One focused experience
One working AI feature, one critical journey, one primary user group, one language (English) and one agreed desktop or mobile environment. Five remote moderated sessions cover up to two tasks in that journey. You identify and invite suitable users; we agree screening criteria and handle the research. This is qualitative research, not a population-level performance benchmark.
Agree access before starting
We confirm the test environment, reference information, participant consent and access before accepting the scope. Automated scenario execution depends on a successful access and feasibility check; the method is agreed in writing. Custom integrations, external recruitment, specialist domain reviews, implementation and ongoing monitoring are outside this package.
Timing and the follow-up
The initial 10-business-day sprint starts once scope, the first payment, access and participant bookings are ready. Delays or changes are discussed and dates revised together. Request the included recheck within 30 days of the findings walkthrough; we schedule it within five business days of receiving a stable revised version. It covers up to six original scenarios, three attempts each, and an updated findings note. Additional user sessions are separate.
We provide evidence and recommendations to inform your decision. The evaluation does not certify safety, compliance or launch readiness, or guarantee a commercial result. The scope and terms are confirmed before payment.
