Hotjar MCP. Find out why users leave your site, not just that they did.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Hotjar MCP Server gives your AI agent full access to user experience analytics. You get heatmaps, session recordings, and detailed feedback surveys, letting you see exactly where visitors get stuck or why they leave your site.
Run full conversion funnel analyses and look up specific user behavior by ID, all through natural conversation.
What your AI agents can do
Check hotjar status
Verifies the connection status to the Hotjar account.
Get feedback widget
Retrieves specific details about a feedback widget.
Get heatmap
Gets detailed click, scroll, and move data for a specified page heatmap.
List funnels and get step-by-step data showing where visitors drop out of a conversion path.
List and get heatmaps to see where users click, scroll, and move attention on specific pages.
Look up a user by ID to get their entire session history and behavior data.
List feedback widgets and list responses to review user-submitted comments and screenshots.
List recordings and get individual recordings to observe how users interact with your site.
List surveys and get survey statistics to analyze NPS scores, completion rates, and response trends.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Hotjar MCP Server: 15 Tools for User Behavior Analysis
These tools let your agent perform every step of the user experience diagnostic process, from listing funnels to retrieving individual user recordings.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Hotjar on Vinkius019dd105check hotjar status
Verifies the connection status to the Hotjar account.
019dd105get feedback widget
Retrieves specific details about a feedback widget.
019dd105get heatmap
Gets detailed click, scroll, and move data for a specified page heatmap.
019dd105get recording
Retrieves details and metadata for a specific user session recording.
019dd105get survey
Gets the full details and settings for a specific survey.
019dd105get survey stats
Retrieves quantitative statistics for a given survey, like NPS scores.
019dd105list feedback
Lists all available feedback widgets configured on the site.
019dd105list feedback responses
Lists all incoming user feedback submissions.
019dd105list funnels
Lists all defined conversion funnels and their steps.
019dd105list heatmaps
Lists all available heatmap snapshots for different pages.
019dd105list recordings
Lists all recorded user sessions available for review.
019dd105list sites
Lists every site tracked within your Hotjar account.
019dd105list survey responses
Lists all individual responses submitted to a survey.
019dd105list surveys
Lists all the surveys you have set up in Hotjar.
019dd105lookup user
Retrieves all session history and behavior data for a specific user ID.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Hotjar, then connect any of our 4,800+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,800+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Hotjar. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 15 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Reading user behavior reports used to mean opening 12 different tabs.
Before this server, diagnosing a user issue meant jumping between the funnel dashboard, the heatmap tool, and the recording library. You'd spend an hour clicking, filtering, and cross-referencing drop-off numbers with vague notes about user frustration. You'd end up with a spreadsheet of numbers and a stack of screenshots, but no clear next step.
Now, you just ask your agent. You can ask, 'Why did the checkout funnel drop 15%?' and the agent pulls the `list_funnels` data, cross-references it with the `get_heatmap` for that step, and pulls the top 3 `list_feedback_responses` simultaneously. You get a clear, prioritized narrative in seconds.
Hotjar MCP Server: Watch the user, don't just count them.
Previously, if you wanted to see a user struggle with a specific flow, you had to manually search the recording library, find the right date, and scrub through the video timeline for 5 minutes. It was tedious, and you rarely saw the *key* moment—just the general struggle.
Now, you tell the agent to look up a user by ID. It pulls the full session history via `lookup_user` and provides the key metrics alongside the specific recording metadata. You pinpoint the exact moment of friction—the one that matters—without digging through endless video files.
What you can do with this MCP connector
Connect your Hotjar account to your AI agent and start seeing why people bail on your site. You don't gotta guess anymore. Your agent lets you pull user experience analytics just by talking to it.
Diagnose Visual Friction Points: You can list all available heatmaps, then get detailed click, scroll, and move data for any page you point it to. This shows exactly where users are clicking, where they're scrolling, and where their attention is going.
Analyze User Flow Drop-off: You can list all defined conversion funnels, and then get step-by-step data showing where visitors drop out of a conversion path. This lets you pinpoint exactly which step is killing your conversion rate.
