Most insurers measure NPS and complaints. Neither answers the question the FCA is actually asking. Consumer Duty requires evidence across four outcomes: products and services, price and value, consumer understanding, and consumer support. This post explains the genuine measurement approach for each outcome, the data sources that make it possible, and what a Consumer Duty board report needs to contain to satisfy the regulator.
Four Metrics That Look Good. Zero Evidence of Good Customer Outcomes.
The Consumer Duty board report is due in three weeks. The head of compliance has her four slides ready. NPS: up 4 points year on year. Complaints: down 12%. Renewal rate: 84%, stable. Claims settlement time: 14 days average, within the 21-day target. The board will be satisfied. The metrics are moving in the right direction. But she is not satisfied.
NPS measures whether customers are happy. It does not measure whether they were treated fairly. Complaints measure the customers who escalated. Most customers who received a poor outcome did not complain. Renewal rate measures how many customers came back. It does not measure whether the product they renewed was still appropriate for their needs. Claims settlement time measures how fast the team worked. It does not measure whether the settlement amount was fair.
She has data that shows the business is performing well. She does not have data that shows the business is delivering the customer outcomes the FCA is actually looking for. The FCA did not ask for four metrics that look good. It asked for evidence that customers are receiving fair value, appropriate products, clear information, and effective support. Those are four different questions. Her four slides answer none of them directly.
The Four Consumer Duty Outcomes: What They Require and How to Measure Them
FCA Consumer Duty insurance obligations centre on four outcomes. Each requires a specific measurement approach. The common gap is that most insurers have data that measures activity and satisfaction, not whether customers actually received what the regulation requires.
| Outcome | What the FCA requires | Common measurement gap | Genuine measurement approach |
|---|---|---|---|
| Products and services | Products must be designed to meet the needs of the target market. Cover must be appropriate for the customers buying it. | Renewal rate measures retention, not whether renewing customers still need what they are buying. | Claims acceptance rate by segment. Coverage gap analysis. Target market vs actual purchaser comparison. |
| Price and value | Price must be reasonable relative to the benefits delivered. Customers must receive fair value. | Premium comparison tracks market position, not whether customers received the value they paid for. | Claims payout ratio by segment. Benefit utilisation rate. Value delivered per NOK of premium paid. |
| Consumer understanding | Customers must understand what they are buying, what is excluded, and how to make a claim. | Complaints about unclear terms measure only the customers who escalated — most confused customers do not complain. | Post-purchase comprehension testing. Claim denial rates attributable to misunderstood exclusions. Contact centre query categorisation. |
| Consumer support | Customers must be able to access support when they need it. Support must be accessible and effective. | Customer satisfaction scores measure experience, not whether support was available when needed. | Abandonment rate at key contact points. First-contact resolution rate. Vulnerable customer support outcomes tracked separately. |
The price and value outcome is the most consistently underdeveloped in insurance. An insurer that charges NOK 4,200 for a home insurance policy and pays out NOK 980 in claims on average has a claims payout ratio of 23%. A payout ratio of 23% overall may mask a 9% payout ratio for customers in a specific age band or geographic area. That is an outcome failure even if the aggregate figure looks acceptable. The FCA expects segmented analysis. Averages conceal segment-level failures.
FCA: Consumer Duty Insurance Sector Review and Supervisory Findings · 2024[1]How Data and AI Enable Genuine Insurance Customer Outcome Measurement
Insurance customer outcome measurement requires data from across the customer lifecycle. A single data source cannot answer the Consumer Duty questions. The table below maps the key data sources to the outcomes they measure and the AI contribution at each point.
| Data source | What it measures | Consumer Duty outcome | AI contribution |
|---|---|---|---|
| Claims outcome data | Acceptance rate, denial rate, denial reason, time to settlement, settlement value vs estimate | Products and services; price and value | Pattern detection on denial reasons by segment; identification of systematic coverage gaps |
| Renewal behaviour | Renewal rate by segment, product age, claim history, coverage level change at renewal | Products and services | Churn prediction; identification of customers likely renewing unsuitable cover |
| Contact centre query data | Query type, resolution rate, first-contact resolution, escalation rate, abandon rate | Consumer understanding; consumer support | NLP classification of query types; identification of comprehension failure patterns |
| Complaints root cause | Complaint category, root cause, time to resolution, vulnerability flag, FOS referral rate | All four outcomes | Root cause pattern detection; Consumer Duty board report data assembly |
| Vulnerable customer outcomes | Identification rate, routing accuracy, resolution rate, support protocol adherence | Consumer support | Vulnerability indicator detection; outcome monitoring by vulnerability type |
| Product usage data | Feature utilisation, coverage usage rate, endorsement uptake, mid-term policy changes | Products and services; price and value | Identification of consistently unused coverage that may indicate product mismatch |
AI pattern detection is particularly important for the consumer understanding outcome. A human analyst reviewing complaint data can identify individual cases of confusion. An AI model processing all contact centre queries, complaints, and post-denial feedback simultaneously identifies the systematic patterns: the specific exclusion clause that generates disproportionate claim denials, the product document section that customers consistently misunderstand, the communication channel where comprehension is lowest. These patterns are invisible in aggregate NPS data. They are visible in claims denial reason data, contact centre query classification, and post-denial customer feedback when processed at scale.
