13. Marketing, Distribution and CRM Analytics

Introduction and Context

Insurance marketing and distribution generate a large volume of information: enquiries, quotations, proposals, underwriting decisions, policy issuances, renewals, cancellations, claims, complaints, service interactions and digital engagements. The strategic value of this information does not lie merely in storing it. It lies in converting it into evidence for better decisions about customers, channels, products and service processes. Marketing, distribution and customer relationship management (CRM) analytics provide the methods through which insurers connect commercial activity with customer outcomes and financial performance.

Analytics is particularly important in insurance because the customer relationship is usually longer and more uncertain than in many other industries. A customer may purchase a policy today, pay premiums over several years, make a claim only after a loss, and renew or lapse according to changing needs and perceptions of value. Consequently, a campaign that appears successful at the point of sale may be unprofitable if it attracts customers with low persistency, high servicing costs or poor claims experience. Similarly, a distribution channel that produces fewer policies may create greater long-term value through better retention and lower complaints.

This chapter develops an integrated analytical approach. It examines marketing funnels, acquisition and conversion measures, persistency and lapse, churn and retention, customer lifetime value, share of wallet, channel profitability, satisfaction and complaint indicators. It also considers attribution, dashboards, data governance, ethical use of customer data and emerging applications of automation and artificial intelligence.

Learning Objectives

After studying this chapter, the student should be able to:

1. Explain the role of analytics in insurance marketing, distribution and CRM.

2. Construct and interpret an insurance marketing funnel and its conversion rates.

3. Calculate acquisition cost, persistency, lapse, retention, churn and customer lifetime value.

4. Compare attribution models and assess channel profitability.

5. Design a management dashboard using financial, behavioural and customer-experience indicators.

6. Analyse customer satisfaction, NPS and complaint data without confusing correlation with causation.

7. Identify data, ethical, regulatory and operational risks in CRM analytics.

8. Apply analytical frameworks to managerial decisions concerning campaigns, channels and customer retention.

From Activity Measurement to Value Measurement

A basic reporting system answers questions such as how many leads were generated, how many policies were issued and how much premium was collected. An analytical system asks more demanding questions: Which customers were acquired? At what cost? Through which channel? What proportion renewed? Which products generated profitable relationships? Did the campaign improve customer value or merely accelerate sales that would have occurred anyway?

This distinction is significant. Activity metrics are useful for operational control, but they do not necessarily indicate business success. A high number of leads may reflect weak targeting. A high conversion rate may be achieved by selling simple products to customers who later cancel. A low complaint count may indicate excellent service, but it may also indicate poor complaint accessibility or under-recording. Sound analytics therefore combines measures of volume, quality, economics, behaviour and experience.

A useful analytical chain is:

Reach and engagement → lead quality → quotation and proposal → issuance → premium and margin → persistency and retention → claims and service experience → advocacy and lifetime value.

Each stage should be linked to customer, product, channel, geography, intermediary and time period. The same customer identifier should ideally connect marketing, policy administration, billing, claims and CRM systems. Without this linkage, the insurer may optimise individual departments while damaging the total relationship.

The Insurance Marketing Funnel

The marketing funnel represents the movement from potential market awareness to profitable and enduring customer relationships. Its precise stages vary by product and channel, but a practical insurance funnel includes:

• target audience or reachable market;
• impressions, responses or website visits;
• leads and enquiries;
• qualified leads;
• quotations or illustrations;
• completed applications and proposals;
• underwriting acceptance;
• policy issuance and first payment;
• activation, onboarding and usage;
• renewal, cross-sell and advocacy.

The funnel should not be interpreted as a simple one-way sequence. Customers may enter through a branch, leave the process, return through a digital channel, contact an intermediary, or purchase several products at different times. In complex insurance, a lead may require advice and underwriting before a decision can be made. Funnel design must therefore reflect the actual customer journey rather than force all interactions into a simplistic digital-commerce model.

