Feedback is clear and helps correct errors
What this criterion checks
Evaluates whether feedback and error messages provided by the application are helpful and actionable. Assesses whether feedback is meaningful and provides additional context, whether error messages include guidance on how to fix the issue, and whether all feedback messages can be understood by the average user. Font size, font style, icon shape, and general layout unrelated to feedback are excluded.
How it's assessed
The agent navigates through the application following realistic user journeys, capturing screenshots at each step. Using AI-powered vision analysis, it evaluates error and feedback messages and checks whether error messages are generic and lack context, whether they offer constructive advice on how to fix the issue, and whether they help educate users about system behavior. It also checks whether error messages use technical jargon instead of plain language, whether links to supplementary information are provided for complex errors, whether help content is focused on the user's task, and whether help lists concrete steps to be carried out.
How reliable it is
This criterion is currently scoring below its expected accuracy range, so confirm each finding before acting on it. See Accuracy & confidence for how the confidence tags are set.