What a Genuinely Useful Consumer Duty Board Report Contains
A Consumer Duty board report that will satisfy the FCA if it asks for the evidence base contains five things. Each is distinct from a standard metrics pack.
For each of the four Consumer Duty outcomes, the report presents the specific data that evidences whether the outcome is being delivered. Claims acceptance rate by segment for products and services. Benefit utilisation rate for price and value. Comprehension failure indicators for consumer understanding. Vulnerable customer support outcomes for consumer support.
Board-level averages conceal segment-level failures. The report shows the distribution of outcomes across customer segments, product types, and distribution channels. It identifies the segments where outcomes are weakest. An aggregate that looks acceptable may mask a specific population receiving consistently poor outcomes.
Where outcomes are not being delivered, the report identifies why. A high claim denial rate attributable to a specific exclusion is a different problem from a high claim denial rate attributable to fraud indicators. Each requires a different response. The FCA expects root cause analysis, not just the headline number.
The FCA expects to see that the board reviewed the evidence, identified failures, took action, and tracked whether the action improved outcomes. A board report without an action log and an outcome tracking section is a metrics pack, not a Consumer Duty governance document.
FCA Consumer Duty insurance requirements specifically expect vulnerable customer outcomes to be tracked and reported separately. They must not be averaged into the general population metrics. Where vulnerable customer outcomes diverge materially from the general population, the divergence requires root cause analysis and a remediation action in the report.
The Finanstilsynet Equivalent for Norwegian Insurers
Norwegian insurers do not operate under the FCA Consumer Duty framework. They operate under Finanstilsynet's tilsynspraksis and the Forsikringsavtaleloven. The Forsikringsavtaleloven contains substantive customer protection requirements: clear pre-contract information, comprehensible policy terms, fair and prompt claims handling, and the customer's right to complain to Finansklagenemnda.
Finanstilsynet's supervisory practice expects insurers to demonstrate that their products are appropriate for their customers and that claims are handled fairly. The evidence base expected is similar in character to Consumer Duty outcome evidence — it differs in specificity and formal structure, not in kind. The measurement infrastructure that generates Consumer Duty evidence also generates the evidence base that Finanstilsynet expects. The data sources are the same. The AI analysis is the same. Only the reporting format and the regulatory submission differ. Specific Norwegian regulatory requirements should be verified with qualified Norwegian legal counsel.
Frequently Asked Questions
We already measure NPS and complaints — is that not sufficient for Consumer Duty?+
No. NPS measures satisfaction. Consumer Duty measures outcomes. The FCA has been explicit: firms that submit NPS and complaints volumes as their primary Consumer Duty evidence are not meeting the standard. NPS does not measure whether customers received fair value. Complaints do not measure customers who experienced a poor outcome but did not escalate. The FCA expects outcome-specific evidence for each of the four Consumer Duty outcomes, segmented by customer group, with root cause analysis where outcomes are failing, and an action log showing what was done about it.[1]
How do we identify customers who received a poor outcome but did not complain?+
Post-claims outcome tracking is the primary method. Customers whose claims were denied are a defined population. Contacting a sample to understand whether the denial was understood and accepted, or experienced as unfair, reveals the silent outcome failures that complaints data misses. Contact centre call classification is a second method: calls about claim denials, coverage queries, and cancellations contain outcome signals that NLP analysis can identify at scale. AI models trained on this data identify the systematic patterns that individual case review cannot see.[2]
What does the FCA expect to see in a Consumer Duty board report?+
The FCA expects four things. Evidence for each of the four outcomes, not just aggregate satisfaction metrics. Segmented analysis that shows outcomes by customer group rather than only overall averages. Root cause analysis where outcomes are poor, explaining why they are failing and what the insurer is doing about it. An action log demonstrating that the board reviewed the evidence, identified failures, took action, and tracked the result. A report that presents improving metrics without identifying any failures will attract FCA scrutiny rather than satisfy it.[1]