Conversion Analysis

A stage conversion rate is calculated as:

Conversion rate = Number progressing to the next stage ÷ Number entering the current stage × 100

For example, if 10,000 campaign responses generate 2,000 qualified leads, the response-to-qualified-lead conversion is 20 per cent. If 500 policies are issued from those qualified leads, qualified-lead-to-issue conversion is 25 per cent. The overall response-to-issue conversion is 5 per cent.

Conversion analysis becomes more useful when segmented. An insurer may discover that digital leads convert quickly but lapse more frequently, while adviser-generated leads convert more slowly but produce higher persistency. Conversion should also be measured by product complexity, customer need, age group, geography, device, intermediary and underwriting outcome. A change in conversion may be caused by pricing, lead quality, sales conduct, operational delays, documentation requirements or changes in underwriting appetite.

Funnel analysis should include time-to-conversion. A channel with a high conversion rate but a long cycle may tie up sales capacity and delay premium collection. Cohort analysis, which follows customers acquired in the same period or campaign, is especially valuable because it reveals later retention and profitability rather than focusing only on initial sales.

Acquisition Economics and Attribution

Customer acquisition cost (CAC) indicates the expenditure required to acquire a new customer or policy. A simple formula is:

CAC = Attributable acquisition and marketing cost ÷ Number of new customers acquired

The numerator may include media expenditure, commissions, lead-purchase costs, campaign technology, sales incentives and relevant staff costs. The organisation must define the boundary consistently. If only advertising expenditure is included in one channel and total sales cost in another, comparisons become misleading.

CAC should be distinguished from cost per lead and cost per issued policy. Cost per lead measures efficiency before quality is known; CAC measures the cost of actual acquisition. A further refinement is quality-adjusted CAC, which divides acquisition cost by customers who remain active after a specified period or by customers reaching a defined contribution margin.

Attribution assigns credit for an outcome, such as a sale or renewal, to marketing and distribution touchpoints. Common models include:

• First-touch attribution: all credit is assigned to the first recorded interaction. It is useful for assessing demand creation but ignores later influence.
• Last-touch attribution: all credit goes to the final interaction before purchase. It is simple but may overvalue the closing channel.
• Linear attribution: credit is divided equally among recorded touchpoints.
• Time-decay attribution: interactions closer to conversion receive greater credit.
• Position-based attribution: greater weight is assigned to the first and last interactions, with the remainder distributed among intermediate contacts.
• Data-driven attribution: statistical or machine-learning methods estimate the incremental contribution of touchpoints.

No attribution model automatically proves causation. A customer who clicks a search advertisement after deciding to buy may have been close to purchase already. A call-centre interaction may appear influential because high-intent customers choose to call. Incrementality testing, holdout groups and controlled experiments provide stronger evidence than observational attribution alone. Privacy restrictions, incomplete identifiers and offline interactions also limit measurement accuracy.

Retention, Persistency, Lapse and Churn

Retention is the proportion of customers or policies that remain active during a specified period. Churn is the proportion that is lost. For a defined customer cohort:

Retention rate = Customers active at end of period ÷ Customers active at beginning of period × 100

Churn rate = Customers lost during period ÷ Customers active at beginning of period × 100

The two measures should specify whether deaths, maturity, switching, non-payment, cancellation, replacement and migration are included. In insurance, policy persistency is often more meaningful than a broad customer-retention measure. Persistency measures the proportion of policies that remain in force at a specified duration, such as 13, 25 or 37 months, depending on product and reporting practice.

Lapse is termination caused by failure to pay premiums or maintain the policy according to its terms. Voluntary cancellation, surrender, replacement and non-renewal may require separate categories. Combining all exits into one churn figure hides different managerial problems. A lapse following a billing failure calls for payment reminders or alternative payment methods; cancellation after a poor claims experience calls for service improvement; replacement may indicate unsuitable advice or aggressive competitor activity.

