You can read a UXDuty scan in 30 seconds for the headline, or 30 minutes if you want to dig into every issue. The page is laid out so that the most important information sits at the top, in the language a non-technical reader needs.
Every finding sits against one of the four FCA Consumer Duty outcomes - Products and services, Price and value, Consumer understanding, Consumer support - and cites the specific PRIN 2A handbook rule it relates to, so the path from a UX or accessibility issue to a regulatory implication is explicit.
Two related but separate concepts run through the report: severity sits on each individual issue, and the Consumer Duty risk band sits on the journey as a whole. Knowing the difference makes the report easier to act on.
The Consumer Duty risk band
The single most important number on the page. It comes in three bands, calculated from the count and severity of issues across every step of the journey, weighted toward issues that disproportionately affect customers in vulnerable circumstances:
- High - there is a clear potential for customer harm. Score below 50. Should be raised promptly with whoever owns the journey.
- Medium - there is meaningful friction or a Consumer Duty shortfall that would warrant attention if reviewed by a regulator. Score 50-79. Worth fixing in the next sprint.
- Low - the journey is broadly safe. Score equivalent to 80 or above on the underlying scale. Customers can complete it without significant friction.
The AI summary
Two or three sentences in plain English describing what the scan found. Designed to be pasted straight into a board paper or shared with a non-technical stakeholder. The summary is grounded in the actual issues - it won't describe problems that aren't in the issue list.
The issue chart
A visual count of issues by severity (High, Medium, Low) or by Consumer Duty outcome (Products and services, Price and value, Consumer understanding, Consumer support). Use it to scan the shape of the problem at a glance - a chart heavy on red means "a few critical issues"; a chart heavy on a single outcome means a particular PRIN 2A rule is at risk across the journey.
The journey list
Each step in the journey appears as a row, in order. Clicking a step expands it to show:
- The screenshot, with red outlines around any element involved in an issue.
- The issue list for that step - title, plain-language description, severity and category.
- A suggested fix for each issue, written by the same model that produced the summary.
Severity (per issue)
Each individual issue is graded on the same three-level scale as the journey:
- High - the issue blocks or significantly impedes a customer from completing the journey, or creates a clear Consumer Duty risk. Fix as a priority.
- Medium - the issue creates real friction but most customers can work around it. Fix in the next planned cycle.
- Low - the issue is a polish item or affects a small subset of users. Worth fixing when you're already in the area.
Severity is decided by a combination of the heuristic that triggered the issue, the specific PRIN 2A rule it sits against, and an AI judgement of how the issue affects a real customer trying to complete the journey. Issues that disproportionately affect customers in vulnerable circumstances are weighted upward.
Consumer Duty outcomes
Every issue is tagged with the Consumer Duty outcome it sits under, and cites the specific PRIN 2A handbook rule that backs it. The four outcomes cover the journey from product design through to ongoing support:
- Products and services (PRIN 2A.3.4R) - target-market fit, headline-claim integrity, FSCS protection signalled where it applies.
- Price and value (PRIN 2A.4.2R) - rate prominence, introductory-rate clarity, exit fees, total-cost illustration, representative examples and sourced comparison claims.
- Consumer understanding (PRIN 2A.5.3R) - plain language, acronyms explained, sentence length, passive voice on obligations, document guidance, comprehension checkpoints.
- Consumer support (PRIN 2A.6.2R, read with the cross-cutting 2A.2.8R foreseeable-harm rule) - help availability, vulnerability support, debt-advice signposting, claims pathway visibility, switching and transfer support, symmetric friction between joining and leaving, plus user-facing accessibility controls.
Accessibility findings are not a separate category - they run as part of each outcome's check set, mapped to the WCAG 2.2 AA success criterion they sit against. An alt-text issue on a product comparison page surfaces under Products and services; a contrast failure on a complaints form surfaces under Consumer support.
Why the band can change between scans
A scan reflects the journey at the moment it was run. If the page changed (a new banner, a copy change, a popup), the next scan can produce a different result even if you didn't intend to change anything. We recommend running a scan on a regular cadence - weekly or monthly is typical - so trends are visible rather than surprising.
Where to start
If you've only got ten minutes:
- Read the AI summary.
- Look at the High-severity issues only.
- Open the screenshot for the first one and decide whether to fix it now, schedule it, or escalate.
The rest can wait - but it's usually a good idea to make sure someone owns the Medium list before the next scan.
All documentation