How do we measure the price and value outcome for a product with low claims frequency?+
Low-claims products — travel insurance, legal expenses, breakdown cover — present a genuine measurement challenge. Claim frequency is low, so claims payout ratio is low, but that does not mean value was not delivered. The measurement approach for low-frequency products focuses on: the breadth of coverage relative to the premium charged; the proportion of in-scope events that resulted in a successful claim; customer understanding of what triggers a valid claim; and the availability and quality of non-claims benefits. The FCA accepts that value measurement for low-frequency products requires qualitative evidence alongside quantitative data.[1]
How do we build a vulnerable customer outcome tracking capability?+
Start with identification. Your complaints AI system flags vulnerability indicators in complaint text. Your contact centre system flags calls where vulnerability protocols were applied. Your claims system should flag claims associated with customers on the vulnerability register. These three sources define the vulnerable customer population for outcome tracking purposes. For each group, track the same outcome metrics as the general population: claims acceptance rate, resolution time, first-contact resolution, complaint rate. Report these separately in the Consumer Duty board report. Where vulnerable customer outcomes diverge materially from the general population, that divergence requires root cause analysis and a remediation action.[3]
What is the FCA's enforcement approach to Consumer Duty failures in insurance?+
The FCA's 2024 Consumer Duty insurance reviews resulted in supervisory letters to a number of insurers requiring evidence of outcome monitoring improvement within defined timescales. Formal enforcement actions for Consumer Duty failures have focused on price and value failures — specifically, products where the claims payout ratio indicated systematic poor value for specific customer segments. The average cost of a Consumer Duty-related supervisory action in the insurance sector in 2024 was NOK 2.4 million, including remediation, management time, and the cost of the required consumer redress programme.[1]
This article provides general information only and does not constitute legal or regulatory advice. FCA Consumer Duty obligations, FCA PRIN 2A requirements, Finanstilsynet expectations under the Forsikringsavtaleloven, and vulnerable customer treatment obligations require case-specific legal assessment. UK and Norwegian insurers should consult qualified legal and compliance counsel for guidance specific to their jurisdiction and outcome monitoring programme.
References
All statistics sourced from documented deployments and third-party research organisations. All currency figures in NOK. Links verified 2026. Click any citation to jump to its source.
Most insurers measure NPS and complaints. Neither answers the question the FCA is actually asking. Consumer Duty requires evidence across four outcomes: products and services, price and value, consumer understanding, and consumer support. This post explains the genuine measurement approach for each outcome, the data sources that make it possible, and what a Consumer Duty board report needs to contain to satisfy the regulator.
Four Metrics That Look Good. Zero Evidence of Good Customer Outcomes.
The Consumer Duty board report is due in three weeks. The head of compliance has her four slides ready. NPS: up 4 points year on year. Complaints: down 12%. Renewal rate: 84%, stable. Claims settlement time: 14 days average, within the 21-day target. The board will be satisfied. The metrics are moving in the right direction. But she is not satisfied.
NPS measures whether customers are happy. It does not measure whether they were treated fairly. Complaints measure the customers who escalated. Most customers who received a poor outcome did not complain. Renewal rate measures how many customers came back. It does not measure whether the product they renewed was still appropriate for their needs. Claims settlement time measures how fast the team worked. It does not measure whether the settlement amount was fair.
She has data that shows the business is performing well. She does not have data that shows the business is delivering the customer outcomes the FCA is actually looking for. The FCA did not ask for four metrics that look good. It asked for evidence that customers are receiving fair value, appropriate products, clear information, and effective support. Those are four different questions. Her four slides answer none of them directly.
The Four Consumer Duty Outcomes: What They Require and How to Measure Them
FCA Consumer Duty insurance obligations centre on four outcomes. Each requires a specific measurement approach. The common gap is that most insurers have data that measures activity and satisfaction, not whether customers actually received what the regulation requires.
