FCA Consumer Duty outcomes monitoring:
what evidence do firms need?

Outcomes monitoring is the hardest part of Consumer Duty

Most firms have made progress on the identification side of Consumer Duty — training colleagues to recognise vulnerability, creating disclosure pathways, updating scripts and processes. The harder work is outcomes monitoring: the ongoing, systematic evidence that vulnerable customers are receiving appropriate treatment and that outcomes are fair.

Outcomes monitoring is harder because it requires linking multiple datasets: who was identified as vulnerable, what treatment they received, what the outcome was, and whether that outcome differed materially from non-vulnerable customers in comparable situations. Most firms have this data in different systems — and the FCA expects it to be connected, not reconstructed when needed.

What the FCA expects to see

Based on FCA guidance documents (FG22/5 and subsequent enforcement communications), adequate Consumer Duty outcomes monitoring includes the following components:

Coverage MI: The proportion of customers and interactions assessed for vulnerability, broken down by channel (phone, branch, digital, collections). The FCA expects coverage rates to be meaningful — a firm that assesses vulnerability in 2% of phone interactions does not have a systematic process.

Identification rates: The proportion of assessed interactions where vulnerability was identified, by driver category (health, life events, resilience, capability). This demonstrates that the identification process is sensitive enough to detect vulnerability at realistic prevalence rates.

Action data: What happened for each identified vulnerable customer — escalation to specialist team, adjusted treatment, referral to support services, complaint handling adaptation. The FCA expects a documented audit trail from identification to action.

Outcomes data: Whether vulnerable customers received appropriate treatment and whether outcomes (complaint resolution, collections outcomes, product decisions) differed materially from non-vulnerable customers in comparable circumstances. Adverse differential outcomes require explanation and remediation.

Board MI: Aggregate monitoring data reported to senior management and the board at defined intervals — demonstrating that outcomes monitoring is embedded in governance, not just operational compliance.

The documentation gap

Many firms that have invested in vulnerability identification have a documentation problem. Colleagues identify vulnerability in real time but the identification is not systematically recorded at the interaction level. Outcomes are tracked in one system; vulnerability status in another. Board MI is compiled manually from multiple sources on a quarterly basis.

When the FCA requests evidence of outcomes monitoring — as they will, increasingly, as Consumer Duty enforcement accelerates — a firm that cannot produce interaction-level documentation from the identification process through to the outcome is in a difficult position. Retrospectively constructing an audit trail is not the same as having one.

The £176M context

The FCA issued £176M in fines in 2024, up 230% year on year. Separately, £479M in customer redress was paid by UK financial services firms. Consumer Duty enforcement is the mechanism through which these figures will grow. The trajectory is clear: firms that cannot evidence proactive identification and fair outcomes monitoring will face enforcement action as the FCA completes its Consumer Duty review cycle.

The FCA has explicitly stated that outcomes monitoring is the area where it expects the most improvement. Identification processes have improved. The evidence that identification translated into fair outcomes — the harder evidencing requirement — has not improved at the same rate.

Building the evidence trail automatically

EchoDepth generates the interaction-level documentation automatically. Every assessed interaction produces a timestamped record: signals detected, severity rating, vulnerability flag (if applicable), and the action taken. This data is structured for direct export into the MI framework the FCA expects.

Coverage rates are calculated automatically from interaction volume and assessment volume. Identification rates are produced by driver category where signal data supports classification. Action data is recorded at flag creation and updated on resolution. The result is a continuous, structured audit trail — not a quarterly manual MI compilation exercise.