Retention analytics should use cohorts. A cohort of policies issued in January can be tracked through monthly or annual survival. A survival curve shows the proportion remaining active over time, while hazard analysis estimates the likelihood of exit at each period. Predictive lapse models can identify customers at risk, but intervention must be appropriate and not discriminatory. Retention efforts should not obstruct a customer’s legitimate decision to cancel or create pressure to retain an unsuitable policy.

Customer Lifetime Value and Share of Wallet

Customer lifetime value (CLV or LTV) estimates the economic value of a customer over the expected relationship. A simplified model is:

LTV = Present value of expected contribution margins – acquisition cost – servicing and retention costs

A more formal discounted model is:

LTV = Σ [(Expected revenue_t – Expected variable cost_t) × Probability of being active_t] ÷ (1 + discount rate)^t – acquisition cost

Insurance revenue may include premium-related margin, fees and commission income, while costs may include claims, reinsurance effects, servicing, payment processing, commissions and capital-related costs. The model must avoid treating gross premium as profit. Claims uncertainty, fraud, catastrophe exposure and regulatory capital requirements make insurance LTV more complex than subscription-business LTV.

LTV is an estimate, not an accounting fact. It depends on assumptions about renewal, premium changes, claims, expenses, discount rates and customer behaviour. Scenario analysis should test optimistic, central and adverse assumptions. LTV should also be calculated at segment or cohort level rather than used to justify unequal treatment of individuals without proper governance.

Share of wallet refers to the proportion of a customer’s relevant insurance expenditure held by the insurer. It is often difficult to observe directly because customers may not disclose all policies. It can therefore be estimated through customer research, product-holding patterns, external market data or propensity models. Cross-sell potential should be based on genuine protection needs and affordability, not simply on the number of products that can be sold.

Channel Profitability and Distribution Analytics

Distribution profitability requires a full view of channel economics. Relevant channels may include tied agents, brokers, bancassurance, branches, direct sales, comparison platforms, digital aggregators, corporate partnerships and affinity groups. Each channel has a different cost structure, control environment, advice capability and customer profile.

A channel contribution statement may include:

Gross contribution = Premium-related contribution and fee income – claims-related variable costs – commissions – acquisition costs – servicing costs – allocated operating costs.

The allocation of shared costs must be transparent. Allocating all head-office costs equally may make a high-volume channel appear unprofitable, while allocating costs according to activity or resource consumption may provide a more useful managerial view. Profitability should be examined over the customer life cycle. A channel producing high first-year commission income may become less attractive after early lapse and elevated complaint costs are included.

Channel comparison should consider more than financial contribution. Important dimensions include sales quality, suitability, persistency, claims assistance, complaints, regulatory breaches, speed of service, customer accessibility and strategic reach. A balanced channel scorecard prevents management from selecting channels solely on short-term sales.

Satisfaction, NPS and Complaint Indicators

Customer satisfaction is usually measured through a rating of a specific interaction or overall relationship. Net Promoter Score (NPS) is commonly derived by asking customers how likely they are to recommend the organisation on a 0–10 scale. Scores of 9–10 are classified as promoters, 7–8 as passives and 0–6 as detractors. NPS is calculated as:

NPS = Percentage of promoters – Percentage of detractors.

NPS is an indicator of sentiment and advocacy, not a complete measure of service quality or profitability. Results depend on sampling, response rates, question wording, cultural context and timing. Transactional NPS after a claim may differ substantially from relationship NPS collected periodically. A responsible dashboard reports the sample size, response rate, segment, survey channel and confidence limitations.

Complaint analytics classifies complaints by product, cause, channel, severity, outcome, turnaround time, repeat occurrence and regulatory significance. A complaint ratio may be expressed per 1,000 policies, per 10,000 customers or per million transactions. Raw complaint counts favour large portfolios and can conceal deterioration in a small segment. The analysis should also track reopened complaints, escalations, ombudsman or regulator referrals, compensation, root cause and remediation completion.

