2. Insurance Operating Models and Value Chains

Introduction and Context

An insurer does not create value through a single transaction. Value emerges from a connected system of activities: product design, risk selection, pricing, distribution, policy issuance, premium collection, claims settlement, investment, customer service, compliance, and learning from operational and financial data. The arrangement through which these activities are designed, governed, resourced, and performed is the insurer’s operating model.

Insurance operating models matter because the product is intangible, payment and performance are separated in time, and the customer often evaluates the insurer most critically at the point of claim. A technically sound product can fail commercially if underwriting is slow, policy records are inaccurate, distribution incentives are poorly controlled, or claims processes are opaque. Conversely, an efficient operating model can improve customer experience, control expenses, strengthen risk management, and support profitable growth.

The value-chain perspective complements the operating-model perspective. A value chain examines how activities contribute to customer and enterprise value. An operating model explains how those activities are organized and executed. This chapter analyses both perspectives and applies them to life, health, and general insurance. It also considers centralized and decentralized structures, shared services, outsourcing, insurtech partnerships, digital capabilities, governance, and strategic choices.

Learning Objectives

After studying this chapter, students should be able to:

1. Explain the meaning and strategic significance of an insurance operating model.

2. Map the principal activities in an insurance value chain.

3. Distinguish functional, process-based, centralized, decentralized, and hybrid structures.

4. Evaluate shared services, outsourcing, and insurtech partnerships.

5. Compare operating-model requirements in life, health, and general insurance.

6. Analyse operational risks, governance responsibilities, technology implications, and performance measures.

7. Recommend an appropriate operating model for a given insurance business situation.

From Insurance Activities to an Operating Model

An operating model is the practical design that connects strategy with execution. It specifies what work must be done, where it will be done, who is accountable, which capabilities and technologies are required, and how decisions and controls will operate. It is broader than an organisation chart. Two insurers may have similar departments but very different operating models because one relies on local authority and manual processing while the other uses centralized platforms, automated decisions, and external service providers.

A useful operating-model description includes six interdependent dimensions:

• Processes: the sequence of activities through which business outcomes are delivered.
• Organisation: business units, functions, reporting lines, and accountability.
• People and capabilities: technical knowledge, judgement, professional skills, and capacity.
• Technology and data: core systems, integration, analytics, automation, and information quality.
• Governance: decision rights, controls, risk ownership, escalation, and oversight.
• Location and sourcing: the physical or digital location of work and the extent to which it is performed internally or externally.

The operating model must follow the insurer’s strategic position. A specialist commercial insurer may compete through expert underwriting and relationship management. A mass-market motor insurer may compete through low expense ratios, rapid claims settlement, and digital distribution. A life insurer may emphasize long-term administration, actuarial discipline, investment management, and policyholder servicing. There is therefore no universally superior operating model.

The Insurance Value Chain

The insurance value chain begins before a policy is sold and continues throughout the policy and claims life cycle. It also includes support activities that enable the primary chain.

Market insight and product development

Insurers identify customer needs, regulatory requirements, competitor offerings, affordability constraints, and emerging risks. Product development converts these insights into policy wording, benefits, exclusions, eligibility criteria, pricing assumptions, distribution rules, and service commitments. Product governance is essential because unclear wording or poorly designed benefits can create conduct risk, disputes, and adverse selection.

Distribution and advice

Distribution may involve tied agents, brokers, bancassurance, direct digital channels, affinity partners, motor dealers, employers, comparison platforms, or embedded insurance arrangements. Distribution activities include prospect identification, quotation, advice, illustration, disclosure, sales confirmation, and commission administration. The operating model must control mis-selling, conflicts of interest, customer consent, and the accuracy of information supplied at point of sale.

Underwriting and risk selection

Underwriting assesses whether a risk should be accepted, on what terms, and at what price. It may use proposal forms, medical evidence, inspection reports, external data, rules engines, actuarial models, or specialist judgement. Effective underwriting balances growth with portfolio quality. Excessive central control may slow decisions, while excessive local discretion may produce inconsistent risk selection.