| Outcome | What the FCA requires | Common measurement gap | Genuine measurement approach |
|---|---|---|---|
| Products and services | Products must be designed to meet the needs of the target market. Cover must be appropriate for the customers buying it. | Renewal rate measures retention, not whether renewing customers still need what they are buying. | Claims acceptance rate by segment. Coverage gap analysis. Target market vs actual purchaser comparison. |
| Price and value | Price must be reasonable relative to the benefits delivered. Customers must receive fair value. | Premium comparison tracks market position, not whether customers received the value they paid for. | Claims payout ratio by segment. Benefit utilisation rate. Value delivered per NOK of premium paid. |
| Consumer understanding | Customers must understand what they are buying, what is excluded, and how to make a claim. | Complaints about unclear terms measure only the customers who escalated — most confused customers do not complain. | Post-purchase comprehension testing. Claim denial rates attributable to misunderstood exclusions. Contact centre query categorisation. |
| Consumer support | Customers must be able to access support when they need it. Support must be accessible and effective. | Customer satisfaction scores measure experience, not whether support was available when needed. | Abandonment rate at key contact points. First-contact resolution rate. Vulnerable customer support outcomes tracked separately. |
The price and value outcome is the most consistently underdeveloped in insurance. An insurer that charges NOK 4,200 for a home insurance policy and pays out NOK 980 in claims on average has a claims payout ratio of 23%. A payout ratio of 23% overall may mask a 9% payout ratio for customers in a specific age band or geographic area. That is an outcome failure even if the aggregate figure looks acceptable. The FCA expects segmented analysis. Averages conceal segment-level failures.
FCA: Consumer Duty Insurance Sector Review and Supervisory Findings · 2024[1]How Data and AI Enable Genuine Insurance Customer Outcome Measurement
Insurance customer outcome measurement requires data from across the customer lifecycle. A single data source cannot answer the Consumer Duty questions. The table below maps the key data sources to the outcomes they measure and the AI contribution at each point.
| Data source | What it measures | Consumer Duty outcome | AI contribution |
|---|---|---|---|
| Claims outcome data | Acceptance rate, denial rate, denial reason, time to settlement, settlement value vs estimate | Products and services; price and value | Pattern detection on denial reasons by segment; identification of systematic coverage gaps |
| Renewal behaviour | Renewal rate by segment, product age, claim history, coverage level change at renewal | Products and services | Churn prediction; identification of customers likely renewing unsuitable cover |
| Contact centre query data | Query type, resolution rate, first-contact resolution, escalation rate, abandon rate | Consumer understanding; consumer support | NLP classification of query types; identification of comprehension failure patterns |
| Complaints root cause | Complaint category, root cause, time to resolution, vulnerability flag, FOS referral rate | All four outcomes | Root cause pattern detection; Consumer Duty board report data assembly |
| Vulnerable customer outcomes | Identification rate, routing accuracy, resolution rate, support protocol adherence | Consumer support | Vulnerability indicator detection; outcome monitoring by vulnerability type |
| Product usage data | Feature utilisation, coverage usage rate, endorsement uptake, mid-term policy changes | Products and services; price and value | Identification of consistently unused coverage that may indicate product mismatch |
AI pattern detection is particularly important for the consumer understanding outcome. A human analyst reviewing complaint data can identify individual cases of confusion. An AI model processing all contact centre queries, complaints, and post-denial feedback simultaneously identifies the systematic patterns: the specific exclusion clause that generates disproportionate claim denials, the product document section that customers consistently misunderstand, the communication channel where comprehension is lowest. These patterns are invisible in aggregate NPS data. They are visible in claims denial reason data, contact centre query classification, and post-denial customer feedback when processed at scale.
What a Genuinely Useful Consumer Duty Board Report Contains
A Consumer Duty board report that will satisfy the FCA if it asks for the evidence base contains five things. Each is distinct from a standard metrics pack.
For each of the four Consumer Duty outcomes, the report presents the specific data that evidences whether the outcome is being delivered. Claims acceptance rate by segment for products and services. Benefit utilisation rate for price and value. Comprehension failure indicators for consumer understanding. Vulnerable customer support outcomes for consumer support.
Board-level averages conceal segment-level failures. The report shows the distribution of outcomes across customer segments, product types, and distribution channels. It identifies the segments where outcomes are weakest. An aggregate that looks acceptable may mask a specific population receiving consistently poor outcomes.
Where outcomes are not being delivered, the report identifies why. A high claim denial rate attributable to a specific exclusion is a different problem from a high claim denial rate attributable to fraud indicators. Each requires a different response. The FCA expects root cause analysis, not just the headline number.
The FCA expects to see that the board reviewed the evidence, identified failures, took action, and tracked whether the action improved outcomes. A board report without an action log and an outcome tracking section is a metrics pack, not a Consumer Duty governance document.
FCA Consumer Duty insurance requirements specifically expect vulnerable customer outcomes to be tracked and reported separately. They must not be averaged into the general population metrics. Where vulnerable customer outcomes diverge materially from the general population, the divergence requires root cause analysis and a remediation action in the report.
The Finanstilsynet Equivalent for Norwegian Insurers
Norwegian insurers do not operate under the FCA Consumer Duty framework. They operate under Finanstilsynet's tilsynspraksis and the Forsikringsavtaleloven. The Forsikringsavtaleloven contains substantive customer protection requirements: clear pre-contract information, comprehensible policy terms, fair and prompt claims handling, and the customer's right to complain to Finansklagenemnda.
