BNPL regulation has arrived: the compliance challenge now is proving good customer outcomes
In a recent
16/07/26
Read more
From sampling to evidence: scaling oversight for Consumer Duty
In a recent Mailock “Beyond Encryption” podcast episode, a clear challenge emerged for regulated firms: how to move beyond limited sampling to achieve a complete and defensible understanding of customer outcomes. The discussion highlights that narrow review coverage creates blind spots across conduct risk and customer journeys, reinforcing that firms must review more interactions and do so in a way that evidences outcomes at scale.
As regulatory expectations continue to evolve, firms are under increasing pressure to move beyond partial insight and demonstrate a clear, defensible understanding of customer outcomes.
Under the Consumer Duty, this shift is explicit. Firms must not only monitor interactions, but assess, test, understand and evidence the outcomes customers receive consistently and at scale.
Traditional sampling approaches are not designed to meet this standard. Reviewing a small proportion of calls, files or customer interactions can highlight isolated issues, but it cannot provide a complete view of conduct risk or customer outcomes. This creates a material compliance gap, particularly where issues relating to disclosures, vulnerable customers or customer understanding may go undetected.
Put simply, small samples surface anecdotes they do not evidence systemic issues or provide the level of assurance regulators expect.
To close this gap, firms need to transition from selective, retrospective review to continuous, evidence-based oversight. This means building the ability to assess large volumes of interactions, identify emerging risks early and demonstrate outcomes in a structured and defensible way.
This is not just about deploying AI, but applying the right capabilities:
- Generative AI can support summarisation of interactions
- Predictive AI enables firms to analyse datasets at scale, detect patterns of risk and prioritise action
This distinction is important. Compliance challenges are not solved by producing more outputs – they are addressed by generating meaningful insight, identifying risk earlier and supporting informed decision-making.
Without this capability, oversight remains reactive. Reviews become slower, remediation becomes more complex, and evidencing outcomes becomes harder to defend under regulatory scrutiny.
By contrast, scalable, data-driven oversight enables firms to:
- Gain complete visibility across customer interactions
- Identify and assess conduct risk proactively
- Evidence good customer outcomes with greater confidence
- Reduce the cost, complexity, and disruption of remediation
Ultimately, the FCA is raising the bar for compliance. Firms must move from partial visibility to comprehensive, outcome-based assurance, and be able to demonstrate that assurance clearly.
Sampling alone cannot deliver this, and generic AI approaches will not provide the depth of insight required. What is needed is a structured, scalable approach to oversight that combines the right technology with a clear focus on compliance and customer outcomes.
Getting it right the first time
We support firms in strengthening governance and oversight frameworks, helping them move beyond sampling towards robust, evidence-based compliance that stands up to regulatory scrutiny and delivers better customer outcomes.
Evidence at scale starts with applying the right AI to the business challenge – which is why we partner with Recordsure to help firms strengthen compliance oversight with purpose-built AI.