2. Process Architecture, Mapping and Customer Value

Introduction and Context

Operational excellence depends on an organisation’s ability to understand how work is actually performed and how that work creates value for customers and other stakeholders. Organisations are often described through departments—marketing, operations, finance, information technology, human resources and customer service. Customers, however, experience neither departments nor organisational charts. They experience processes: requesting a service, placing an order, receiving a product, making a payment, registering a complaint or obtaining technical support.

A process perspective therefore shifts attention from isolated activities to the complete flow of work across organisational boundaries. It asks: What triggers the work? Who performs it? What information, materials or decisions are required? Where are delays and errors introduced? Which activities create value from the customer’s point of view? Who owns the overall outcome?

Process architecture provides the structural view of an organisation’s processes. Process mapping provides a visual and analytical representation of how work moves through that architecture. Customer-value analysis ensures that improvement is not limited to internal efficiency but remains connected to customer requirements, experience and outcomes. Together, these disciplines form an important foundation for Lean management, Six Sigma, business process management, quality management and operational excellence.

Learning Objectives

After studying this chapter, students should be able to:

1. Explain process architecture, process hierarchy, process ownership, boundaries, inputs, outputs and stakeholders.

2. Distinguish core, support, management and enabling processes.

3. Select and apply SIPOC diagrams, flowcharts, value-stream maps, service blueprints and voice-of-customer methods.

4. Identify customer value, non-value-adding activity, waste, failure points and improvement opportunities.

5. Interpret process maps critically rather than treating them as merely graphical descriptions.

6. Analyse an end-to-end process and recommend improvement actions with appropriate governance and measures.

Seeing the Organisation as a System of Processes

A process is a linked set of activities that transforms defined inputs into outputs for a customer or recipient. The customer may be external, such as a purchaser of a product, or internal, such as a department receiving an approved purchase order. A process is more than a list of tasks. It has a purpose, a trigger, resources, controls, decision logic, performance measures and an accountable owner.

The transformation may be physical, informational, transactional or experiential. Manufacturing a component transforms materials; approving a loan transforms information and risk assessments into a decision; treating a patient transforms clinical information and resources into improved health; resolving a complaint transforms a dissatisfied experience into a response and, ideally, restored confidence.

A process should be distinguished from a function. A function groups people according to expertise or managerial responsibility. A process crosses functions to produce an outcome. Confusing the two creates local optimisation: each department may achieve its own targets while the overall customer journey becomes slow, costly or unreliable.

Process hierarchy and architecture

Process architecture is a structured model showing how processes relate to one another and to organisational objectives. A typical hierarchy contains several levels:

• The value chain or process landscape presents major end-to-end streams, such as develop offering, acquire customer, fulfil order and provide after-sales support.
• A major process or process group is a substantial business capability, such as order fulfilment.
• A subprocess decomposes the major process, such as validate order, allocate inventory, arrange shipment and confirm delivery.
• An activity describes a meaningful unit of work, such as check customer credit status.
• A task is a specific action performed by an individual or system, such as enter the credit reference number.

The number of levels is not universal. A useful architecture is governed by purpose: senior leaders need a sufficiently aggregated view for strategic decisions, while improvement teams require enough detail to understand variation and failure. Excessive detail produces an unreadable map; insufficient detail conceals handoffs and causes.

Processes are commonly classified as core, support, management and enabling processes. Core processes directly deliver value to external customers, for example design, production, delivery and service. Support processes provide resources or capabilities, such as procurement, recruitment, facilities and payroll. Management processes establish direction and control, including strategic planning, performance review, risk management and governance. Enabling processes, often supported by technology, connect or strengthen other processes; data management and workflow administration are examples. The categories are analytical conveniences rather than rigid truths. A process such as information security may be enabling, supportive and governance-related at the same time.

Ownership, boundaries and accountability

Process ownership is accountability for the performance and continuing improvement of an end-to-end process. The owner may not personally perform every activity. Rather, the owner defines the process purpose, requirements, measures, controls, interfaces and improvement priorities, and coordinates the functions involved.

Boundaries determine where analysis begins and ends. A process may begin when a customer submits an order and end when the order is delivered and accepted. If the boundary begins only when the warehouse receives the order, important causes in sales or order entry may be excluded. If it ends at dispatch, customer experience after delivery is ignored. Boundaries should therefore be explicit and justified by the improvement question.

A process has multiple stakeholders. Customers receive the output; suppliers provide inputs; employees execute or control activities; regulators establish obligations; owners allocate resources; and technology providers may influence performance. Stakeholder analysis is particularly important where requirements conflict. For example, a customer may value speed, a regulator may require additional verification, and an organisation may need cost control. Good process design makes such trade-offs visible rather than allowing them to emerge as informal workarounds.