Customer-experience indicators should be connected to operational and financial outcomes. A long claims settlement time may reduce satisfaction and increase churn. A high first-contact resolution rate may reduce repeat contacts and cost. Correlation is not proof of causation, however. Customers with complex claims may be less satisfied because of the claim itself, even when service quality is good. Qualitative review and root-cause investigation are necessary complements to statistical analysis.

Designing an Executive Analytics Dashboard

An effective dashboard is designed around decisions rather than the availability of data. A senior distribution dashboard might contain five linked perspectives:

1. Growth: reach, qualified leads, quotations, policies issued, premium and conversion.

2. Economics: CAC, commission, contribution margin, LTV, payback period and channel profitability.

3. Relationship health: persistency, retention, lapse, churn, cross-sell and share of wallet.

4. Experience and conduct: satisfaction, NPS, complaints, claims turnaround, remediation and sales-quality indicators.

5. Risk and data quality: unusual cancellations, vulnerable-customer outcomes, consent status, missing identifiers, model drift and access exceptions.

Each metric requires a definition, owner, source system, calculation logic, refresh frequency, target, tolerance and escalation rule. A dashboard should distinguish leading indicators, such as web engagement and pending complaints, from lagging indicators, such as annual profitability and persistency. Traffic-light colours can support rapid attention, but they should not replace interpretation.

Drill-down capability is essential. Management should be able to move from total portfolio results to product, cohort, branch, agent, region, campaign and customer segment. Data lineage ensures that users can identify how a number was produced. Reconciliation with policy, finance and claims systems protects trust in the dashboard.

An Integrated Analytical Process

A robust analytics process usually follows six stages. First, define the managerial decision: for example, whether to expand a bancassurance campaign. Second, define the unit of analysis, such as lead, policy, customer, household, claim or channel. Third, establish data definitions and quality rules. Fourth, analyse descriptive and diagnostic patterns. Fifth, estimate future outcomes using forecasting or predictive models. Sixth, test and implement an intervention, then measure its effect.

Descriptive analytics explains what happened. Diagnostic analytics examines why it happened. Predictive analytics estimates what may happen. Prescriptive analytics recommends an action subject to constraints. These categories are complementary. A predictive lapse score is not useful unless the insurer knows which intervention is affordable, lawful and likely to help the customer.

Segmentation may use demographics, needs, behaviour, value, life stage or service preference. Care is needed when using sensitive or proxy variables. A model can unintentionally disadvantage protected or vulnerable groups even when sensitive attributes are excluded. Validation should examine accuracy, stability, calibration and fairness across relevant segments.

Illustrative Case: Meridian Assurance’s Retention and Channel Review

Background

Meridian Assurance, a fictional multi-line insurer, sold health and protection products through agents, bank branches and a direct digital channel. Management observed that the digital channel had the lowest acquisition cost and therefore planned to transfer a larger marketing budget to it.

Situation and Problem

The digital channel generated many quotations and policies. However, preliminary reports showed weaker 13-month persistency and a higher volume of payment-related complaints. Agent sales had a higher CAC but stronger renewal performance. The existing dashboard measured issued policies and first-year premium but did not connect acquisition data with later policy behaviour.

Analysis

The analytics team created cohorts by channel and month of issue. It linked campaign costs, policy issuance, payment history, service contacts, complaints and renewal status. The analysis showed that digital CAC was 30 per cent lower than agent CAC, but digital first-year lapse was materially higher. When servicing cost, refunds, complaint handling and expected second-year contribution were included, the difference in channel profitability was much smaller than the initial report suggested.

Further investigation found that many digital customers misunderstood waiting periods and payment dates. The quotation journey used concise product language, but the pre-issuance explanation did not adequately highlight these conditions. Customers who received a post-sale call and a simplified payment reminder had better early persistency than comparable customers who did not receive the intervention.

