conformance to Perceived Value

Traditional quality management often emphasizes conformance: whether output conforms to a documented specification. Conformance remains essential, particularly for safety, legal compliance and reliability. However, customer value includes more than technical conformity. It may include convenience, emotional reassurance, transparency, accessibility, environmental responsibility and the quality of interactions.

A useful distinction is between objective quality and perceived quality. Objective quality concerns measurable characteristics such as response time, defect rate or dimensional accuracy. Perceived quality concerns the customer’s interpretation of performance in relation to expectations, alternatives and circumstances. The two are related but not identical.

A customer’s overall evaluation can be represented conceptually as:

Benefits may include performance, reliability, status and service. Sacrifices may include price, time, effort, risk and inconvenience. This is a conceptual relationship rather than a universal accounting equation; benefits and sacrifices are not always measured in the same units. Organizations should therefore avoid treating perceived value as a precise financial calculation unless a validated measurement model has been developed.

Customers and stakeholders

A customer purchases, uses or receives an organization’s output. The purchaser, end user, payer and decision-maker may be different people. For example, a procurement department may purchase software, employees may use it, and an executive may approve renewal.

A stakeholder is any person or group that can affect, be affected by, or perceive itself to be affected by an organization’s decisions or activities. Stakeholders may include:

• customers and end users;
• employees and managers;
• owners and investors;
• suppliers and logistics partners;
• regulators and accreditation bodies;
• local communities;
• professional associations;
• advocacy groups and the natural environment.

Stakeholder requirements may conflict. Customers may want rapid delivery, employees may require realistic workloads, regulators may demand extensive controls, and owners may seek cost reduction. Quality management must make these tensions visible and resolve them through evidence, risk analysis and governance rather than by simply satisfying the loudest group.

2. Translating Voice into Requirements

The voice of the customer (VOC) is the structured representation of customer needs, expectations, experiences and priorities. It is not merely a list of quotations. A credible VOC process separates what customers say from what they need and from how the organization will design a response.

A practical translation chain is:

customer statement → interpreted need → quality attribute → measurable requirement → process or design control → verification method

For example:

A good requirement is sufficiently specific to be tested, but not so narrowly defined that it prevents innovation. Requirements should normally identify a measurable characteristic, target or limit, operating conditions and method of verification.

Needs, wants and requirements

A need is a fundamental outcome sought by a customer, such as safety or dependable access. A want is a preferred feature or form of delivery, such as a particular interface colour. A requirement is a documented condition that must be satisfied. A want can become a requirement if research, regulation, contract or strategy establishes that it is important.

Requirements should be checked for:

• clarity: different readers interpret them similarly;
• measurability: achievement can be evaluated;
• traceability: the origin and downstream control are known;
• feasibility: the organization can achieve them with available resources;
• non-contradiction: they do not conflict with higher-priority requirements;
• relevance: they contribute to value, risk control or compliance.

Methods for collecting VOC

No single method captures all customer knowledge. Qualitative methods reveal language, motives and unanticipated problems; quantitative methods estimate prevalence and priority.

Sampling must reflect the organization’s customer segments. Averages can conceal serious dissatisfaction among vulnerable, high-risk or low-volume users. Researchers should consider response rates, non-response bias, accessibility, language, privacy and the difference between a customer’s stated preference and actual behaviour.

3. Satisfaction Analysis

Customer satisfaction is an evaluation formed by comparing perceived performance with a reference point, often expectations. A simplified conceptual model is:

This is not a universal computational formula. It indicates that the same performance can produce different evaluations depending on expectations and the importance of the attribute.

Measuring satisfaction

Common measures include attribute ratings, overall satisfaction, repurchase intention, retention, complaint frequency and recommendation intention. A frequently used recommendation measure is the Net Promoter Score (NPS):

Respondents are commonly classified on a 0–10 recommendation scale: promoters are scores 9–10, passives 7–8 and detractors 0–6. NPS ranges from −100 to +100. It is a descriptive indicator, not a complete measure of quality, loyalty or profitability. Results depend on sampling, question wording, industry context and response behaviour. It should be combined with operational and diagnostic measures.