Policy administration

Policy administration includes data capture, identity verification, document generation, premium billing, endorsements, renewals, cancellations, beneficiary changes, reinstatements, and record maintenance. Accurate administration is particularly important because the policy record is the operational expression of the contract. Errors can affect coverage, premiums, claims, financial reporting, and legal defensibility.

Premium collection and financial administration

The insurer must collect premiums, allocate receipts, manage instalments, reconcile accounts, administer refunds, and monitor arrears. Premium processes interact with underwriting, policy status, commissions, accounting, and customer service. Weak reconciliation can cause inaccurate policy status, leakage, complaints, and misstated financial information.

Claims management

Claims management includes notification, registration, coverage verification, investigation, assessment, reserving, approval, settlement, recovery, fraud detection, and closure. Claims are both a cost centre and a major source of customer value. Claims data also informs pricing, product design, risk engineering, fraud controls, and capital management.

Investment and capital management

For many insurers, especially life insurers, premium income creates investable funds. Investment management seeks to match assets with liabilities while meeting liquidity, regulatory, risk, and return requirements. Capital management links underwriting risk, market risk, credit risk, operational risk, and solvency requirements. These functions may be internal or provided through specialist entities, but accountability remains with the insurer and its governing body.

Customer service and relationship management

Service includes enquiries, complaints, policy changes, status updates, documentation, and communication during claims. Omnichannel service requires consistent information across call centres, branches, agents, websites, mobile applications, and messaging platforms. Service quality is not merely a marketing concern; it affects retention, conduct outcomes, expense levels, and regulatory reputation.

Enabling activities

Finance, actuarial services, legal affairs, compliance, risk management, human resources, procurement, information security, data management, and technology support the primary chain. These functions may not be visible to customers, but they determine whether the insurer can operate safely and sustainably.

A value-chain analysis should not assume that every activity adds value in the same way. Some activities create customer value directly, such as transparent advice and fair claims settlement. Others protect value by preventing errors, fraud, regulatory breaches, or financial instability. The appropriate question is not simply whether an activity is expensive, but whether its cost is justified by the risk reduced or value created.

Organisational Logics: Functional, Process-Based, and Product Structures

A functional structure groups expertise into departments such as underwriting, claims, actuarial, finance, sales, and information technology. It supports professional specialization, standard setting, and economies of scale. Its weakness is the possibility of functional silos. A customer’s claim may pass through several departments, none of which owns the complete outcome.

A process-based structure organizes around end-to-end journeys such as quote-to-bind, policy issuance, premium-to-cash, first-notice-of-loss-to-settlement, or complaint-to-resolution. A process owner is accountable for performance across functional boundaries. This structure improves visibility of delays, handoffs, rework, and customer effort. It requires strong process data and may challenge traditional departmental authority.

A product or market structure creates dedicated units for life, health, motor, property, commercial, or geographic markets. It encourages market responsiveness and clear profit accountability. However, duplicated technology, actuarial resources, procurement, and service capabilities may increase cost.

Most mature insurers use a hybrid. Product units retain underwriting and market expertise, while common platforms, finance, procurement, cybersecurity, and selected service activities are shared. The design should reflect the degree of product similarity, regulatory separation, customer needs, risk concentration, and required speed of decision-making.

Centralization, Decentralization, and the Hybrid Choice

Centralization concentrates authority, expertise, systems, and processing in a corporate or regional centre. It can reduce duplication, improve consistency, strengthen controls, and support investment in advanced technology. Centralized claims or underwriting hubs may also create specialist communities of practice.

Decentralization delegates authority to branches, subsidiaries, product units, or local teams. It can improve responsiveness, local market knowledge, broker relationships, and adaptation to jurisdictional conditions. Its risks include inconsistent pricing, variable service, duplicated infrastructure, weak oversight, and difficulty aggregating data.

The central issue is not whether centralization is inherently good. It is which decisions should be centralized and which should remain close to the customer or risk. A practical decision framework considers:

1. Standardization: Are the activity and rules largely common across markets?

2. Scale: Are meaningful cost or capability benefits available through concentration?

3. Local responsiveness: Does the activity depend on local law, culture, relationships, or physical assessment?

4. Risk sensitivity: Would inconsistent decisions create material underwriting, conduct, or operational risk?

5. Customer proximity: Does speed, empathy, or local judgement materially affect the outcome?

6. Data and technology feasibility: Can systems provide secure, timely access across locations?

A hybrid model may centralize platforms, policy standards, actuarial methodology, procurement, and quality assurance while decentralizing relationship management, complex commercial underwriting, field investigation, or jurisdiction-specific compliance. Decision rights must be documented; otherwise a hybrid model becomes an ambiguous structure in which accountability is shared but ownership is unclear.