The team also tested attribution. Last-touch reporting credited the digital payment page with almost every sale, while a controlled holdout indicated that some conversions would have occurred without the final reminder. This prevented management from overestimating the incremental effect of the reminder campaign.

Approach and Solution

Meridian adopted a balanced channel scorecard including quality-adjusted CAC, 13-month persistency, complaint rate, expected LTV, claims and service indicators, and conduct measures. It redesigned the digital explanation of key terms, introduced consent-based reminders, offered payment-date choices and created an early-life service contact for selected customers. The agent channel was not reduced immediately; instead, it was used for products requiring advice or more complex needs.

Outcome

In this illustrative case, management concluded that the correct decision was not to select one universally superior channel. Digital distribution was retained for suitable, lower-complexity products and improved through clearer communication. Adviser channels remained important where explanation and needs assessment created value. The case demonstrates how a short-term acquisition metric can lead to a poor decision when persistency, complaints, attribution and lifetime economics are ignored.

Learning Points

The principal lessons are that acquisition cost must be quality-adjusted, channel profitability must be measured over time, attribution does not equal causation, and customer-experience data can identify both commercial and conduct risks. A retention intervention should be tested for genuine incremental impact rather than assumed to work.

Risks, Limitations and Common Analytical Errors

Several errors recur in insurance analytics. The first is inconsistent definitions: one department may count a policy as retained when it is merely active for a few days, while another uses renewal payment. The second is survivorship bias, where only continuing customers are analysed. The third is confusing correlation with causation. The fourth is using averages that conceal segment differences. The fifth is treating predicted LTV as certain value.

Other risks include duplicate customer records, missing intermediary identifiers, disconnected offline and online journeys, delayed claims data, incorrect cost allocation and data leakage in predictive modelling. A model trained using information that became available only after the decision date will appear accurate but cannot support real-time decisions. Model drift can occur when pricing, regulation, customer behaviour or distribution strategy changes.

Dashboards can also create behavioural risks. Sales teams may pursue volume targets by encouraging unsuitable sales. Call-centre staff may reduce recorded complaints to meet performance targets. Managers may focus on metrics that are easy to improve rather than those that matter to customers. Balanced scorecards, audit trails, independent review and outcome-based incentives reduce these risks.

Governance, Ethics and Regulation

CRM analytics must respect lawful processing, purpose limitation, data minimisation, accuracy, security, retention controls and customer rights under applicable data-protection and insurance regulation. Consent and legitimate business purposes should not be treated as interchangeable in every context. Customers should receive understandable information about relevant data use, especially profiling, automated decisions and targeted marketing.

Fairness is central to insurance. Analytics may support risk classification, but it should not produce unjustified discrimination or exploit vulnerability. Price, retention and marketing models require governance for prohibited variables, proxy discrimination, explainability and adverse outcomes. Customers should have access to appropriate human review where automated decisions significantly affect them.

A governance framework should assign accountability to business owners, data stewards, compliance and risk functions, model validators, information-security teams and senior management. Important controls include role-based access, encryption, audit logs, data-quality monitoring, model documentation, bias testing, incident response and periodic review of vendor data. Ethical analytics asks not only whether a prediction is accurate, but whether the proposed action is fair, proportionate and in the customer’s legitimate interest.

Technology, Automation and Emerging Practice

Modern insurers increasingly use customer data platforms, application programming interfaces, cloud data warehouses, visualisation tools and event-based CRM systems. These technologies can unify online behaviour, policy records, claims events and service interactions. Automation may trigger a welcome message, payment reminder, renewal communication or complaint escalation. Such automation should include consent controls, suppression rules, human oversight and testing for message appropriateness.

