behavioural research
Behavioural Research for Payment Processors: Find the Friction Before the FCA Does
Payment journeys are regulated, scrutinised, and abandoned. OpenScouter runs remote behavioural research sessions with neurodivergent participants to surface the exact points where checkout flows, dispute forms, and onboarding screens lose users and lose revenue.
Payment Processors Face a Specific Usability Problem
The FCA's Consumer Duty, introduced under PS22/9, places a direct obligation on firms to evidence that their communications and interfaces support informed consumer decisions. For payment processors, that means checkout screens, fee disclosures, chargeback flows, and authentication steps are all in scope. Good intentions are not evidence. Observed behaviour is.
Neurodivergent users, people with ADHD, dyslexia, or autism, represent a significant share of the population completing payments every day. They encounter friction that neurotypical users often navigate around without noticing: ambiguous error messages on declined cards, multi-step 3DS authentication flows that lose context mid-journey, and fee summaries that require re-reading several times before the total is clear. These are not edge cases. They are conversion and compliance risks.
Analytics tools show where users drop off. They do not show why. A rage click on a payment confirmation button tells you something is wrong. It does not tell you whether the user misread the currency, missed a required field, or simply did not trust the interface. Behavioural research closes that gap with voice, interaction signals, and facial expression captured in parallel.
Communications must equip consumers to make effective, timely and properly informed decisions, supported by tests that the communications are likely to be understood
For payment processors, this FCA requirement is not abstract. Every screen in a checkout or dispute journey is a communication in the regulatory sense: a fee summary, an authentication prompt, a refund timeline. The obligation to test whether those communications are likely to be understood means that self-assessment and assumption are insufficient. Observed behavioural evidence, of the kind OpenScouter produces, is the method that closes the gap between what a design team believes is clear and what a real user actually understands under the conditions of a live payment decision.
Our approach
Three-Stream Capture Across Real Payment Journeys
Participants complete live or prototype payment flows while OpenScouter captures interaction signals, think-aloud voice, and facial expression simultaneously. All three streams are correlated by the AI pipeline so that a hesitation on a fee disclosure screen is matched to the spoken comment and the micro-expression recorded at the same moment.
Neurodivergent Panel as a Higher-Signal Usability Test
Our panel includes participants with ADHD, autism, dyslexia, and related cognitive differences. They are not an accessibility-only cohort. They are a higher-signal usability panel. Issues they surface in a payment flow, such as unclear authentication prompts or ambiguous refund timelines, are typically present for all users but harder to detect with neurotypical-only testing.
Human-Confirmed Reports Your Compliance and Product Teams Can Use
Every report is reviewed by a human researcher before delivery. Findings are structured around specific journey steps, not general observations. Your team receives evidence that is usable in a product backlog, a design review, and, where relevant, a Consumer Duty board report.
What you receive
- Behavioural research report covering named payment journey steps, including checkout, authentication, and dispute initiation
- Timestamped session clips matched to specific friction points in the flow
- AI-correlated findings across interaction, voice, and facial expression streams
- Prioritised issue list with severity ratings and recommended interface changes
- Summary section formatted for inclusion in Consumer Duty outcome monitoring documentation
Frequently asked
- Which payment journeys can OpenScouter test?
- We can test any journey you can present in a browser or prototype: card checkout flows, open banking payment screens, 3DS authentication steps, chargeback and dispute forms, fee disclosure pages, and onboarding screens for merchant or consumer accounts. If your team can share a staging URL or a Figma prototype, we can run sessions against it.
- How does this relate to our FCA Consumer Duty obligations?
- Consumer Duty PS22/9 requires firms to evidence that their communications support informed consumer decisions. Behavioural research sessions produce observed evidence of how real users interact with your interfaces. Our reports are not legal opinion, but they are structured evidence that your compliance team can reference in outcome monitoring and board reporting.
- Why use neurodivergent participants specifically for payment research?
- Neurodivergent users tend to follow instructions more literally, tolerate less ambiguity, and verbalise confusion more clearly than neurotypical users. In a payment context, that means they surface issues with unclear fee breakdowns, confusing authentication prompts, and ambiguous error messages that other testing methods miss entirely. They are a higher-signal panel, not a specialist-only one.
- How long does a typical engagement take?
- Most engagements move from study brief to delivered report within a matter of days. The exact timeline depends on the number of journey steps in scope and the size of the participant cohort. We will confirm a schedule before work begins.
- Is OpenScouter a replacement for our existing analytics or research tools?
- No. OpenScouter is a complement to tools like Hotjar, Maze, or Dovetail. Analytics platforms tell you where users drop off in a payment flow. Behavioural research tells you why. The two types of evidence work together. We do not ask you to replace anything.
Keep reading
Talk to a behavioural researcher
Tell us about the vertical, the journey, and the evidence you need. We will scope a pilot in days, not weeks.