Shared Services and Centres of Expertise

A shared-service centre performs common activities for multiple business units. Typical examples include finance operations, human resources administration, procurement, document processing, contact centres, information security monitoring, and certain claims or underwriting services. A centre of expertise, by contrast, concentrates scarce specialist knowledge such as actuarial modelling, fraud analytics, catastrophe risk, regulatory interpretation, or data science.

Shared services can lower unit costs and improve standardization, but only if service levels and accountability are explicit. Internal customers should know what service is provided, within what time, at what quality, and with which escalation route. Service-level agreements may include turnaround time, accuracy, availability, complaint resolution, and data-security measures.

Common implementation failures include transferring work without redesigning the process, measuring volume rather than outcomes, underestimating transition costs, and treating business-unit users as passive recipients. A shared service should be managed as an operating capability, not merely as a cost-cutting exercise.

Outsourcing and Insurtech Partnerships

Outsourcing occurs when an external provider performs an activity that could otherwise be performed internally. Insurance examples include claims administration, medical examination networks, roadside assistance, cloud hosting, document management, contact centres, fraud analytics, actuarial support, and payment processing. Insurtech partnerships may provide artificial-intelligence tools, digital distribution, identity verification, telematics, application programming interfaces, automated valuation, or embedded-insurance platforms.

The decision to outsource should evaluate more than price. Relevant criteria include strategic importance, customer impact, data sensitivity, regulatory expectations, operational resilience, substitutability of the provider, concentration risk, intellectual property, transition difficulty, and exit feasibility. Activities involving key underwriting judgement, claims fairness, customer data, or regulatory reporting require particularly strong oversight.

Outsourcing does not transfer ultimate accountability. The insurer must retain knowledge sufficient to challenge the provider, monitor performance, investigate incidents, and make informed strategic decisions. Contracts should define scope, service levels, audit rights, data ownership, confidentiality, subcontracting, business continuity, incident notification, regulatory access, recovery obligations, and termination assistance.

Insurtech partnerships can accelerate innovation, but integration risk is substantial. A technically impressive tool may produce biased decisions, unexplained outcomes, poor data lineage, or unreliable results outside its training population. Pilot projects should therefore use controlled populations, clear success measures, human review, privacy safeguards, and a plan for model validation and withdrawal.

Life, Health, and General Insurance: Different Operating Imperatives

Life insurance

Life insurance often involves long-duration contracts, complex benefits, medical and financial underwriting, beneficiary administration, policy loans, surrender values, bonuses, annuities, and investment-linked features. Operating models must preserve records over decades and manage changes in legislation, mortality assumptions, investment conditions, and customer circumstances. Actuarial, investment, finance, and policy administration functions are closely interconnected.

Straight-through processing is useful for simple products, but exceptions require skilled review. Service design must address vulnerable customers, bereavement, beneficiary disputes, lapses, reinstatements, and long-term communication. Data quality and system longevity are critical because migration errors can affect contractual rights over an extended period.

Health insurance

Health insurance involves relationships among members, employers, providers, administrators, brokers, and healthcare networks. Core processes include eligibility, pre-authorisation, provider contracting, medical coding, utilisation review, billing, claims adjudication, care management, and fraud detection. Speed and accuracy are important because delays may affect access to treatment.

Health operations must protect sensitive personal and medical information. Automated claims or authorisation systems require clinical governance, explainability, appeal mechanisms, and controls against discriminatory outcomes. Provider-network management is as important as policy administration because customer value depends partly on the availability and quality of care.

General insurance

General insurance commonly covers motor, property, marine, liability, engineering, travel, and commercial risks. Policies are often shorter-term and require frequent quotation, renewal, endorsement, and claims activity. Operating models must handle high transaction volumes for personal lines while retaining specialist expertise for complex commercial and specialty risks.

