User researcher¶

The guardrails an user researcher leads, what to show in each phase, the patterns that help and when to work with an architect.

Prototype - testing with users

This section is an early prototype that we are testing with users. It may change a lot, and some of it may be wrong. Check with your solution design authority before relying on it, and tell us what works and what does not through the feedback link above.

You find out what users need and test whether the service meets it. Research also tests architecture decisions - offline working, acting for others, human oversight of AI - and research data must be handled safely.

You lead 8 guardrails, often with other roles. Leading means making sure the team meets them and can show it, not doing all the work. See them all in the guardrail library, filtered to your role.

Your guardrails, phase by phase¶

Discovery¶

Guardrail	Level	What to show
GR-DATA-10 Handle research data safely	Should	Consent forms and privacy notice in use, recordings and notes stored only in approved places, and a deletion date set
GR-DATA-11 No real personal data in prototypes	Should	Any prototype uses made-up data
GR-DIG-01 Challenge paper and manual processes	Should	Discovery findings showing paper, email and manual steps in the current process and how the new design removes or justifies each one
GR-FIELD-01 Choose devices that suit the job	Should	User research on the working environment, and the device choice recorded in an ADR
GR-FIELD-04 Check connectivity before you design	Could	Connectivity in the places the service will be used checked, and the approach recorded

Alpha¶

Guardrail	Level	What to show
GR-AI-03 Keep a human accountable	Must	Design of the human oversight and challenge route for decisions with significant effects, tested with users in prototypes
GR-DATA-10 Handle research data safely	Should	The same for alpha research, with DPIA screening done for the research and any new research tool assessed
GR-DATA-11 No real personal data in prototypes	Should	Prototypes and research materials use made-up data, including data a participant types in during a session
GR-DIG-01 Challenge paper and manual processes	Should	Discovery findings showing paper, email and manual steps in the current process and how the new design removes or justifies each one
GR-DIG-03 Provide assisted digital and offline routes	Should	Assisted digital and offline routes designed and tested with users who need them
GR-FIELD-01 Choose devices that suit the job	Should	User research on the working environment, and the device choice recorded in an ADR
GR-FE-05 Design for low bandwidth and rural users	Should	Page weight budget, save-progress design and testing on slow connections
GR-FIELD-04 Check connectivity before you design	Could	Connectivity in the places the service will be used checked, and the approach recorded

Beta¶

Guardrail	Level	What to show
GR-AI-03 Keep a human accountable	Must	Human review and challenge built into the service and tested with users and staff
GR-DATA-10 Handle research data safely	Should	Research data from earlier phases deleted on schedule, and the same controls for beta research
GR-DATA-11 No real personal data in prototypes	Should	Test and research environments use synthetic or anonymised data; any exception agreed through a DPIA
GR-DIG-03 Provide assisted digital and offline routes	Should	Assisted digital and offline routes designed and tested with users who need them
GR-FE-05 Design for low bandwidth and rural users	Should	Page weight budget, save-progress design and testing on slow connections

Live¶

Guardrail	Level	What to show
GR-AI-03 Keep a human accountable	Must	Records of human review, challenges raised and their outcomes, reviewed regularly
GR-DATA-10 Handle research data safely	Should	Research data handled the same way for ongoing research, and deletion checked
GR-DIG-03 Provide assisted digital and offline routes	Should	Assisted digital and offline routes designed and tested with users who need them

Patterns that help¶

No patterns yet. See the patterns section.

Working with architects¶

Share research findings that affect technical choices with the architect, such as connectivity, devices or who acts for whom.
Invite the architect to research playbacks.