For an attribute-based satisfaction survey, a simple weighted priority index may be used:

Here, i denotes an attribute; Importance_i and Performance_i must use compatible scales; and “Maximum score” is the upper end of the performance scale. The index assumes that a performance gap and importance are meaningful and approximately multiplicative. It is useful for screening priorities, but it should not be treated as a causal model.

Worked example: prioritizing service improvements

Suppose customers rate three attributes from 1 to 5. Importance is also rated from 1 to 5.

Waiting time has the highest screening priority, followed by information clarity. The result does not prove that waiting time is the only cause of dissatisfaction. Managers should investigate capacity, scheduling, process variation and customer segmentation before selecting an intervention.

4. Service Quality and the SERVQUAL Perspective

Services have characteristics that make quality management difficult. They are often intangible, produced and consumed together, variable across employees and customers, and perishable because unused capacity cannot normally be stored. Service quality therefore includes both the result and the process of delivery.

The SERVQUAL framework is commonly associated with five dimensions:

1. Reliability: performing the promised service dependably and accurately.
2. Responsiveness: willingness and speed in helping customers.
3. Assurance: employee knowledge, courtesy and ability to inspire trust.
4. Empathy: individualized care and attention.
5. Tangibles: facilities, equipment, appearance and communication materials.

A traditional gap score for attribute i is:

Both terms must be measured on the same scale and under comparable conditions. A negative gap indicates that perceived performance is below stated expectation. Gap scores are easy to communicate but have limitations: expectations may be unstable, the difference score can have lower reliability than its components, and the model may not capture all industry-specific dimensions. Organizations should validate dimensions for their context rather than applying them mechanically.

Service quality also depends on the service recovery experience. A failure followed by a rapid, respectful and fair recovery may produce a better evaluation than a service that was merely adequate. This does not justify poor first-time performance; recovery consumes resources and may not restore trust.

5. Kano Analysis: Different Effects of Requirements

The Kano model classifies attributes according to their relationship with satisfaction. It helps organizations avoid the assumption that every feature creates value in the same way.

• Must-be attributes: Their presence is taken for granted; their absence causes strong dissatisfaction. Examples may include safety controls or basic billing accuracy.
• One-dimensional attributes: Satisfaction generally increases as performance improves, and dissatisfaction increases as performance declines. Waiting time and battery life often behave in this way.
• Attractive attributes: Their presence creates delight, but their absence does not normally cause dissatisfaction because customers do not expect them.
• Indifferent attributes: Customers show little reaction to their presence or absence.
• Reverse attributes: Some customers prefer the attribute not to be present, perhaps because it adds complexity or cost.
• Questionable responses: Inconsistent answers may indicate misunderstanding or poor questionnaire design.

Kano classification is obtained through paired functional and dysfunctional questions. For an attribute such as live chat, the respondent is asked how they feel if it is present and how they feel if it is absent. The response combination is classified using a Kano evaluation table.

Kano categories can change over time. An attractive innovation may become a one-dimensional expectation and eventually a must-be condition. Competitive benchmarking and repeated research are therefore necessary. Kano analysis should guide prioritization, not replace safety analysis, legal review or direct observation.

6. Complaint Management as a Quality System

A complaint is more than an expression of dissatisfaction. It is an information signal about a failure in a product, process, communication or expectation-setting system. Effective complaint management has two purposes: restore the individual customer’s confidence and prevent recurrence across the system.

A robust complaint process includes:

1. Access: customers can complain through convenient, accessible channels.
2. Acknowledgement: the complaint is recorded and receipt is confirmed.
3. Classification: severity, type, product, location and customer impact are identified.
4. Containment: immediate action protects the customer from further harm.
5. Investigation: evidence is examined without assuming blame.
6. Resolution: an appropriate remedy is offered within a defined timeframe.
7. Communication: the customer receives clear, respectful information.
8. Root-cause analysis: systemic causes are identified where appropriate.
9. Corrective action: controls are changed and effectiveness is verified.
10. Trend review: management monitors recurring patterns and emerging risks.

Useful measures include acknowledgement time, resolution time, repeat complaint rate, unresolved-case backlog, complaint rate per transaction, recurrence after corrective action and customer perception of fairness. A low complaint count does not necessarily indicate high quality. Customers may consider complaining difficult, pointless or risky. Complaint data should therefore be interpreted alongside retention, interviews, observed failures and service records.