From Customer Requirements to Process Requirements

Customer value is not identical to internal activity or cost. An activity creates customer value when it contributes to an outcome the customer needs, is performed correctly the first time, and changes the product or service in a way for which the customer would reasonably recognise value. Some activities do not directly create value but are necessary under current conditions, such as regulatory checks, safety inspections or legally required records. Other activities are pure waste: they consume resources without improving the output or satisfying a necessary requirement.

This distinction should be applied carefully. A customer may not observe an activity but may depend on its result. Preventive maintenance, cybersecurity controls and quality verification may be invisible yet essential. The appropriate question is not simply whether the customer sees the activity, but whether the activity is necessary to deliver a conforming, safe and legitimate outcome.

Voice of customer (VOC) methods translate expressed and unexpressed needs into operational requirements. Direct methods include interviews, surveys, focus groups, complaint analysis, observation, usability testing and customer panels. Indirect sources include returns, warranty data, service logs, social-media comments, search behaviour and frontline employee observations. Quantitative data reveals frequency and variation; qualitative inquiry explains meaning and context.

A sound VOC process distinguishes customer statements from underlying needs. “I need a faster approval” may represent a need for certainty, reduced waiting, better communication or avoidance of repeated submissions. The statement can be converted into a critical-to-quality requirement, such as “application status is communicated within one business day” or “complete applications receive a decision within the published service standard.” Requirements should be specific enough to measure and trace to process performance.

Customer journeys add another perspective. They describe the customer’s experience across touchpoints, including activities performed outside the organisation. A journey may reveal that an internally efficient process still creates customer effort because the customer must repeat information, switch channels or wait without visibility.

Choosing a Map That Answers the Right Question

No single mapping technique is universally superior. The map should be selected according to the purpose, audience and type of process.

SIPOC: establishing the analytical frame

SIPOC stands for Suppliers, Inputs, Process, Outputs and Customers. It is a high-level scoping tool, normally prepared before detailed mapping. A SIPOC for online order fulfilment might identify suppliers such as customers, inventory systems and payment providers; inputs such as order data, stock information and payment authorisation; a small number of process stages; outputs such as confirmed order, shipped product and delivery notification; and customers such as the purchaser and the recipient.

SIPOC prevents premature immersion in task-level detail. It helps a team agree on what process is being studied, who receives the output and what inputs are critical. It also exposes assumptions—for example, whether “customer” means the person placing the order, the user of the product, or an internal distribution centre.

Flowcharts and cross-functional maps

A flowchart represents sequence, decisions, loops and sometimes delays. Standard symbols are less important than consistent interpretation. A cross-functional flowchart, or swimlane map, places activities in lanes for departments, roles, systems or suppliers. It makes handoffs visible and helps identify unclear responsibility, duplicated entry and communication loss.

A current-state flowchart should describe reality, including exceptions and rework, rather than an ideal procedure. Interviews alone may produce the documented process rather than the work actually performed. Observation, transaction records and discussion with frontline employees help reveal informal steps.

Value-stream mapping

Value-stream mapping (VSM) examines the end-to-end flow of material and information required to deliver a product or service. It commonly records process time, waiting time, inventory or queue levels, information signals, demand and constraints. The central distinction is between processing time and lead time. A transaction may require only ten minutes of actual work but remain in the system for five days because of queues, batching, approvals or incomplete information.

A current-state VSM identifies the present pattern of flow; a future-state map expresses a designed condition based on demand, capability and improvement principles. Useful questions include: Where does work wait? Where are batches created? How often is information re-entered? Where does rework occur? Is work released according to demand or according to local convenience? VSM is especially useful for Lean improvement, but it must be adapted to knowledge work and services where work-in-process may be electronic or invisible.

Service blueprinting

A service blueprint maps the customer’s actions, frontstage employee interactions, backstage activities, support processes and physical or digital evidence. It is particularly valuable for services because production and consumption may occur simultaneously and the customer often participates in the process.

For example, in a hospital appointment process, the customer actions may include searching, booking, attending and receiving results. Frontstage interactions include reception and clinician communication. Backstage activities include scheduling, records retrieval and laboratory processing. Support processes include information systems, staffing and facility management. The blueprint can reveal a failure invisible in an internal process map: the laboratory may complete its work on time, but the patient receives no understandable notification.