Review Specific User Behavior: Give your AI client a specific user ID, and it pulls up that user's entire session history and behavior data. You get the full picture of what they did on your site.
Watch Past User Sessions: You can list all recorded user sessions available for review, and then get details and metadata for any single recording. This lets you observe how users actually interact with your site.
Manage Survey Data: You can list all your Hotjar surveys and then get quantitative statistics for any specific survey, like NPS scores, completion rates, and response trends. You can also list all individual survey responses and get the full details and settings for a specific survey.
Deep Dive into User Feedback: You can list all available feedback widgets and then list all incoming user feedback submissions. You can also get specific details about a feedback widget, and review all individual responses submitted to a survey, which often include screenshots.
Site and Account Management: You can list every site tracked in your Hotjar account, and you can also check the connection status to your Hotjar account.
019dd105-2c4c-73ea-bf8f-0a3ee9c9bae6 
How Hotjar MCP Works
- 1 Subscribe to the Hotjar MCP Server and provide your Hotjar Client Secret.
- 2 Your AI agent uses the tools to request specific data (e.g., 'list_funnels' for drop-off data, or 'get_heatmap' for clicks).
- 3 The server returns the structured data, which your AI client synthesizes into a plain-language report detailing user behavior.
The bottom line is, your AI client turns raw user data into actionable insights without you having to click through 15 separate dashboards.
Who Is Hotjar MCP For?
Product Managers who need to know why a feature isn't adopted. UX Researchers who need to prove a hypothesis about user interaction. Growth teams who are tired of guessing what page needs fixing. This is for anyone whose job depends on understanding the user's actual journey.
Tracks conversion funnels, identifies precise drop-off points, and monitors NPS scores to prioritize product fixes.
Reviews survey responses, analyzes heatmap patterns, and watches session recordings to understand user pain points.
Analyzes feedback trends and optimizes page engagement using data from the Hotjar MCP Server.
What Changes When You Connect
- See precise drop-off points. Use
list_funnelsto map conversion paths step-by-step. Pinpoint exactly which step causes users to abandon the process. - Visualize user attention with
get_heatmap. See where users actually click, scroll, and move their mouse on any page, finding visual pain points instantly. - Watch the problem live.
list_recordingsandget_recordinglet you play back actual user sessions. You see the friction firsthand, not just the metric. - Know the 'why.' Use
list_feedback_responsesandlist_feedbackto gather direct user quotes and screenshots of what they hate. - Quantify satisfaction. Running
get_survey_statsgives you NPS scores and completion rates, backed by the full data set fromlist_surveys. - Track specific users. The
lookup_usertool lets you pull a complete behavioral report for a single user ID, perfect for bug reproduction.
Real-World Use Cases
Checkout Funnel Failure
The conversion rate dropped 15% last month. The Product Manager asks the agent to run list_funnels and list_heatmaps. The agent identifies that the drop-off happens between the 'plan selection' and 'signup form' steps, and the heatmap shows users are repeatedly scrolling past the CTA button.
Investigating High Support Tickets
A support agent needs to understand why users can't use the export feature. They ask the agent to lookup_user for a sample user ID. The agent returns the user's session history, revealing they visited the report page multiple times but never interacted with the export button.
Understanding Low NPS Scores
The team sees an NPS score that is too low. The Growth Manager asks the agent to list_feedback_responses. The agent pulls the top submissions, revealing recurring complaints about mobile speed and confusing navigation.
Feature Adoption Mystery
The team suspects a new feature isn't being used. They run list_surveys to check the completion rate and then use get_survey to check the specific questions, confirming if the feature was even visible to the user.
The Tradeoffs
Dashboard Overload
Trying to manually correlate 15 different metrics (e.g., open rate, bounce rate, NPS, click count) across 5 tabs in a dashboard. This gives you data, but no single narrative.
→
Don't look at dashboards. Ask your agent to combine tools. For example, run list_funnels to scope the drop-off, then use get_heatmap on the failing step, and finally list_feedback_responses to find the qualitative reason. This gives a narrative, not just numbers.
Focusing on Correlation
Seeing that users who view the pricing page also submit bad feedback. This is a correlation, but you don't know why they feel that way.
