Maze
Maze is a continuous product discovery platform powered by AI. The platform is designed to enhance product research workflow by delivering ...
Last verified:
What is Maze?
Maze is a continuous product discovery platform that brings together every method, every team, and every stage of user research in a single place. It enables product teams to collect and analyze customer insights to speed up product development, combining qualitative and quantitative research methods including moderated interviews, prototype tests, usability studies, surveys, tree testing, card sorting, and live website/mobile testing.
Key features include an intuitive drag-and-drop maze builder with 17 research block types, access to a panel of over 6 million participants across 130+ countries with 400+ demographic filters, AI-powered functionalities for automating recruiting, moderating, and summarizing research, automated visual-rich reports with interview clips and AI transcripts, and integrations with Figma, Figma Make, Loveable, and Bolt for importing prototypes. The platform supports both unmoderated and moderated research, in-product prompts, participant management, and screener questions.
Maze is designed for product teams including designers, product managers, researchers, and product leaders who care about customer experience. It's used by over 60,000 teams worldwide, from startups to enterprises like industry-leading companies. The platform enables everyone to run credible research, make better product decisions, and scale insights company-wide without bottlenecks.
The platform emphasizes research at the pace of change, turning insights into clarity teams can act on. All discoveries live in one hub ready to share, with automated reports highlighting relevant clips, stats, and stories that turn research into evidence driving decisions. Maze is SOC 2 Type II certified, GDPR compliant, and offers encrypted transmission, access control, SSO, and private workspaces with role-based access.
Maze pricing
Pricing model: Freemium
Free Plan: $0/month includes 1 study per month, 1 active project, up to 3 mazes per project, 5 seats, essential prototype testing, surveys, and pay-per-use panel credits. Starter Plan: $99/month (or $1,188/year) offers 1 study per month, unlimited active projects and mazes, 5 seats, and pay-per-use credits for additional testers from the Maze panel. Team Plan: $15,000/year provides up to 20 studies annually, unlimited active projects and mazes, up to 2,000 rows for CSV uploads, and 5 seats. Organization Plan: Custom pricing includes unlimited projects, unlimited seats, moderated interview studies, AI insights, advanced permissions, SSO, custom contracts & invoicing, dedicated customer success manager, priority support, and full mobile testing. Enterprise Plan: Custom pricing built for organizations scaling user research with custom study limits, unlimited seats, Maze panel access, moderated and AI-moderated interviews, all Maze AI features, automated analysis, presentation-ready reports, enterprise-grade security, priority support, and dedicated CSM. Credits are billed separately across all plans.
Maze pros
- Access to 6 million+ participants across 130+ countries
- 400+ demographic filters for precise participant targeting
- Drag-and-drop builder with 17 research block types
- AI-powered study builder creates ready-to-launch studies in seconds
- AI moderator automates interview scheduling and summarization
- Supports both unmoderated and moderated research methods
- Integrates directly with Figma, Figma Make, Loveable, and Bolt
- Automated visual-rich reports with interview clips and AI transcripts
- Tree testing and card sorting for information architecture validation
- Live website and mobile app testing capabilities
- 5-second tests for quick sentiment measurement
- A/B testing on prototypes with variant comparison
- In-product prompts for contextual participant recruitment
- Free plan available for individuals starting usability testing
- SOC 2 Type II certified with enterprise-grade security
- GDPR compliant for data protection
- Private workspaces with role-based access control
- Turnaround times shortened from days to hours
- Over 60,000 teams trust the platform worldwide
- Templates library for practical research examples
Maze cons
- Unmoderated tests lack ability to probe deeper during research
- Limited for exploratory research compared to validation-focused studies
- Participant quality can be variable when using Maze Panel
- Less comprehensive than full-spectrum research platforms
- Primarily designed for prototypes, less suitable for live products
- AI features and data analysis may seem complex to beginners
- No physical customer support available
- Can be cost-prohibitive for small businesses or independent researchers
- Study templates constrain highly bespoke methodologies
- Complex branching logic may require workarounds
- Advanced permissions and compliance controls depend on subscription tier
Frequently asked questions about Maze
What is Maze?
Maze is a user research platform that allows product teams to collect and analyze customer insights to speed up product development. It brings together every method, every team, and every stage of research in a single place, so teams can go deeper, move faster, and influence more decisions.
What is Maze used for?
Maze is used by product teams to learn from their customers and use the insights and feedback to improve user experiences and build the right products faster. Practical use cases include collecting quantitative and qualitative insights to monitor customer satisfaction, perform usability testing, validate ideas and concepts, test prototypes and wireframes, and measure product sentiment.
Who is Maze for?
Maze is a valuable platform for those who care about customer experience, particularly product teams including designers, product managers, researchers, and product leaders. It's used extensively by over 60,000 teams, including some of the world's top companies.
What is a maze in Maze?
In the Maze platform, a maze is a succession of tasks and/or questions for testers to complete. It is the fundamental unit for testing in Maze. Each maze has its own results and reporting, and test results are collected on a per-maze basis. Mazes are composed of separate blocks including missions (tasks), questions, and additional screens.
How do you create a maze?
Creating a maze is made simple with Maze's intuitive drag-and-drop builder that comes with a variety of research blocks. For each task, question, and extra screen you want to add to your maze, you use a block. All block types are found on the right-side panel of the maze draft page, and you need at least one block in your maze.
What can you test with Maze?
Maze allows you to test anything including concepts and ideas, customer satisfaction, product sentiment, prototypes and wireframes, content, information architecture, or full live websites and apps. You can run any kind of study from big surveys to one-on-one interviews.
Can Maze do moderated interviews?
Yes. Interview Studies allow you to schedule, run, and analyze user interviews seamlessly with Maze. The platform offers solutions for both unmoderated and moderated product research, with AI moderator capabilities to automate the moderating process.
Can Maze do A/B testing?
Yes! Maze supports A/B testing on prototypes by letting you import multiple variants directly from tools like Figma, Figma Make, Loveable and Bolt. Maze randomly routes users to each version, measures success metrics like task completion and time on task, and generates a clear report so you can identify the optimal design quickly.
Do I get both quantitative and qualitative results?
Yes. Apart from best-in-market reporting functionalities for quantitative analysis, Maze allows you to capture and analyze audio, video, and text answers from research studies with AI. The platform delivers both quantitative usability analysis and qualitative analysis including clips recordings and interview transcripts.
Who are the test participants in Maze Panel?
Participants in Maze Panel are real people (not AI-generated) from more than 130 countries. You can target with precision the type of participants you want in your studies with over 400 demographic filters including role, language, income, qualifications, gender, and more. The panel includes over 6 million engaged participants worldwide.