Digital process discovery

Event logs from enterprise systems can reconstruct actual process paths, frequencies, cycle times and rework. Process mining can identify variants, bottlenecks and deviations from a prescribed model. It is powerful when records are complete and events are consistently defined. It is less reliable when work occurs outside systems, timestamps do not represent real activity, or a single case is split across multiple identifiers. Digital analysis should therefore complement, not automatically replace, engagement with people who perform and experience the process.

A disciplined mapping method

A practical mapping exercise begins by defining the improvement question, customer, process owner, start event, end event and expected output. The team should then assemble representatives from relevant functions and, where possible, include a customer or customer-facing employee. A SIPOC establishes scope before the team maps the current state.

The current-state map is developed using evidence. The team records activities, decisions, handoffs, queues, systems, defects, exceptions and rework. It should distinguish what the procedure says from what actually happens. Data is then added: volume, demand, process time, waiting time, first-pass yield, error rates, backlog and customer-perceived effort. The map is validated through walkthroughs with the people who perform the work.

Analysis should focus on the relationship between customer requirements and process performance. Questions include:

• Which outputs fail to meet explicit or implicit customer requirements?
• Where is ownership unclear?
• Which handoffs create delay or information loss?
• Where are decisions based on incomplete or inconsistent data?
• Which controls prevent failure, and which merely detect it later?
• What variation is caused by customer demand, and what variation is created internally?

The future state should not be a wish list. It should reflect a reasoned design: simplified flow, fewer handoffs, clearer decision rules, suitable automation, balanced controls, visible status and measures linked to customer outcomes. Implementation can then be organised through experiments, standard work, training, technology changes and governance reviews.

An illustrative case: redesigning a university admissions process

Background

The following is an illustrative case. A university receives applications for a postgraduate programme through an online portal. The process involves applicants, admissions staff, academic departments, finance, information technology and external document-verification providers.

Situation and problem

Applicants complain that decisions are unpredictable and that they must submit the same information more than once. The university’s reported average processing time is twelve working days, but some applications remain open for more than a month. Each department believes that another unit causes the delay. Senior management asks for an end-to-end improvement analysis without compromising academic standards or data protection.

Analysis

The team first establishes a SIPOC. The trigger is a completed application; the desired output is an accurate and communicated admission decision. The applicant is the primary external customer, but programme administrators, academic reviewers and the student-records office are also recipients of outputs.

A swimlane map shows seven handoffs. Admissions staff manually re-enter data from uploaded documents into a student system. Academic reviewers receive applications in weekly batches, even when complete applications are available. Missing-document requests are sent by email, but applicants cannot see a consolidated status. A document-verification provider returns results through a separate portal, creating an additional queue.

A service blueprint adds the applicant perspective. The applicant’s principal pain point is not only elapsed time; it is uncertainty. The applicant cannot tell whether the application is awaiting academic review, verification or correction. VOC interviews also reveal that applicants value a reliable decision date more than an occasional very rapid decision.

The team compares process time with lead time. Most applications require less than two hours of staff and reviewer effort, but waiting accounts for most elapsed time. Defects include incomplete submissions, duplicate data entry and inconsistent interpretation of programme prerequisites. These defects create rework and make performance appear more variable than the underlying academic assessment.

Approach and solution

The future-state design introduces a completeness check at submission, using clear programme-specific requirements and immediate feedback. A single case identifier links the application portal, verification service and student system. Applications that pass completeness checks are routed continuously rather than in weekly batches. A dashboard displays status to applicants and staff, while exception rules direct unusual cases to specialist review.

The university retains human academic judgement for admissions decisions. Automation is used for data validation, routing and status communication, not for making opaque decisions about applicant suitability. A process owner is appointed from admissions leadership, with defined responsibilities for academic review, information security and service performance.

Measures are redesigned. In addition to average processing time, the university tracks the percentage of complete applications decided within the published standard, first-pass completeness, number of handoffs, rework rate, applicant contacts about status and variation between programmes.

Outcome and learning points

The redesign demonstrates that the visible problem—slow decisions—was connected to less visible causes: batch processing, fragmented systems, unclear status and inconsistent requirements. The case does not imply that every process should be fully automated. Rather, appropriate digital support removes avoidable administration while preserving expert judgement and governance.

The principal learning points are that process boundaries should follow the customer outcome, current-state maps must reflect actual work, waiting often dominates total lead time, and VOC can redefine the problem from “speed” to “certainty and transparency.” A local departmental improvement would not have solved the end-to-end experience.

Challenges, risks and professional considerations

Mapping can create a false sense of precision. A visually attractive map may still omit exceptions, informal work, emotional experience or political realities. Teams may also defend their department, describe the ideal process, or treat the map as an audit of individual performance. Effective facilitation frames mapping as system improvement rather than blame allocation.

