Microsoft Clarity

@Microsoft Clarity export the last 7 days of session recordings where users abandoned checkout, filtered by mobile devices

Exported 142 session recordings from your Clarity project matching the criteria: mobile users who reached checkout but didn't complete purchase between Jan 15-22. Data includes session IDs, device types, rage clicks (18 sessions), dead clicks (31 sessions), and exit pages. The export is available as a JSON file with timestamps, user paths, and interaction heatmaps. Note: recordings are anonymized per your project's privacy settings.

This example shows Clarity's core export capability for behavioral analysis. Requires a valid API key with read access to your Clarity project. Export size depends on traffic volume — high-traffic sites may hit rate limits or need pagination. The MCP returns structured data, not the actual video files.

@Microsoft Clarity pull rage click data for our pricing page from the past 30 days and summarize the top 3 problem areas

Retrieved rage click events from /pricing over the last 30 days (847 total events). Top problem areas: (1) 'Enterprise Plan' CTA button — 312 rage clicks, likely a z-index issue covering the clickable area. (2) FAQ accordion headers — 198 rage clicks, users expect instant expansion but there's a 400ms delay. (3) Currency selector dropdown — 127 rage clicks, mobile users can't tap it reliably on screens under 375px width. Data exported includes coordinates, device types, and session replay IDs for each cluster.

Demonstrates pairing Clarity's raw export with AI synthesis to surface UX friction. The MCP fetches event data; the AI interprets patterns. Accuracy depends on Clarity's event tracking setup — if custom events aren't configured, you'll only see default interactions. Rage click thresholds are Clarity's defaults (3+ clicks in 1 second).

@Microsoft Clarity export conversion funnel data for our signup flow, then compare drop-off rates between desktop and mobile users this month

Exported funnel data for the 4-step signup flow (landing → form → verification → dashboard) from Jan 1-22. Desktop users: 68% completion rate, primary drop-off at email verification (22% abandon). Mobile users: 41% completion rate, major drop-off at form step (34% abandon), likely due to keyboard obscuring field labels. Data includes 3,200 desktop sessions and 1,850 mobile sessions. The export contains step-by-step timestamps, device metadata, and exit URLs for incomplete funnels.

Shows how Clarity's export enables cross-device analysis when combined with AI reasoning. The MCP pulls raw funnel metrics; you must have funnel goals pre-configured in your Clarity dashboard. Export doesn't include A/B test segments unless you've tagged sessions with custom properties. Rate limits apply to projects with >1M monthly sessions.

When session replay data needs to live in your team workspace

A 5-person product team ships a checkout redesign and wants to review session replays during their weekly retro. The Microsoft Clarity MCP lets you pull heatmap and session data directly into Switchy threads instead of switching tabs to the Clarity dashboard. You can export rage-click sessions or drop-off funnels, paste them into a shared context, and annotate them alongside Linear tickets or Figma mockups. This works best when your team reviews 10-30 sessions per week—enough to justify the export step, but not so many that you need a dedicated analytics tool. If you're already using Clarity for free session recording, this MCP closes the loop between observation and action. Worth adding if your team treats UX bugs like code bugs: triaged, assigned, and tracked in one place.

Pulling Clarity replays into customer support threads

A 3-person support team gets a spike in "checkout broken" tickets and needs to see what users actually did. The Clarity MCP exports session recordings filtered by error events or page URLs, so you can drop them into Switchy threads next to the Zendesk ticket and the Sentry stack trace. You avoid the context-switch tax of opening Clarity in a separate tab, copying the session ID, and pasting it into Slack. This scenario assumes you've already tagged Clarity sessions with user IDs or error codes—the MCP doesn't add filtering logic, it just moves the data. If your support volume is under 50 tickets per week and you're already using Clarity, this MCP saves 2-3 minutes per escalation. Skip it if you don't have session tagging set up or if your support team doesn't need to see replays to resolve tickets.

When you need Clarity data in a recurring standup format

A 4-person growth team runs weekly experiments on landing pages and wants to review Clarity heatmaps and funnel drop-offs in their Monday standup thread. The MCP exports aggregated data—click maps, scroll depth, session counts—so you can compare last week's variant to this week's without opening the Clarity dashboard. You paste the export into a Switchy thread, tag the designer and the copywriter, and decide whether to ship the variant or roll back. This works when your team runs 2-4 experiments per month and needs a lightweight review cadence. If you're running 10+ experiments or need real-time data, you'll hit the export API's rate limits or find the manual export step too slow. Add this MCP if your growth process is already thread-based and you want one less dashboard to check during standups.