→
Use list_feedback_responses first. Find the exact quote about the pricing page. Then, use get_heatmap to confirm if the issue is a visual element (like a hidden CTA) or if it's truly the content.
Ignoring the Full Picture
Only looking at the 'Completed Signups' metric and assuming the problem is the final button. You miss the friction that happened 10 steps earlier in the journey.
→
Always start with list_funnels. It forces you to look at the entire path. If the drop-off is early, use list_sites to narrow the focus, and then get_recording to watch the initial user interaction.
When It Fits, When It Doesn't
Use this if you need to answer the question, 'Why?' or 'Where exactly?' If you're just checking if a button works, use a basic testing tool. Use this when you need a full, multi-layered diagnosis: first, use list_funnels to find the drop-off where. Then, use list_heatmaps to see what the user was looking at. Next, use list_feedback_responses to understand why they left. Don't use it just because you have the tools; use it when you have a specific, high-stakes problem that requires combining quantitative metrics (funnels, heatmaps) with qualitative evidence (recordings, feedback). If the root cause is organizational (e.g., the product requires a separate account setup), this tool will tell you the data, but you still have to fix the process.
Common Questions About Hotjar MCP
How do I use the list_funnels tool to find a drop-off point? +
You use the agent to call list_funnels. This tool lists all conversion funnels on your site. The agent then gives you step-by-step drop-off data, showing the exact drop-off percentage and the transition between steps.
What does the get_heatmap tool actually show? +
The get_heatmap tool gives you click, scroll, and move data for a specific page. It shows a visual representation of where users focus their attention and how far down the page they scroll.
Can I check the NPS score using get_survey_stats? +
Yes, the get_survey_stats tool pulls the quantitative data for any survey. It provides the NPS score, the percentage breakdown of Promoters, Passives, and Detractors, and the overall completion rate.
How do I get detailed user behavior data for a single person? +
Use the lookup_user tool. You provide the user ID, and the agent retrieves the user's complete session history, total time on site, and pages viewed over time.
Is list_feedback_responses the right tool for qualitative data? +
Yes. list_feedback_responses pulls individual user submissions. You get the text, the rating, and crucially, the attached screenshots, giving you the raw qualitative evidence.
How do I use list_sites to find out which pages are being tracked? +
The list_sites tool provides a complete inventory of every domain or subsite currently configured in your Hotjar account. It's useful for verifying tracking coverage or checking if a new page needs to be added before analysis.
What is the difference between list_heatmaps and get_heatmap? +
list_heatmaps shows a list of all available heatmap snapshots you've generated. The get_heatmap tool retrieves the detailed data for one specific snapshot, allowing you to inspect click, scroll, and move metrics.
When should I use list_survey_responses instead of list_feedback_responses? +
Use list_survey_responses for structured, quantitative data from formal surveys. Use list_feedback_responses when you're looking at raw, qualitative user comments from the general feedback widget.
Can I view survey responses and NPS scores? +
Yes. Use list_surveys to see all surveys with response counts and status. Use get_survey_stats for aggregate metrics including completion rate, NPS score, and response trends. Use list_survey_responses to read individual user responses. Use get_survey for the full question list and settings.
Can I analyze heatmap data for specific pages? +
Yes. Use list_heatmaps to browse all heatmap snapshots, then get_heatmap with a specific Heatmap ID to retrieve click, scroll, and move interaction data along with the page URL. This reveals where users focus attention and which elements they interact with most.
Can I track conversion funnel drop-offs? +
Yes. The list_funnels tool retrieves all configured conversion funnels with step-by-step drop-off data. Each funnel shows the number of users entering each step and the percentage that continue to the next, helping you identify exactly where users abandon the flow.
Multi-server workflows that include Hotjar MCP
Score Leads by Site Behavior Using MCP Servers
Heatmaps analyzed, CRM contacts scored, behavioral insights logged , your website visitors reveal what surveys never capture
Track Post-Email User Behavior Using MCP
Your campaign had a 42% open rate but zero conversions , because everyone clicked, landed on the page, and bounced at the pricing section you never tested
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.