Claims may depend on physical inspection, repair networks, legal assessment, catastrophe response, salvage, and recoveries. Location can therefore matter more than in some life processes. Catastrophe events create sudden volume shocks, making workforce flexibility, supplier networks, digital notification, and resilient claims platforms important.

The distinction among classes should not become a rigid separation. Shared capabilities such as identity management, payments, cybersecurity, procurement, and data governance may serve all classes, while product-specific expertise remains essential.

Measuring Operating-Model Performance

Financial measures remain important, but a balanced assessment combines cost, quality, speed, risk, and customer outcomes. Relevant indicators include expense ratio, cost per policy or claim, quote-to-bind conversion, first-contact resolution, policy-issuance turnaround, claims cycle time, settlement accuracy, complaint frequency, renewal retention, leakage, fraud savings, rework, system availability, outsourcing incidents, and employee capability measures.

Metrics must be interpreted carefully. A low claims cost may reflect efficiency, but it may also indicate unjustified settlement delays. A high automation rate may hide inappropriate approvals or exclusions. A useful dashboard links operational indicators to strategic outcomes and includes leading indicators such as backlog growth, data-quality exceptions, control failures, and employee turnover.

Process mining, customer-journey analytics, predictive models, and real-time operational dashboards can reveal bottlenecks and variation. However, analytics should support managerial judgement rather than replace governance. Data definitions, lineage, access controls, model validation, and fairness testing are essential.

Governance, Ethics, and Operational Risk

Operating-model design raises governance questions about accountability, fairness, privacy, resilience, and professional conduct. The board and senior management remain responsible for the soundness of the insurer’s operations even when activities are decentralized, shared, automated, or outsourced.

Important risks include:

• inaccurate or incomplete policy and claims data;
• unauthorized access, cyberattack, and privacy breaches;
• model error, bias, or lack of explainability;
• excessive dependence on a vendor or cloud provider;
• unclear ownership at process handoffs;
• inadequate business continuity during catastrophes or system outages;
• conflicts in distribution and incentive arrangements;
• claims unfairness, vulnerable-customer harm, and poor complaint handling;
• regulatory non-compliance and weak record retention.

A sound control environment assigns first-line ownership to business processes, independent challenge to risk and compliance functions, and assurance to internal audit. Segregation of duties, approval limits, exception monitoring, audit trails, reconciliations, quality reviews, and incident reporting should be proportionate to risk. Ethical design requires the insurer to consider not only whether a decision is legally permissible, but also whether it is understandable, fair, and consistent with the purpose of insurance.

Illustrative Case: Redesigning Claims Operations at Meridian General Insurance

Background

Meridian General Insurance is a fictional mid-sized insurer offering motor and household insurance through brokers, agents, and digital channels. Its claims department is organized by function: notification, assessment, payments, recovery, and complaints. Each function has separate targets and systems.

Situation and problem

Claim volumes have increased, while customers report repeated requests for information and inconsistent settlement times. Management observes that simple motor claims are handled by experienced assessors, but claimants wait several days before assignment. Complex claims are sometimes routed to teams lacking specialist authority. The insurer is considering a technology vendor offering automated damage assessment and triage.

Analysis

A value-chain review shows that the main problem is not simply staff capacity. The process contains multiple handoffs, duplicate data entry, and functional targets that reward local completion rather than end-to-end settlement. Claims information is not consistently available to customer-service staff. The proposed technology could improve triage, but historical images are incomplete and the vendor’s model has not been validated for all vehicle types. Outsourcing the decision entirely would create conduct and accountability risks.

Approach

Meridian adopts a hybrid process-based model. It creates an end-to-end claims owner, redesigns the journey into express, standard, and complex pathways, and centralizes common digital intake and payment controls. Local teams retain authority for complex property assessments and broker-sensitive cases. The technology is introduced first as a triage and damage-estimation aid, with human review for exceptions and adverse decisions.

The insurer establishes measures for cycle time, accuracy, customer complaints, reopen rates, fraud referrals, and fairness across customer groups. The vendor contract includes audit access, data-security requirements, model-performance reporting, incident notification, and exit support. Staff are trained to challenge automated recommendations rather than accept them mechanically.

