Product discovery & usability research · PUBLIC RESEARCH BRIEF
MazeWhat should a prototype score be allowed to decide?
Maze describes moderated interviews and unmoderated surveys, usability tests and prototype tests, along with participant recruitment, collaboration and AI-assisted summaries and themes. Combining task metrics with qualitative feedback can speed interpretation, but task configuration, prototype fidelity and sampling still determine what a result can support.
Updated 2026-10-20 · Simulation results not yet generatedCHANGE ONE THING. LEARN WHAT MATTERS.
Three questions for the GTM team.
How should teams distinguish a navigation problem from a prototype-fidelity problem before treating misclicks or task failure as a product finding?
Set up this study →Would non-researchers design safer unmoderated studies with templates, required review gates or visible warnings tied to sampling, leading questions and unsupported conclusions?
Set up this study →PROPOSED AUDIENCE
Who should weigh in?
North American product designers, UX researchers, product managers, design leaders and research-operations teams evaluating Maze. Include researchers who design studies and non-researchers who run or consume them. Proposed audience only; no Maze project, participant response or customer account is included.
TWO TIME HORIZONS
Trial today. A habit tomorrow?
Near term · 0–90 days
Over 0–90 days, run the same fictional task through prototypes with deliberately varied fidelity, success paths and question wording. Compare issue detection, causal overreach, escalation to a researcher and the quality of the next-study plan.
Longer term · 3–12 months
Over 3–12 months, follow approved self-serve research practices across teams and product stages. Study template drift, repeated participant exposure, reuse of evidence and whether quick tests are corroborated after launch.
What would make the result actionable?
Use versioned prototype variants, known usability defects, task definitions, response scripts, sample metadata and blinded adjudication. A product decision should be validated with appropriate participants and real behavioral outcomes; a prototype metric alone cannot establish business impact.
A Gather simulation returns hypothetical customer reactions. Quantifying revenue, traffic or retention needs actual business inputs and validation against observed behavior.