Over-mapping is another common error. A team may spend weeks documenting every click without identifying the customer requirement or decision to be made. The level of detail should be proportionate to risk, complexity and improvement purpose. Conversely, an excessively high-level map may hide the precise handoff where failure occurs.

Maps can also become obsolete. Changes in technology, regulation, staffing, suppliers or customer behaviour can invalidate assumptions. Process governance should specify who maintains the map, how changes are approved and how performance is reviewed.

Ethical and governance issues are significant where maps involve personal, financial, health or employee data. Access should be restricted according to role, data should be collected only for legitimate purposes, and retention and sharing should comply with applicable law and organisational policy. Process improvement must not use mapping to conceal accountability, intensify unreasonable work demands or disadvantage particular customer groups. Accessibility, language, digital exclusion and reasonable accommodation should be considered in service design.

Automation introduces additional risks. A system may accelerate a flawed process, reproduce biased rules, or make decisions difficult to explain. Before automation, the organisation should simplify the process, clarify decision rights, validate data quality and establish human escalation. Performance dashboards should not reward speed at the expense of accuracy, safety or customer dignity.

Managerial application and best practice

Managers should treat process architecture as an operating model, not merely a documentation exercise. A coherent architecture connects strategic objectives to end-to-end process owners, customer requirements, measures and improvement portfolios. It helps leaders decide where standardisation is beneficial and where local adaptation is necessary.

Strong practice generally includes:

• defining processes around outcomes rather than departmental boundaries;
• naming one accountable owner while preserving shared execution;
• starting with customer requirements and critical-to-quality characteristics;
• mapping the current state with evidence and frontline participation;
• distinguishing value, necessary non-value-adding work and avoidable waste;
• measuring lead time, flow, quality, reliability and customer effort together;
• treating exceptions and rework as sources of learning;
• validating proposed changes through controlled experiments;
• integrating process maps with risk controls, training and technology design; and
• reviewing maps and measures after significant organisational or regulatory change.

Current developments include process mining, digital twins of operations, workflow orchestration, robotic process automation and AI-supported analysis of customer feedback. These technologies can improve visibility and decision support, but they do not remove the need for sound process definitions, accountable ownership, representative data and ethical oversight. The enduring principle is that technology should serve a clearly understood value proposition rather than automate activity for its own sake.

Chapter Summary

Process architecture shows how organisational processes are arranged and connected to strategic and customer outcomes. Process hierarchy ranges from value chains to tasks, while process ownership establishes accountability across functional boundaries. Clear boundaries, inputs, outputs, suppliers and stakeholders are essential to meaningful analysis.

SIPOC provides high-level scope; flowcharts and swimlane maps reveal sequence, decisions and handoffs; value-stream maps expose lead time, queues and information flow; and service blueprints connect customer actions with frontstage, backstage and support activities. VOC methods convert customer experiences and expectations into measurable process requirements. Effective improvement compares these requirements with actual performance, identifies waste and failure, and designs a future state supported by appropriate measures, governance and technology.

Process mapping is not an end in itself. Its value lies in enabling better decisions about ownership, flow, controls, customer experience and improvement priorities. Ethical data use, human judgement, accessibility and accountability must remain central when processes are digitised or automated.

Key Terms

Process: A linked set of activities that transforms inputs into outputs for a customer or recipient.

Process architecture: A structured representation of how an organisation’s processes relate to one another and to objectives.

Process hierarchy: The levels used to decompose a value chain into processes, subprocesses, activities and tasks.

Process owner: The person accountable for end-to-end process performance, governance and improvement.

Process boundary: The defined start and end points of a process under analysis.

Stakeholder: A person or group that affects, is affected by, or has an interest in a process.

Core process: A process that directly delivers value to an external customer.

Support process: A process that provides resources or services to enable other processes.

SIPOC: A high-level view of Suppliers, Inputs, Process, Outputs and Customers.

Flowchart: A visual representation of activities, sequence, decisions and loops.

Swimlane map: A cross-functional map that assigns activities to roles, departments or systems.

Value-stream mapping: A method for analysing end-to-end material and information flow, including processing and waiting.

Service blueprint: A service map linking customer actions, frontstage interactions, backstage work and support processes.

Voice of customer: Systematic collection and interpretation of customer needs, expectations and experiences.

Critical-to-quality requirement: A measurable characteristic essential to customer satisfaction or conformity.

Value-adding activity: Work that contributes directly to an outcome the customer needs and recognises.

