Beyond the Policy Paperwork: A Board Scorecard for Proving Consumer Duty Outcomes in 2026

Consumer Duty has reached a more demanding stage. For boards of specialist assistance and insurance groups, the question in 2026 is no longer whether the right policies, registers and fair value assessments exist. It is whether those arrangements can be shown to produce good outcomes for customers across partner products, embedded insurance and assistance services. That shift matters because distribution in this sector is rarely simple. Products are often designed with partners, sold through intermediaries, administered by third parties and used at a moment of stress. A board scorecard should therefore follow the customer journey from product design through to claim, repair, replacement or recovery.

The shift from design to evidence

Early implementation work naturally focused on governance, documentation and board accountability. Those foundations remain necessary. In 2026, however, supervisors, partners and boards are asking for something more granular: what changed for customers, and how do we know? The evidence base should combine leading indicators, such as comprehension testing and call quality, with lagging indicators, such as complaint uphold rates, claim decision times and remediation volumes.

Documentation is the scaffolding, not the building. A firm can have a comprehensive Consumer Duty framework and still struggle to explain why a particular customer cohort waited too long for a repair, why a partner product generated repeat complaints, or why vulnerable customers needed to contact the business three times to resolve a simple issue. The scorecard should expose those gaps rather than smooth them over.

A practical board scorecard

Scorecards work best when they are small enough to be read and specific enough to prompt action. The following seven dimensions can be adapted to a specialist assistance or insurance group, whether the products are sold directly, through affinity partners or as embedded cover.

1. Product and partner oversight

Evidence should show that target markets are defined, product triggers are clear and partner arrangements do not create obvious conflicts. Useful metrics include the proportion of products reviewed on schedule, the number of partner products with unresolved outcome actions, and the extent to which customer understanding has been tested for embedded products. Partner oversight should look at customer results, not only contractual controls.

2. Price and value

Fair value is not a single ratio. It is a judgement supported by evidence across price, benefits actually used, claim acceptance, renewal and add-on value. Boards should see benefit utilisation, claim acceptance rates, average time to benefit, complaint reasons and retention by customer segment. A product can look competitive on price and still offer poor value if customers cannot use it when they need it.

3. Consumer understanding

Can customers explain what they have bought, what is covered and what happens next? Metrics can include comprehension testing results, cancellation and cooling-off reasons, call recordings tagged for misunderstanding, and sign-off of partner sales material. Plain English is not a branding preference; it is an outcome driver.

4. Customer support and claims

Assistance is judged when something has gone wrong. The scorecard should track call answer times, abandonment rates, first-contact resolution, repair or replacement cycle times, claim decision times, complaint uphold rates and Financial Ombudsman Service referrals. These are operational measures, but they are also conduct evidence.

5. Vulnerable customers

Identification is not the same as support. Boards should ask whether vulnerable customers experience different outcomes and whether staff can adapt conversations, hand off to specialists and provide forbearance where needed. Metrics can include repeat contact rates, quality assurance findings, training completion and outcomes for customers flagged as vulnerable. A group-level green rating can hide a red rating for a specific cohort.

6. Complaints and remediation

Complaints are an outcome signal, not a nuisance. The scorecard should show root-cause categories, time to redress, repeat complaints, remediation backlog and the number of proactive remediations. Boards should be able to see whether issues are isolated or systemic, and whether partner products are over-represented.

7. Data and board reporting

The scorecard should show evidence quality as well as performance. Each metric needs an owner, a source, a reporting frequency and a RAG status. Boards should ask how old the evidence is, how complete it is and whether actions have closed. A scorecard that cannot be traced back to source data is a presentation, not a management tool.

Reading the scorecard

Boards should avoid averaging away poor outcomes. A single group-level rating can obscure a product, partner or customer segment that is performing badly. The scorecard should be read by product, partner and customer cohort, and discussed alongside conduct risk, operational resilience and financial performance.

Three tests are useful. First, can the firm trace an outcome from product design to delivery? Second, can it show a change that resulted from evidence? Third, can it identify where it does not yet know? The third test is often the most revealing. Mature firms are willing to say that their evidence is incomplete and explain how they will close the gap.

The next phase

By 2026, the competitive and regulatory conversation is less about whether a firm has a Consumer Duty programme and more about whether it can demonstrate repeatable outcomes. For specialist assistance providers, that means treating claims and assistance data as conduct evidence, not just operational data. It means partner oversight that examines customer results, not only contractual controls. And it means vulnerable-customer support that is visible in outcomes, not just in policy language.

For a group such as CPP Group plc, the scorecard has to reflect the reality of partner products, embedded distribution and assistance services delivered under pressure. It will not answer every question. It should, however, give boards a defensible way to ask better ones, and a clearer basis for proving that good outcomes are not merely intended but evidenced.