Machine learning can improve propensity, lapse and next-best-action models, but complex models may be difficult to explain. Generative AI can assist with dashboard commentary, customer-service summaries and data exploration, yet it introduces risks of inaccurate outputs, confidential-data exposure and uncontrolled recommendations. Human validation, restricted data environments and clear accountability remain necessary.

Emerging practice is moving from campaign reporting toward continuous decision systems that evaluate customer value, experience and conduct together. Experimentation platforms, causal inference, privacy-enhancing technologies and real-time journey analytics are becoming more important. The strategic direction is not simply greater automation; it is more disciplined use of evidence in which commercial performance and customer outcomes are assessed jointly.

Best Practices for Managers and Analysts

A mature analytics function should:

• begin with a decision and a clearly defined customer or policy unit;
• maintain a common metric dictionary and reliable customer identifiers;
• report acquisition, persistency, profitability and experience together;
• use cohort analysis and appropriate time horizons;
• distinguish attribution from incremental causal impact;
• test interventions through controlled experiments where feasible;
• disclose assumptions, uncertainty and data limitations;
• monitor model fairness, drift and unintended consequences;
• protect privacy and restrict access to necessary data;
• connect dashboards to accountable owners and actions; and
• review whether metrics are improving genuine customer and organisational outcomes.

Chapter Summary

Marketing, distribution and CRM analytics convert insurance data into decisions about growth, value, retention and customer treatment. The marketing funnel identifies movement from awareness to policy issuance and subsequent relationship stages, while conversion analysis reveals where customers are gained or lost. CAC becomes more meaningful when adjusted for policy quality and later persistency. Attribution models organise credit across touchpoints, but controlled testing is required to estimate incremental impact.

Retention, persistency, lapse and churn must be defined precisely and analysed through cohorts. LTV combines expected contribution, retention probability, costs and discounting, while share of wallet indicates broader relationship potential. Channel profitability should include acquisition, commission, servicing, claims-related economics and long-term customer value. Satisfaction, NPS and complaint indicators provide important experience and conduct evidence but require careful sampling and root-cause interpretation.

An effective dashboard links growth, economics, relationship health, experience and risk. Data governance, privacy, fairness, explainability and human oversight are essential, particularly when automation and AI are used. The central managerial principle is to optimise durable, fair and profitable relationships rather than isolated sales metrics.

Key Terms

Acquisition Cost: Expenditure attributable to obtaining a new customer or policy.

Attribution: Assignment of credit for a conversion or outcome to marketing or distribution touchpoints.

Churn: Loss of customers or policies during a specified period.

Cohort Analysis: Study of customers grouped by a common characteristic, usually acquisition or issue period.

Conversion Rate: Proportion of entities progressing from one funnel stage to the next.

Customer Lifetime Value: Present value of expected future contribution from a customer after relevant costs.

Customer Relationship Management Analytics: Analytical use of customer data to improve acquisition, service, retention and value.

Funnel: Staged representation of movement from market contact to purchase and continuing relationship.

Incrementality: The additional outcome caused by an intervention compared with what would have occurred without it.

Lapse: Termination of a policy, commonly because required premium payments are not maintained.

Net Promoter Score: Percentage of promoters minus percentage of detractors from a recommendation question.

Persistency: Proportion of policies remaining in force at a specified duration.

Quality-Adjusted CAC: Acquisition cost evaluated against customers meeting a later quality or retention criterion.

Retention Rate: Proportion of customers or policies remaining active over a defined period.

Share of Wallet: Proportion of a customer’s relevant expenditure captured by the insurer.

Survival Curve: Representation of the proportion of a cohort remaining active over time.

Touchpoint: A customer interaction with an insurer, intermediary, product or campaign.

Channel Profitability: Economic contribution of a distribution channel after relevant income and costs.