Rework: Repetition or correction required because work was incomplete, incorrect or nonconforming.

Lead time: Elapsed time from process start to completion, including processing and waiting.

Process mining: Use of event-log data to analyse actual process paths, timing and variation.

Local optimisation: Improvement of one function or activity that harms overall process performance.

Review Questions

Short Answer Questions

1. What is the difference between a process and a function?

2. Define process ownership and explain why it may cross departmental boundaries.

3. What information does a SIPOC diagram provide?

4. Distinguish processing time from lead time.

5. What is meant by a critical-to-quality requirement?

Long Answer Questions

1. Explain process hierarchy and discuss how different levels support strategic and operational decision-making.

2. Compare flowcharts, value-stream maps and service blueprints in terms of purpose, content and application.

3. Describe a disciplined method for mapping a current-state process and designing a future state.

4. Explain how VOC methods can transform customer statements into measurable process requirements.

5. Discuss the governance, ethical and data-related risks associated with digital process mapping and automation.

Analytical / Application Questions

1. A claims process has ten minutes of processing time but an average lead time of eight days. Using process-mapping concepts, identify likely causes and propose data that should be collected before redesign.

2. Design a SIPOC and recommend a complementary mapping method for a university student-grievance process. Justify your choices.

3. A call centre reduces average handling time but receives more repeat calls and complaints. Analyse this outcome using customer value, process measures and local-optimisation concepts.

22. Multiple Choice Questions with Answers

Question 1

Which statement best describes process ownership?

A. Responsibility for performing every task in a process personally

B. Accountability for end-to-end performance, governance and improvement

C. Authority limited to the department with the largest number of activities

D. Responsibility for maintaining only the process flowchart

Correct Answer: B. Accountability for end-to-end performance, governance and improvement

Question 2

What is the primary purpose of a SIPOC diagram?

A. To document every task and keystroke in a process

B. To calculate the statistical capability of a process

C. To establish high-level process scope and identify suppliers, inputs, outputs and customers

D. To replace customer interviews with internal assumptions

Correct Answer: C. To establish high-level process scope and identify suppliers, inputs, outputs and customers

Question 3

A value-stream map shows that a transaction requires 20 minutes of work but remains in the system for four days. This most directly indicates a problem with:

A. Lead time caused largely by waiting or queues

B. Customer segmentation

C. Product design complexity only

D. The definition of an output

Correct Answer: A. Lead time caused largely by waiting or queues

Question 4

Which feature distinguishes a service blueprint from a conventional internal flowchart?

A. It excludes customer participation from the process

B. It focuses only on financial transactions

C. It links customer actions with frontstage, backstage and support activities

D. It can be used only for manufacturing processes

Correct Answer: C. It links customer actions with frontstage, backstage and support activities

Question 5

An applicant says, “I need a faster decision,” but interviews reveal that uncertainty about status is the main concern. The best process-improvement response is to:

A. Reduce all review activities regardless of their purpose

B. Translate the underlying need into measurable requirements for decision timing and status communication

C. Ignore the statement because it is subjective

D. Measure only average staff processing time

Correct Answer: B. Translate the underlying need into measurable requirements for decision timing and status communication

Question 6

Which mapping practice is most likely to produce an unreliable current-state map?

A. Validating the map with frontline employees

B. Including exceptions and rework

C. Describing the documented ideal process without checking actual work

D. Adding lead-time and defect data

Correct Answer: C. Describing the documented ideal process without checking actual work

Question 7

A department reduces its own workload, but customers now repeat information to three different departments. This is an example of:

A. Critical-to-quality improvement

B. Local optimisation harming end-to-end performance

C. Successful service blueprinting

D. Necessary non-value-adding work

Correct Answer: B. Local optimisation harming end-to-end performance

Question 8

In process mining, the principal source used to reconstruct process paths is:

A. Event-log data from information systems

B. Only informal employee opinions

C. A high-level organisation chart

D. A customer satisfaction slogan

Correct Answer: A. Event-log data from information systems

Question 9

Which activity may be non-value-adding from the customer’s direct perspective but still necessary?

A. Repeated correction of an avoidable data-entry error

B. Duplicating information across systems without control need

C. A legally required safety or compliance check

D. Waiting for a weekly batch when continuous flow is possible

Correct Answer: C. A legally required safety or compliance check

Question 10

Before automating an unstable approval process, an organisation should first:

A. Increase the number of automated decisions immediately

B. Simplify the process, clarify decision rules and validate data quality

C. Remove all human escalation routes

D. Measure only speed and transaction volume

Correct Answer: B. Simplify the process, clarify decision rules and validate data quality