← Back to ResourcesConsumer Duty
What Consumer Duty evidence does the FCA expect to see in 2026?
Karthigeyan R J#Consumer Duty#FCA#2026
In 2026 the FCA expects a Consumer Duty evidence pack built on outcome monitoring: per-product metrics with thresholds across the four outcomes, distributions rather than averages, forward-looking harm mapping and an annual board report that shows action. Its April 2026 observations on Year 2 board reports and its July 2026 good and poor practice examples say exactly where firms fall short.
Three things this article will leave you able to do
Assemble the six components of an outcome-monitoring pack that survives review.
Replace average-based MI with the distribution views the FCA now asks about.
Check your next board report against the FCA's published poor-practice list.
Three years into the Consumer Duty, the FCA has stopped explaining the rules and started publishing marking schemes. Its observations on Year 2 board reports arrived in April 2026. Good and poor practice examples on outcome monitoring, drawn from multi-firm work including the insurance review updated on 13 March 2026, followed in July. Between them, firms now have something rare: a written description of what the regulator considers good evidence.
Which removes the last excuse. In 2026 a thin pack is not a maturity problem; it is a choice.
Who this applies to
Every firm in scope of the Consumer Duty, which since July 2023 means nearly every FCA-authorised firm with retail customers in its chain, whether or not it faces them directly. Closed products came into scope in July 2024. Cryptoasset firms join from 25 October 2027 and should read this as their preview: the expectations below are the ones their authorisation applications are being marked against already.
The six components of the pack
Outcome metrics, per product, with thresholds. For each of the four outcomes (products and services, price and value, consumer understanding, consumer support), defined metrics per product or segment, each with a threshold that triggers action. A metric without a threshold is a chart, not a control.
Distributions, not averages. The FCA's reviewers now ask for outcome distributions. An average complaint resolution time of four days can hide a tail of customers waiting forty. The pack shows the spread, the tail and what is being done about the tail.
The vulnerable customer split. Every metric cut by vulnerability characteristics. If outcomes for vulnerable customers cannot be distinguished in your data, that fact is itself a finding to report and fix.
Forward-looking harm mapping. A maintained map of foreseeable harms per product, refreshed when products, markets or technology change (the FCA has flagged AI-introduced risks specifically), each harm linked to the metric that would detect it.
Distribution chain evidence. Data exchanged with manufacturers or distributors: what you sent, what you received, what you did when the numbers looked wrong. Silence in the chain reads as a firm that has not asked.
The actions log. Where monitoring found poor outcomes: what changed, when, owned by whom and whether the metric recovered. The FCA's poor-practice examples are full of firms that monitored diligently and acted never.
The board report that carries it
The annual board report is where the pack becomes accountable. The FCA's April 2026 observations describe the good ones: results of outcome monitoring including the poor outcomes, actions with dates, strategy alignment and a fair value analysis that engages with what actually drives profitability across customer segments. The weak ones share a signature: backward-looking, average-based, harm mapping last updated at implementation and vulnerable customers mentioned once, in the glossary.
What you must evidence
How firms handle this
Most firms have MI; fewer have monitoring. The difference is thresholds, distributions and an actions log. It is visible within minutes of opening a pack. Some firms rebuild the pack manually each quarter, which consumes the team the Duty was meant to redirect towards customers. Our AuditGeniee keeps outcome evidence attached to the PRIN 2A obligations it proves, so the board pack assembles from live artefacts rather than from a quarterly scramble. However you build it, test it against the FCA's own published lists: they have told you the marking scheme.
Primary sources
FCA, observations on Consumer Duty board reports (April 2026) and good and poor practice examples on monitoring consumer outcomes (July 2026).
FCA, insurance multi-firm review: outcomes monitoring under the Consumer Duty. Updated 13 March 2026.
FCA Handbook, PRIN 2A.
Volatile. Re-verify before each republish: the exact titles and dates of the April and July 2026 FCA publications · any new multi-firm review findings · the crypto timetable reference (25 October 2027).
#Consumer Duty#FCA#2026