How does this fit with the GOV.UK Service Standard?

Points 1, 2, and 5 of the Service Standard expect ongoing user research with users who have access needs. Our reports provide qualitative behavioural evidence you can submit at alpha, beta, and live assessments run by the Central Digital and Data Office.

Do you replace our in-house user researchers?

No. We complement in-house teams. Your researchers own the strategic questions and the service vision. We run the behavioural sessions with neurodivergent participants and deliver structured findings your team can act on.

Can you test pre-release prototypes as well as live services?

Yes. We test Figma prototypes, HMRC and DWP style alpha builds, and live production journeys. The three-stream capture works on any web-based interface a participant can reach in a browser.

How do you handle participant data and privacy?

Facial expression analysis runs locally on the participant device. Only derived signals leave the device. Recordings and transcripts are handled under UK GDPR and the Data Protection Act 2018, and we sign departmental data processing agreements where required.

What does a typical engagement look like?

A study brief, panel recruitment matched to your service's audience, remote sessions on the journey you specify, AI correlation of the three behavioural streams, human review by our research team, and a structured report. Usually days, not months.

Behavioural research for GOV.UK services that meet the public where they are

Remote testing with neurodivergent participants on live GOV.UK journeys, start pages, eligibility checkers, and account flows. Three streams of behavioural evidence, correlated and human-confirmed, so service teams ship changes backed by what users actually did.

Try OpenScouter for Free All research methods

The regulator, the standard, and the evidence gap

Public sector digital services in the UK sit under the Public Sector Bodies (Websites and Mobile Applications) (No. 2) Accessibility Regulations 2018, which require WCAG 2.2 AA conformance and a published accessibility statement. The Government Digital Service runs the GOV.UK Service Standard and the Service Manual, and the Central Digital and Data Office assesses services against both. Conformance is the floor, not the ceiling.

Service assessments increasingly ask teams to show evidence that real users, including disabled and neurodivergent users, can complete the journey. An accessibility audit tells you whether a Universal Credit eligibility check or a Self Assessment sign-in page passes a checklist. It does not tell you whether a citizen with ADHD abandons at the address lookup, or whether a dyslexic user misreads a National Insurance prompt.

That gap, between automated conformance and observed citizen behaviour, is where service teams lose completion. Behavioural research is the evidence method that closes it.

Evidence

GOV.UK content design guidance targets a reading age of 9 for content for the public, in line with the principle that government services must be usable by the widest possible audience

GOV.UK Style Guide, Content design and plain English · 2024

For GOV.UK services this anchor is not a stylistic preference, it is the operating principle the rest of the service is built around. If guidance, error messages, and confirmation screens are written to be understood by the widest possible audience, then the evidence question for a service team is behavioural: do citizens actually parse the content, complete the step, and leave with the right outcome. A neurodivergent panel testing live journeys is a direct way to check that the plain English target is doing the job it was set, particularly on high-stakes flows like benefits eligibility, tax returns, and identity verification through GOV.UK One Login.

Our approach

Tested on live GOV.UK patterns

We run sessions on real service journeys: start pages, GOV.UK One Login flows, eligibility checkers, document upload steps, and confirmation screens. Participants use the service as they would at home, on their own devices.

Three behavioural streams, correlated

Interaction signals (clicks, scrolls, rage clicks, hesitation), think-aloud voice, and on-device facial expression analysis are captured in parallel and correlated by our AI pipeline. Every insight is human-confirmed before it reaches your service team.

Neurodivergent panel as higher-signal users

Participants with ADHD, autism, dyslexia, and low vision surface friction that neurotypical testers miss. For GOV.UK content, where comprehension drives completion, that signal is directly relevant to whether the service works.

What you receive

Behavioural findings mapped to specific steps in the GOV.UK journey, with timestamped video evidence
Comprehension and hesitation analysis on form fields, error messages, and guidance content
A prioritised list of usability issues ranked by observed impact on task completion
Direct quotes and think-aloud transcripts you can cite in Service Standard assessments
A plain summary suitable for sharing with policy leads, content designers, and delivery managers

Frequently asked

How does this fit with the GOV.UK Service Standard?: Points 1, 2, and 5 of the Service Standard expect ongoing user research with users who have access needs. Our reports provide qualitative behavioural evidence you can submit at alpha, beta, and live assessments run by the Central Digital and Data Office.
Do you replace our in-house user researchers?: No. We complement in-house teams. Your researchers own the strategic questions and the service vision. We run the behavioural sessions with neurodivergent participants and deliver structured findings your team can act on.
Can you test pre-release prototypes as well as live services?: Yes. We test Figma prototypes, HMRC and DWP style alpha builds, and live production journeys. The three-stream capture works on any web-based interface a participant can reach in a browser.
How do you handle participant data and privacy?: Facial expression analysis runs locally on the participant device. Only derived signals leave the device. Recordings and transcripts are handled under UK GDPR and the Data Protection Act 2018, and we sign departmental data processing agreements where required.
What does a typical engagement look like?: A study brief, panel recruitment matched to your service's audience, remote sessions on the journey you specify, AI correlation of the three behavioural streams, human review by our research team, and a structured report. Usually days, not months.