Accessible Experiences
Part of Accessible customer experiences
Recording accessibility barriers without guessing customer abilities
Write barrier records that separate observed actions, service responses, customer accounts and explanations still to check.
Record the service step, what happened and its effect on the customer's task. Keep any explanation separate until it is checked. “The participant could not submit the form using a keyboard in this session” gives a team something to investigate; a guess about the participant's abilities does not.
Describe the encounter
A useful barrier record identifies the task, service version or channel, starting point, action attempted, service response, assistance given and outcome. Add the date and a reference to the permitted research note or case record. If the outcome could not be checked, write “unknown”.
For example, in a hypothetical appointment task, a participant tries to reach a time-selection control with a keyboard, but focus moves past it. Record the control and attempted action. The possible cause might be the control, the test setup or something else. Assign a technical check before choosing an explanation.
Separate four kinds of statement
Record / Example from the hypothetical session
- Observed action
- The participant pressed Tab while moving through the booking form.
- Service response
- Focus moved past the time-selection control.
- Participant account
- The participant said they could not choose a time.
- Interpretation to check
- The time-selection control may be unavailable by keyboard.
Use the participant's own words only when captured accurately and permitted for use. Do not add a diagnosis or infer a person's literacy, confidence or capability from one interaction. If you need to know why they took an action, ask them.
UK Government research guidance advises researchers not to assume what a participant can do or why they did something; that principle also helps keep barrier notes accurate.
Types of Statements in Accessibility Barrier Records
- Observed Action
- The participant pressed Tab while moving through the booking form.
- Service Response
- Focus moved past the time-selection control.
- Participant Account
- The participant said they could not choose a time.
- Interpretation to Check
- The time-selection control may be unavailable by keyboard.
Keep the record useful and proportionate
Record enough context for the responsible team to reproduce or investigate the issue, without circulating unnecessary personal detail. A person's diagnosis is rarely needed to describe a failed control or unclear instruction. Access needs may still matter to arranging participation or interpreting a specific interaction; handle that information through the organisation's approved process.
For organisations covered by the Australian Privacy Principles, collection of personal information must meet the applicable necessity rules. Sensitive information has additional conditions, including consent unless an exception applies. Review the record under the organisation's actual privacy obligations rather than storing every detail disclosed during a session.
Finish with the customer consequence, the next check and an owner. If another participant completes the same task, retain that contrary case. A small study can identify a barrier worth fixing, but it cannot establish how many customers encounter it. After a proposed repair, test the affected task again and record the observed result.
Privacy Considerations for Accessibility Research in Australia
- APPI Requirement
- Collection must be necessary and proportionate
- Sensitive Information
- Requires consent unless an exception applies
- Diagnosis Not Required
- To describe a barrier, personal diagnosis is rarely needed
- Retention Rule
- Only record what’s needed for investigation or repair