Root-cause methods may include the five whys, cause-and-effect diagrams, process mapping, Pareto analysis and fault-tree analysis. The five whys is a questioning discipline, not a guarantee that exactly five questions will reveal a root cause. Investigators should distinguish immediate cause, contributing conditions and systemic management causes.

Ethics and governance in complaints

Complaint systems must protect confidentiality, prevent retaliation and provide additional support for vulnerable customers. Automated triage should not systematically downgrade complaints from particular languages, regions or demographic groups. Staff incentives based only on closing cases quickly may encourage premature rejection rather than fair resolution. Governance should define escalation for safety, discrimination, privacy, fraud and regulatory issues.

7. Quality Function Deployment

Quality Function Deployment (QFD) is a structured method for translating customer requirements into technical characteristics, process controls and deployment priorities. Its best-known tool is the House of Quality (HOQ).

The main HOQ components are:

1. Customer requirements, or WHATs: what customers need.
2. Importance ratings: relative significance of each need.
3. Technical characteristics, or HOWs: measurable organizational or engineering responses.
4. Relationship matrix: strength of the relationship between WHATs and HOWs.
5. Correlation roof: positive and negative interactions among HOWs.
6. Competitive assessment: customer perception of the organization and competitors.
7. Technical priorities and targets: decisions for design and improvement.

A common prioritization calculation is:

where i indexes customer requirements, j indexes technical characteristics, and Relationship_{ij} is a chosen relationship score, often 0, 1, 3 or 9. The scores are ordinal conventions, not physical measurements. They should be used consistently within one analysis and interpreted as decision-support information rather than as precise utility values.

If improvement difficulty or competitive significance is included, organizations may use an adjusted priority such as:

The improvement factor must be explicitly defined. Since different organizations use different conventions, the formula is not meaningful unless the scale and rationale are documented.

Worked QFD example

An online appointment service identifies three customer requirements:

Three technical characteristics are selected:

• interface task steps;
• confirmation processing time;
• information validation rate.

Use relationship scores 9 for strong, 3 for medium, 1 for weak and 0 for none.

For task steps:

Correction: using the table above, the task-step priority is 50, not 71. For processing time:

For validation rate:

Thus the corrected priorities are 50, 62 and 64. Validation rate receives the highest calculated priority, followed by processing time and task-step reduction. This example illustrates why arithmetic verification and transparent tables are essential. The result still requires managerial judgement: a high-priority technical characteristic may be expensive, dependent on another characteristic or constrained by regulation.

QFD is not a one-time meeting. Requirements should flow through successive stages: product or service planning, part or subsystem planning, process planning and production or service-control planning. Traceability should connect the original customer requirement to a control plan and evidence of achievement.

8. Integrated Application: Illustrative Case

Background

A fictional regional healthcare clinic receives increasing complaints about appointment access. Patients report long telephone waits, uncertainty about booking status and occasional errors in appointment dates. Staff report that the booking system is difficult to use and that repeated calls increase workload.

Problem

The clinic initially proposes adding more telephone lines. However, management lacks evidence about which aspects of access matter most and whether the proposed solution addresses the underlying causes.

Analysis

The quality team conducts interviews with patients, observes booking transactions, analyses complaints and surveys patients using importance and performance ratings. It identifies four principal needs:

• appointments should be easy to request;
• confirmation should be timely;
• information should be accurate;
• alternative access should exist for people with limited digital access.

Stakeholder analysis reveals additional requirements: staff need manageable workloads, clinicians need accurate schedules, regulators require privacy protection, and vulnerable patients need accessible channels.

Kano questioning suggests that accurate appointment information is a must-be attribute, while a self-service reminder function is attractive for some users. Survey analysis shows that confirmation delay has high importance and low performance. Complaint records reveal that many “waiting time” complaints actually result from incomplete information and failed call-backs.

Approach and solution

The clinic constructs a QFD matrix. Customer requirements are translated into technical characteristics including confirmation processing time, data-validation controls, call-back completion rate, interface simplicity and accessibility of channels. Targets are established, for example:

• 95% of booking requests acknowledged within 10 minutes;
• 99.5% accuracy for appointment date and time in confirmed records;
• 90% of promised call-backs completed within the stated period;
• accessibility testing with representative users before release.