Outcome

In this illustrative case, the redesigned model reduces avoidable handoffs and gives managers visibility over the complete claim journey. More importantly, it clarifies accountability: the technology supports assessment, the claims function owns the decision, and the insurer retains responsibility for customer outcomes. The model also provides a basis for continuous improvement through claims data and post-settlement review.

Learning points

The case demonstrates that operating-model improvement begins with process diagnosis, not technology purchase. Centralization is valuable for common controls and information, while local expertise remains important for complex risks. Automation should be governed as a decision-support capability, and performance should be judged through customer, risk, quality, and financial outcomes together.

Current Developments and Best-Practice Direction

Insurance operating models are increasingly shaped by cloud platforms, application programming interfaces, embedded insurance, digital identity, telematics, robotic process automation, generative artificial intelligence, ecosystem partnerships, and advanced analytics. These developments can reduce manual work and enable more personalized products, but they also increase dependence on data, models, vendors, and digital infrastructure.

Emerging practice is moving toward modular platforms, event-driven integration, real-time monitoring, reusable services, and product teams that combine business and technology expertise. Insurers are also strengthening resilience testing, third-party risk management, model governance, and responsible use of artificial intelligence.

Best practice is not the maximum use of automation or the lowest apparent operating cost. It is the disciplined alignment of strategy, customer needs, risk appetite, process ownership, talent, technology, and governance. Effective insurers periodically reassess their value chains, remove unnecessary handoffs, preserve critical internal knowledge, and ensure that accountability remains clear wherever work is performed.

Chapter Summary

An insurance operating model explains how an insurer organizes people, processes, technology, governance, and sourcing to execute strategy. The insurance value chain includes product development, distribution, underwriting, policy administration, premium collection, claims, investment, customer service, and enabling functions. Functional structures support specialization, while process-based structures strengthen end-to-end accountability; product and hybrid structures balance market expertise with shared capabilities.

Centralized models offer consistency and scale, whereas decentralized models provide local responsiveness. Shared services and centres of expertise can improve efficiency when service standards and ownership are clear. Outsourcing and insurtech partnerships may provide capability and innovation, but they do not remove the insurer’s accountability for customers, data, resilience, and regulatory compliance. Life, health, and general insurance require different operating emphases because their contract duration, data, claims, provider, underwriting, and service requirements differ. Sound operating models combine performance measurement, technology, professional judgement, ethical conduct, and strong governance.

Key Terms

Operating model: The design linking strategy with processes, organisation, capabilities, technology, governance, and sourcing.

Value chain: The connected activities through which an insurer creates, delivers, protects, and captures value.

Functional structure: An organisation grouped by specialist functions such as underwriting, claims, finance, and sales.

Process-based structure: An organisation designed around end-to-end customer or business processes.

Centralization: Concentration of authority, resources, systems, or processing in a common centre.

Decentralization: Delegation of authority and activities to local, regional, or product units.

Hybrid operating model: A combination of centralized standards or capabilities and decentralized decision-making.

Shared service: A common internal service that performs activities for multiple business units.

Centre of expertise: A specialist unit that concentrates scarce technical knowledge and sets standards.

Outsourcing: Use of an external provider to perform an activity on behalf of the insurer.

Insurtech: Technology-enabled innovation applied to insurance products, processes, distribution, or risk management.

Straight-through processing: Automated processing with minimal manual intervention from initiation to completion.

Underwriting: Assessment and selection of risks, including acceptance terms and pricing.

Policy administration: Management of policy records, changes, billing, documents, renewals, and status.

Claims adjudication: Evaluation of coverage, liability, evidence, and settlement entitlement.

Operational resilience: The capability to continue important services during disruption and recover effectively.

Process owner: The person accountable for the performance of an end-to-end process.

Service-level agreement: A documented commitment defining a provider’s service scope, quality, timing, and controls.

Model governance: Oversight of the development, validation, use, monitoring, and retirement of analytical models.

Value leakage: Loss of economic or customer value through errors, fraud, delay, rework, or ineffective controls.