Review Questions

Short Answer Questions

1. What is the difference between cost per lead and customer acquisition cost?

2. Define persistency and explain why it is important in insurance.

3. What does NPS measure, and what does it fail to measure directly?

4. Why is cohort analysis useful in retention management?

5. Distinguish lapse from the broader concept of churn.

Long Answer Questions

1. Explain the stages of an insurance marketing funnel and discuss how conversion should be interpreted across segments.

2. Compare first-touch, last-touch, linear, time-decay and data-driven attribution models.

3. Discuss the construction and limitations of a customer lifetime value model for an insurer.

4. Explain how an executive dashboard can integrate growth, economics, customer experience and conduct indicators.

5. Examine the governance and ethical requirements for predictive CRM analytics in insurance.

Analytical / Application Questions

1. An insurer has a low-CAC digital channel with weak 13-month persistency and a high-CAC adviser channel with strong renewals. Design a comparative analysis to determine whether the digital budget should be increased.

2. A campaign receives many leads but produces few issued policies and a high complaint rate. Identify the data you would examine and recommend a diagnostic process.

3. An insurer’s last-touch model credits a payment reminder with nearly all conversions. Explain why this may overstate the reminder’s value and propose a stronger evaluation method.

22. Multiple Choice Questions with Answers

Question 1

Which measure best evaluates whether newly acquired insurance customers remain active after the initial sale?

A. Cost per impression

B. Quality-adjusted acquisition cost

C. Number of website visits

D. First-touch attribution

Correct Answer: B. Quality-adjusted acquisition cost

Question 2

A policy cohort begins with 2,000 policies and has 1,600 active policies at the end of the measurement period. What is the retention rate, assuming no other adjustments are required?

A. 20%

B. 40%

C. 80%

D. 125%

Correct Answer: C. 80%

Question 3

Which attribution approach gives greater weight to interactions occurring closer to conversion?

A. Linear attribution

B. Time-decay attribution

C. First-touch attribution

D. Random attribution

Correct Answer: B. Time-decay attribution

Question 4

Why should gross premium not be used as the sole measure of customer lifetime value?

A. Premium is never recorded in insurance systems

B. Premium excludes all customer information

C. Claims, commissions, servicing and other costs affect contribution

D. Premium cannot be compared across products

Correct Answer: C. Claims, commissions, servicing and other costs affect contribution

Question 5

An insurer records 25% promoters and 15% detractors in an NPS survey. What is the NPS?

A. 10

B. 15

C. 25

D. 40

Correct Answer: A. 10

Question 6

Which situation most clearly demonstrates the difference between attribution and incrementality?

A. A campaign has a defined budget

B. A customer interacts with two channels

C. A final reminder receives credit, although many customers would have purchased without it

D. A policy is issued after underwriting

Correct Answer: C. A final reminder receives credit, although many customers would have purchased without it

Question 7

Which dashboard indicator is most directly associated with a conduct risk?

A. Website impressions

B. Complaint escalation and remediation status

C. Number of campaign colours

D. Average screen-loading time only

Correct Answer: B. Complaint escalation and remediation status

Question 8

What is the main purpose of cohort analysis in lapse management?

A. To eliminate the need for customer identifiers

B. To compare customers grouped by a common acquisition or issue period over time

C. To assign all sales credit to the final channel

D. To calculate impressions without considering policies

Correct Answer: B. To compare customers grouped by a common acquisition or issue period over time

Question 9

A predictive lapse model identifies vulnerable customers. Which action is most consistent with responsible analytics?

A. Automatically deny all future products to those customers

B. Use the model without checking for bias or accuracy

C. Offer an appropriate, consent-based intervention and monitor fairness

D. Delete all customers predicted to lapse

Correct Answer: C. Offer an appropriate, consent-based intervention and monitor fairness

Question 10

Which is the strongest reason for combining channel profitability with persistency and complaint data?

A. It makes all channels appear identical

B. It prevents short-term sales from being mistaken for durable value

C. It removes the need to calculate acquisition cost

D. It guarantees that every customer will renew

Correct Answer: B. It prevents short-term sales from being mistaken for durable value