The clinic introduces a digital booking option but retains telephone and in-person access. Complaint categories are standardized, and serious privacy or clinical-risk complaints are escalated immediately. A monthly review examines performance, complaint themes and the experiences of patients who use each channel.

Outcome

In this illustrative case, confirmation delays decline and repeat calls decrease. The most important improvement is not the technology alone but the translation of patient needs into measurable controls. The clinic also avoids an inequitable “digital-only” solution by treating accessibility as a stakeholder and quality requirement.

Learning points

1. A proposed solution should not be selected before the problem is understood.
2. Complaint themes may conceal different underlying causes.
3. Customer needs must be balanced with staff, regulatory and accessibility requirements.
4. QFD creates traceability between voice, design decisions and controls.
5. Measures should include both performance and customer experience.

9. Implementation Principles and Limitations

Organizations seeking market-driven quality should observe the following principles:

• segment customers rather than relying only on grand averages;
• combine qualitative discovery with quantitative measurement;
• distinguish stated preferences from observed behaviour;
• use representative samples and document limitations;
• translate needs into testable requirements;
• connect customer measures to operational process measures;
• review requirements as markets, technologies and expectations change;
• include accessibility, safety, privacy and sustainability in stakeholder analysis;
• verify that improvement benefits are distributed fairly;
• close the feedback loop by informing customers how their input was used.

Several limitations require critical judgement. Customers may not be able to articulate latent needs. Surveys can be affected by wording, timing, culture and response bias. A high satisfaction score may coexist with low loyalty when alternatives are limited. NPS and other summary indicators can be overinterpreted. Kano categories are context-dependent and may change. QFD matrices can become bureaucratic, with false precision created by arbitrary numbers or excessive detail. Finally, customer preference cannot override legal duties, safety requirements, professional ethics or the rights of affected non-customers.

Technology and artificial intelligence

Digital channels, text analytics, speech analytics and artificial intelligence can classify complaints, identify themes and detect changes in sentiment. These tools can improve scale and speed, but they do not automatically produce valid VOC. Models may misinterpret sarcasm, dialect, multilingual expressions or accessibility-related communication. Training data may reproduce historical bias. Organizations should maintain human review for high-impact decisions, document model limitations, protect personal information and provide escalation routes.

A responsible analytics process separates exploratory signals from confirmed findings. An apparent sentiment decline should be checked against sampling changes, service events and direct customer evidence. Automation should support, not replace, accountability for complaint resolution and requirement approval.

Chapter Summary

Market-driven quality treats customer and stakeholder value as central inputs to quality planning. The voice of the customer must be translated from statements and experiences into clear, measurable and traceable requirements. Interviews, observation, surveys, complaints and operational data provide complementary evidence.

Satisfaction analysis compares perceived performance with expectations and importance, while service-quality models help examine reliability, responsiveness, assurance, empathy and tangibles. Kano analysis distinguishes must-be, one-dimensional, attractive, indifferent and reverse attributes, recognizing that features do not affect satisfaction uniformly. Complaint management converts individual failures into opportunities for recovery, learning and systemic corrective action.

QFD provides a disciplined link between customer requirements and technical or process characteristics. Its calculations are prioritization aids, not exact measures of customer utility. Effective implementation requires representative research, ethical governance, accessibility, privacy protection, critical interpretation and continuous review. The ultimate purpose is not simply to collect opinions, but to design and control products and services that create reliable, fair and meaningful value.

Key Terms

Voice of the customer: Structured representation of customer needs, expectations and experiences.

Stakeholder: A person or group that can affect, be affected by, or perceive itself to be affected by an organization.

Perceived quality: A customer’s evaluation of quality based on experience and interpretation.

Requirement: A documented condition that must be satisfied.

Specification: A measurable statement defining a characteristic, limit or target.

Customer satisfaction: Evaluation formed from perceived performance in relation to expectations and importance.

Net Promoter Score: Percentage of promoters minus percentage of detractors.

SERVQUAL: A service-quality perspective involving reliability, responsiveness, assurance, empathy and tangibles.

Service recovery: Action taken to resolve a service failure and restore customer confidence.

Kano model: Framework classifying attributes according to their effect on satisfaction.