Review Questions

Short Answer Questions

1. What is meant by an insurance operating model?

2. Identify four primary activities in the insurance value chain.

3. State two benefits and two risks of centralization.

4. What is the difference between a shared service and a centre of expertise?

5. Why does outsourcing not transfer ultimate accountability from an insurer?

Long Answer Questions

1. Explain the insurance value chain and show how its activities are interdependent.

2. Compare functional, process-based, product, and hybrid organisational structures.

3. Discuss the factors that should guide the choice between centralized and decentralized operations.

4. Analyse the operating-model differences among life, health, and general insurance.

5. Explain how governance, technology, and performance measurement should be integrated in an insurance operating model.

Analytical / Application Questions

1. A national insurer has inconsistent underwriting decisions across branches and rising operating costs. Recommend which activities should be centralized and which should remain local, giving reasons.

2. An insurer plans to use an artificial-intelligence tool to reject low-value health claims automatically. Identify the operational, ethical, and governance controls that should be required before implementation.

3. Using the Meridian General Insurance case, design three performance indicators that would reveal whether the new claims operating model is improving customer and business outcomes.

22. Multiple Choice Questions with Answers

Question 1

Which statement best describes an insurance operating model?

A. A schedule of premium rates approved by the actuary

B. The design linking strategy, processes, people, technology, governance, and sourcing

C. A legal document defining exclusions in an insurance contract

D. A list of all insurance products sold by a company

Correct Answer: B. The design linking strategy, processes, people, technology, governance, and sourcing

Question 2

Which activity most directly converts a customer risk into an insurable product offering?

A. Product development

B. Premium reconciliation

C. Claims recovery

D. Investment settlement

Correct Answer: A. Product development

Question 3

A process-based structure is particularly valuable because it:

A. Eliminates the need for specialist expertise

B. Ensures every decision is made locally

C. Creates accountability across an end-to-end customer journey

D. Prevents the use of shared technology platforms

Correct Answer: C. Creates accountability across an end-to-end customer journey

Question 4

Which activity is generally most suitable for centralization when rules and requirements are common across business units?

A. Local relationship management

B. Jurisdiction-specific commercial negotiation

C. Common cybersecurity monitoring

D. Physical catastrophe inspection

Correct Answer: C. Common cybersecurity monitoring

Question 5

What is a principal risk of decentralizing underwriting authority without adequate controls?

A. Excessive consistency across branches

B. Inconsistent risk selection and pricing

C. Loss of all local market knowledge

D. Automatic elimination of operating expenses

Correct Answer: B. Inconsistent risk selection and pricing

Question 6

Which statement about outsourcing is correct?

A. Outsourcing removes the insurer’s regulatory accountability

B. Outsourcing is justified whenever the vendor offers the lowest price

C. The insurer must retain oversight, knowledge, and effective vendor controls

D. Outsourced activities no longer require business-continuity planning

Correct Answer: C. The insurer must retain oversight, knowledge, and effective vendor controls

Question 7

Why are long-term data and system capabilities especially important in life insurance?

A. Life policies never require customer servicing after issue

B. Life contracts and policy records may remain active for decades

C. Life insurance does not use actuarial information

D. Life claims are always settled automatically

Correct Answer: B. Life contracts and policy records may remain active for decades

Question 8

Which feature is especially significant in health-insurance operations?

A. Provider-network management and protection of medical information

B. Elimination of claims adjudication

C. Exclusive reliance on physical property inspections

D. Absence of customer eligibility controls

Correct Answer: A. Provider-network management and protection of medical information

Question 9

In the Meridian case, why was the artificial-intelligence tool initially used for triage rather than unrestricted claims rejection?

A. Triage requires no data or validation

B. The insurer wanted to retain human review and manage model risk

C. The vendor prohibited automated processing

D. Claims decisions do not affect customer outcomes

Correct Answer: B. The insurer wanted to retain human review and manage model risk

Question 10

Which metric combination best evaluates whether a claims redesign is genuinely successful?

A. Only the number of claims closed per employee

B. Only the reduction in claims payments

C. Cycle time, settlement accuracy, complaints, reopen rates, and relevant cost measures

D. Only the percentage of claims processed automatically

Correct Answer: C. Cycle time, settlement accuracy, complaints, reopen rates, and relevant cost measures