Finanstilsynet's supervisory practice expects insurers to demonstrate that their products are appropriate for their customers and that claims are handled fairly. The evidence base expected is similar in character to Consumer Duty outcome evidence — it differs in specificity and formal structure, not in kind. The measurement infrastructure that generates Consumer Duty evidence also generates the evidence base that Finanstilsynet expects. The data sources are the same. The AI analysis is the same. Only the reporting format and the regulatory submission differ. Specific Norwegian regulatory requirements should be verified with qualified Norwegian legal counsel.
Frequently Asked Questions
We already measure NPS and complaints — is that not sufficient for Consumer Duty?+
No. NPS measures satisfaction. Consumer Duty measures outcomes. The FCA has been explicit: firms that submit NPS and complaints volumes as their primary Consumer Duty evidence are not meeting the standard. NPS does not measure whether customers received fair value. Complaints do not measure customers who experienced a poor outcome but did not escalate. The FCA expects outcome-specific evidence for each of the four Consumer Duty outcomes, segmented by customer group, with root cause analysis where outcomes are failing, and an action log showing what was done about it.[1]
How do we identify customers who received a poor outcome but did not complain?+
Post-claims outcome tracking is the primary method. Customers whose claims were denied are a defined population. Contacting a sample to understand whether the denial was understood and accepted, or experienced as unfair, reveals the silent outcome failures that complaints data misses. Contact centre call classification is a second method: calls about claim denials, coverage queries, and cancellations contain outcome signals that NLP analysis can identify at scale. AI models trained on this data identify the systematic patterns that individual case review cannot see.[2]
What does the FCA expect to see in a Consumer Duty board report?+
The FCA expects four things. Evidence for each of the four outcomes, not just aggregate satisfaction metrics. Segmented analysis that shows outcomes by customer group rather than only overall averages. Root cause analysis where outcomes are poor, explaining why they are failing and what the insurer is doing about it. An action log demonstrating that the board reviewed the evidence, identified failures, took action, and tracked the result. A report that presents improving metrics without identifying any failures will attract FCA scrutiny rather than satisfy it.[1]
How do we measure the price and value outcome for a product with low claims frequency?+
Low-claims products — travel insurance, legal expenses, breakdown cover — present a genuine measurement challenge. Claim frequency is low, so claims payout ratio is low, but that does not mean value was not delivered. The measurement approach for low-frequency products focuses on: the breadth of coverage relative to the premium charged; the proportion of in-scope events that resulted in a successful claim; customer understanding of what triggers a valid claim; and the availability and quality of non-claims benefits. The FCA accepts that value measurement for low-frequency products requires qualitative evidence alongside quantitative data.[1]
How do we build a vulnerable customer outcome tracking capability?+
Start with identification. Your complaints AI system flags vulnerability indicators in complaint text. Your contact centre system flags calls where vulnerability protocols were applied. Your claims system should flag claims associated with customers on the vulnerability register. These three sources define the vulnerable customer population for outcome tracking purposes. For each group, track the same outcome metrics as the general population: claims acceptance rate, resolution time, first-contact resolution, complaint rate. Report these separately in the Consumer Duty board report. Where vulnerable customer outcomes diverge materially from the general population, that divergence requires root cause analysis and a remediation action.[3]
What is the FCA's enforcement approach to Consumer Duty failures in insurance?+
The FCA's 2024 Consumer Duty insurance reviews resulted in supervisory letters to a number of insurers requiring evidence of outcome monitoring improvement within defined timescales. Formal enforcement actions for Consumer Duty failures have focused on price and value failures — specifically, products where the claims payout ratio indicated systematic poor value for specific customer segments. The average cost of a Consumer Duty-related supervisory action in the insurance sector in 2024 was NOK 2.4 million, including remediation, management time, and the cost of the required consumer redress programme.[1]
This article provides general information only and does not constitute legal or regulatory advice. FCA Consumer Duty obligations, FCA PRIN 2A requirements, Finanstilsynet expectations under the Forsikringsavtaleloven, and vulnerable customer treatment obligations require case-specific legal assessment. UK and Norwegian insurers should consult qualified legal and compliance counsel for guidance specific to their jurisdiction and outcome monitoring programme.
References
All statistics sourced from documented deployments and third-party research organisations. All currency figures in NOK. Links verified 2026. Click any citation to jump to its source.
What good customer outcomes look like in insurance and how carriers are measuring them