Must-be attribute: Requirement whose absence causes strong dissatisfaction.

One-dimensional attribute: Requirement for which better performance generally produces greater satisfaction.

Attractive attribute: Unexpected feature that can create delight.

Complaint management: System for receiving, resolving, analysing and learning from complaints.

Root cause: Underlying condition that permits a problem to occur or recur.

Quality Function Deployment: Method for translating customer requirements into technical and process priorities.

House of Quality: Matrix-based QFD tool linking customer requirements with technical characteristics.

WHATs: Customer requirements in a QFD analysis.

HOWs: Technical or process characteristics used to respond to customer requirements.

Traceability: Ability to link a requirement to its source, implementation and verification evidence.

Review Questions

Short Answer Questions

1. Distinguish between a customer and a stakeholder.
2. What is meant by translating the voice of the customer?
3. Identify the five dimensions commonly associated with SERVQUAL.
4. What is the purpose of complaint trend analysis?
5. Define a must-be requirement in the Kano model.

Long Answer Questions

1. Explain how qualitative and quantitative VOC methods should be combined in quality planning.
2. Discuss the strengths and limitations of satisfaction measurement and NPS.
3. Explain the structure and purpose of the House of Quality.
4. Describe an effective end-to-end complaint-management process.
5. Critically evaluate the usefulness of Kano analysis for prioritizing product or service improvements.

Analytical / Application Questions

1. A university service receives high ratings for staff courtesy but low ratings for response time and information accuracy. Design a measurement and improvement plan using satisfaction analysis, complaint data and QFD.
2. A company wants to replace all telephone support with an artificial-intelligence chatbot. Analyse the proposal from customer, stakeholder, accessibility, privacy and quality-governance perspectives.
3. Construct a small QFD relationship matrix for a service of your choice. State at least three customer requirements, three technical characteristics, your relationship scale and the resulting priorities.

Multiple Choice Questions with Answers

Question 1

Which statement best describes the voice of the customer?

A. A list of internal engineering tolerances

B. A structured representation of customer needs, expectations and experiences

C. A financial forecast of customer revenue

D. A regulatory inspection report

Correct Answer: B. A structured representation of customer needs, expectations and experiences

Question 2

If perceived performance is 3 and expectation is 5 on the same rating scale, what is the SERVQUAL-style gap?

A. −2

B. 0

C. +2

D. +8

Correct Answer: A. −2

Question 3

Which Kano category describes an attribute whose absence causes strong dissatisfaction but whose presence is largely taken for granted?

A. Attractive

B. Indifferent

C. Must-be

D. Reverse

Correct Answer: C. Must-be

Question 4

What is the principal purpose of complaint management?

A. To reduce the number of complaints recorded

B. To resolve individual problems and enable systemic learning

C. To transfer all responsibility to front-line staff

D. To replace customer research entirely

Correct Answer: B. To resolve individual problems and enable systemic learning

Question 5

In QFD, the WHATs represent:

A. Customer requirements

B. Production costs

C. Audit findings only

D. Supplier invoices

Correct Answer: A. Customer requirements

Question 6

Using NPS categories, a respondent scoring 8 is normally classified as a:

A. Promoter

B. Detractor

C. Passive

D. Complainant

Correct Answer: C. Passive

Question 7

Which requirement is most measurable?

A. The service should feel better

B. Customers should be happier

C. Ninety-five percent of requests should be acknowledged within 10 minutes

D. The interface should be modern

Correct Answer: C. Ninety-five percent of requests should be acknowledged within 10 minutes

Question 8

In the QFD example, what was the calculated priority for validation rate?

A. 50

B. 62

C. 63

D. 64

Correct Answer: D. 64

Question 9

Which practice creates a major risk in automated complaint triage?

A. Documenting model limitations

B. Providing human escalation

C. Checking for language and demographic bias

D. Automatically rejecting complaints from low-confidence categories

Correct Answer: D. Automatically rejecting complaints from low-confidence categories

Question 10

Why should customer averages be interpreted cautiously?

A. Averages always exaggerate dissatisfaction

B. They can conceal important differences among customer segments

C. They cannot be calculated from survey data

D. They are only valid for manufacturing organizations

Correct Answer: B. They can conceal important differences among customer segments