Skip to main content

Find where users are struggling

Ask Fullstory MCP what's frustrating your users and it will surface the biggest friction points — ranked by user count, grounded in real session evidence, and ready to investigate.

Example questions​

What are the biggest frustrations across our site in the last 7 days? Rank by user count.
Fullstory MCP
What's causing the most friction for mobile users on our checkout flow?
Fullstory MCP
What got worse this week vs. last week? Anything trending up sharply?
Fullstory MCP
Show me the top rage-clicked and dead-clicked elements across our product pages.
Fullstory MCP

What you get back​

A ranked list of friction signals with user counts, where each one happens, and a short description of what users are actually doing. For example:

AI Response

Top frustrations across the site (last 7 days)

  1. Rage clicks on "Apply Promo Code" button: 412 users
    Concentrated at /checkout, 73% mobile. Users click 4 to 8 times before leaving the page without completing purchase.

  2. Dead clicks on the header cart icon: 287 users
    Affects /products and /collections. The icon looks clickable but does not respond on first tap on Safari mobile.

  3. Form abandonment on /signup at the password field: 218 users
    Higher rate on first-time visitors than returning. Average time on the field before leaving: 22 seconds.

  4. 504 timeouts on /api/orders during peak hours: 156 users
    Concentrated 7 to 9 PM ET. Users see a generic "Try again" message and abandon checkout.

Follow-up questions​

Tell me more about that promo code issue. Which devices and browsers are most affected, and is it getting worse?
Fullstory MCP
Pull 3 sessions where users hit the dead cart icon. What were they trying to do before they got stuck?
Fullstory MCP
Of the rage clicks across the product pages, which ones look like real bugs vs. user confusion?
Fullstory MCP

Example follow-up​

Fullstory MCP
Tell me more about that promo code issue. Which devices and browsers are most affected, and is it getting worse?
Fullstory MCP
AI Response

"Apply Promo Code" rage clicks (last 30 days)

  • 1,617 affected users
  • Device split: 73% mobile, 27% desktop
  • Top affected page: /checkout
  • Frustration rate vs. the rest of the site: +34%

Reviewed 4 sessions. The common pattern:

  1. User enters a valid promo code in the field
  2. Clicks "Apply"
  3. A small toast appears reading "Promo applied"
  4. The cart total stays the same
  5. User clicks "Apply" again, 4 to 8 times
  6. Eventually leaves the page

The toast is misleading. The promo is not actually being applied, but the UI reads as if it was. Users have no way to tell the difference between a successful and a failed application.

For a browser-level cut or a week-over-week user count, the MCP would chain a follow-up build_metric call grouped by browser or as a daily trend. Those numbers aren't part of the live-stats response on their own.

Behind the scenes​

The MCP routes the chain above through: get_opportunities, get_opportunity_stats, get_sessions_for_opportunity, and get_session_events. The live-stats response includes user count, device split, page breakdown, and rate-of-change percentages; richer dimensional cuts (browser, custom user property) require chained build_metric or build_segment calls.