The FCA’s targeted support regime launches in April 2026 for pensions and investments. Protection didn’t make the first wave, for understandable reasons. But treating that as an all-clear would be a mistake.
The principles underpinning targeted support, consumer understanding, outcome evidence, and demonstrable oversight are already embedded in Consumer Duty. They are shaping how the FCA assesses firms across retail financial services today. If the pensions and investments pilot succeeds, extending the model into protection will be hard to resist.
This is not hypothetical. The FCA’s pure protection market study highlights a significant protection gap, with around 72% of needs unmet. That kind of figure makes scaled consumer support an obvious regulatory priority.
Why protection is such a difficult test case
Protection advice is inherently personal. Suitability depends on health, dependants, debt, affordability, employer benefits, and exclusions. Unlike many investment decisions, it is difficult to guide someone without quickly encountering factors that determine whether the product is appropriate at all.
That makes the line between guidance and recommendation exceptionally narrow and easy to cross unintentionally. It rarely happens through explicit statements. More often, it emerges in small moments: simplifying exclusions, implying suitability, or introducing urgency to maintain momentum. At scale, these moments multiply. What begins as isolated behaviour can quickly become systemic.
The greatest conduct risk rarely sits in policy or training. It sits in the gap between what firms expect to happen and what customers actually experience.
The real challenge: evidencing outcomes
Historically, firms have been judged largely on intent and process, such as policies, scripts, training, and attestations. Targeted support shifts the emphasis decisively towards outcomes.
Regulators will not just ask what was designed or approved. They will expect firms to demonstrate what actually happened across every interaction and channel.
That includes showing how customer segments were defined, what support was delivered in practice, whether customers genuinely understood the information - not just whether they agreed - how vulnerability was identified and addressed, and what outcomes followed.
This is where many firms will struggle. Manual QA based on small samples may have been acceptable in a stable environment. In a scaled support model, it is not. Small errors become systemic risks.
In protection, misunderstandings often surface late, when a customer cancels because the cover wasn’t what they expected, or when a claim dispute reveals that exclusions were never fully understood. By then, the harm is done. Firms are left to manage remediation rather than prevent it.
Five conduct risks to focus on now
The first is misunderstanding risk. Customers often agree without comprehension. Signals like repeated clarification, vague responses, and confusion are easy to miss in live conversations and almost invisible in sampled QA.
The second is affordability and persistency risk. Support that nudges customers towards action can mask affordability issues, which later appear as lapses, and cancellations are patterns difficult to justify under Consumer Duty.
Thirdly, vulnerability amplification. Illness, bereavement, financial stress, and anxiety are central to protection. Treating vulnerability as a checklist rather than a lived reality creates evidential gaps.
Fourth is distribution inconsistency. Around 80% of pure protection sales go through intermediaries. A well-defined framework at the manufacturer level is often reinterpreted at each hand-off, leading to inconsistent customer experiences within the same segment.
The fifth is late-emerging dissatisfaction. Most dissatisfied customers simply disengage. Early indicators of confusion or frustration are some of the most valuable signals firms have, yet many fail to capture them.
What effective oversight looks like at scale
The next evidence standard cannot be met through human-only oversight. The volume and complexity of real interactions make that unrealistic.
AI-led interaction monitoring changes the equation, not by replacing human judgment, but by making patterns visible. Moving from sampling to analysing 100% of interactions creates a fundamentally stronger evidence base.
It enables firms to detect language drift before it becomes systemic, identify vulnerability signals in real time, link frontline behaviour to downstream outcomes, and demonstrate control in a way regulators can trust.
Crucially, it shifts the risk model. Instead of discovering issues through complaints or enforcement, firms can show they identified and addressed problems before harm occurred. That distinction, preventative control versus reactive remediation, is increasingly central to regulatory expectations.
The takeaway for protection leaders
Protection may not formally enter the targeted support regime in the near term. But the underlying expectations are already taking shape. Consumer Duty has established that disclosure is not a proxy for understanding. The FCA's pure protection market study is examining how support is delivered and how outcomes are evidenced across the distribution chain.
The firms that will be best placed, whether or not targeted support formally extends to protection, are those that invest now in understanding what customers actually hear, understand, and act upon. Not what the training materials say should happen. Not what sampled QA suggests is happening. What is actually happening, at scale, in real conversations.
The biggest risk isn't that targeted support falls short. It's that it works exactly as intended, and that activity scales faster than the ability to evidence, monitor, and intervene. At that point, the absence of robust oversight isn't a compliance gap, but an exposure.
