This month is about reach and visibility: MCP opens up to every Enterprise team, your results pages surface the detail you used to dig for, and admins finally get a clear view of where credits go. Watch the full walkthrough below, then dig into each release.
What's in the roundup:
Maze MCP Now available to every Enterprise team — no application or waitlist. Connect Claude, ChatGPT, Cursor, and more to ask questions against your real participant data while the decision is still open, and choose exactly which teams your agent can reach.
Screener results in Analyze Your screener now appears as a result block alongside everything else, with qualifying rate, qualified and disqualified counts, and per-question answer distributions in one place. Click any participant row to review their responses or jump straight into their session.
Fewer repeated questions from the AI Moderator The AI Moderator now moves on as soon as a learning goal is covered instead of stretching to fill its allotted time, so participants stop getting asked the same thing three different ways. Studies can now cover up to 8 learning goals instead of 5, and conversation depth is now Brief or Thorough — studies already running on Deep are unaffected.
Element clicks Prototype results now come with a ready-made list of what participants clicked on each screen, labeled with the real Figma layer name. Each card shows clicks, participants, and a thumbnail; find it in the insights modal next to the heatmap, with Click Areas still there for anything you want to measure by hand.
Build AI Moderated studies with AI Describe your study instead of walking through the wizard — paste a description, upload a file or image, or drop in a link, and Maze generates the learning goals, questions, probing instructions, and stimulus mapping. You land in Build mostly filled in and ready to review. Available to Enterprise teams with AI Moderator access.
See where your credits go Admins can now see how their team is spending studies and panel credits, broken down by workspace and user. Team settings → Usage gives you overview cards with remaining balance, filterable breakdown tables, and a full usage history of every study that counted toward your total — all included in exports.
Coming soon:
Start from your research goal Describe what you want to learn on the home page and Maze will recommend the right method — Unmoderated, AI Moderated, or Moderated — then draft the study for you. Rolling out in beta now, so you may already see it in your workspace.
Credit allocation by workspace Admins will be able to divide panel credits across workspaces, watch the shared pool update as they go, and get a heads-up before a workspace runs low.
Reuse a past audience Pick from your team's recent filter and screener setups at the top of the Audience step, review what's included, and apply it to your order in one go. No rebuilding from scratch.
Catch every release as it ships right here in the changelog.