GrowingIO MCP. Analyze product behavior and pull metrics via chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
GrowingIO MCP Server gives your AI agent access to product analytics data from GrowingIO. Use it to list events, query metrics, audit funnels, and segment users via natural language.
It translates complex product behavior data into conversational insights, letting you monitor everything from feature adoption to campaign performance without touching the web UI.
What your AI agents can do
Get event
Retrieves specific metadata details for a single tracked behavioral event.
Get funnel
Gets the configuration and data for a specific conversion funnel.
Get metrics
Runs a quantitative query against your project's core performance metrics.
The agent lists and retrieves detailed metadata for all behavioral events tracked in your project.
The agent lists and retrieves user segments, allowing you to monitor defined high-value user groups.
The agent executes quantitative queries to retrieve specific performance metrics using natural language.
The agent retrieves detailed data for conversion funnels, helping you identify where users drop off.
The agent lists and retrieves metadata for tracked advertising campaigns.
The agent retrieves general project information and metadata.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
GrowingIO MCP Server: 10 Tools for Product Analytics
Access every core data function—from listing variables to querying live metrics—using your AI agent in a conversational way.
019d8444get event
Retrieves specific metadata details for a single tracked behavioral event.
019d8444get funnel
Gets the configuration and data for a specific conversion funnel.
019d8444get metrics
Runs a quantitative query against your project's core performance metrics.
019d8444get project info
Retrieves general metadata and configuration details about the GrowingIO project.
019d8444get segment users
Gets a list of users belonging to a specific defined user segment.
019d8444list ads
Lists and retrieves metadata for all tracked advertising campaigns.
019d8444list events
Lists all the behavioral event types tracked across your entire project.
019d8444list log sources
Lists all the data log sources connected to your GrowingIO project.
019d8444list segments
Lists all the defined user segments available for analysis.
019d8444list variables
Lists all the custom variables currently tracked within your project.
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 GrowingIO, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ 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
What you can do with this MCP connector
This GrowingIO MCP Server lets your AI agent run product analytics queries straight from your GrowingIO data. You can list all tracked events, get detailed metadata for specific behavioral events, list all defined user segments, and retrieve a list of all custom variables. You'll also find tools to list all data log sources and list all advertising campaigns.
When you need to know how users are doing, your agent can run quantitative queries against your project's core performance metrics using natural language. You can check the configuration and data for specific conversion funnels to see where users drop off. You'll also get general project metadata and configuration details for the entire GrowingIO project.
To figure out who your best users are, your agent can pull a list of users belonging to a specific defined user segment. Your agent can use the available tools to audit conversion funnels, run metrics queries, and pull user lists. Your AI client translates complex product behavior data into actionable insights, letting you monitor everything from feature adoption to campaign performance without touching the web UI.
This gives you access to every piece of data you need, from viewing campaign metadata using list_ads to getting general project info with get_project_info.
How GrowingIO MCP Works
- 1 Subscribe to the server and input your GrowingIO credentials (Project ID, Client ID, API Token).
- 2 Instruct your AI client to perform a task, like 'What was the DAU last week?'
- 3 The server runs the necessary tool (e.g.,
get_metrics) and returns the resulting data structure directly to your client.
The bottom line is, you get analytical answers in plain language without ever leaving your AI workspace.
Who Is GrowingIO MCP For?
This is for Product Managers who need to audit feature adoption without digging through dashboards. It's for Growth Engineers who need to coordinate user segments and monitor campaigns directly from their AI workflow. Data Analysts use it to run funnel analysis and retrieve system metrics via natural language, skipping manual report generation entirely.
Checks feature adoption rates and audits user behavior by asking the agent questions like, 'What are the top 3 events for new users?'
Coordinates user segments and monitors campaign performance, pulling metrics directly into an automated workflow from the AI agent.
Runs quantitative queries to retrieve system metrics or performs a full funnel analysis by simply asking, 'Show me the drop-off rate for the checkout process.'
What Changes When You Connect
- Audit Behavior in Minutes: Instead of clicking through event logs or dashboards, you ask the agent to list events or retrieve segment metadata. This cuts the time spent on basic product health checks from hours to seconds.
- Pinpoint Drop-offs Fast: The
get_funneltool allows you to audit conversion funnels and see exactly where users drop off. You don't need to manually calculate drop-off rates; the agent provides the details directly. - Query Data, Not Dashboards: Using
get_metrics, you run quantitative queries in natural language. This lets you compare DAU vs. WAU across segments without writing a single SQL statement or building a custom report. - Manage Users by Group: The agent uses
list_segmentsandget_segment_usersto identify specific high-value user cohorts. This lets you focus your analysis on the people who matter most, bypassing general user data. - Track Growth Campaigns: The
list_adstool lets you browse all tracked advertising campaigns. You can quickly identify which campaigns are driving the most signups or revenue without checking a separate ad platform. - Unified Source of Truth: By consolidating all this data—events, segments, metrics, and funnels—into the AI chat, you keep your entire product workflow in one place. It eliminates context switching between analytics platforms and your daily coding environment.
Real-World Use Cases
Determining Feature Adoption for a New Checkout Flow
A PM needs to know if a new payment button is being seen and used. They ask the agent to list_events for 'payment_button_click' and then run a get_metrics query to compare the click rate against the overall user base. The agent combines this data, showing the adoption rate and identifying if certain segments (retrieved via get_segment_users) are ignoring the feature.
Investigating a Sudden Drop in Signups
The Growth Engineer sees a dip in signups. They prompt the agent to get_funnel for the 'Sign Up' flow. The agent returns the funnel details, showing the biggest drop-off is at the 'Email Verification' step. This immediately tells the team where to focus their engineering efforts.
Analyzing Top-Tier User Behavior
A Data Analyst wants to understand the 'Power User' group. They first use list_segments to confirm the segment exists, then use get_segment_users to pull the user list. Finally, they ask the agent to get_metrics on the average session length for only those specific users.
Quickly Assessing Campaign Impact
The PM needs a quick read on how the last marketing push performed. They ask the agent to list_ads to see the campaign names, then use get_metrics to query the resulting conversion rate for each campaign, getting an immediate performance scorecard.
The Tradeoffs
Manually stitching together reports
Exporting raw event data from list_events and then manually joining it with segment data from list_segments in a spreadsheet just to count users. This is slow, prone to versioning errors, and requires hours of cleanup.
→
Instead, let the agent run a targeted query. Use get_metrics combined with list_segments to get a summary count, or use get_segment_users to pull the list and then run metrics on that subset. Keep the process inside the chat.
Asking for raw data dumps
Asking the agent to 'give me every single event record from the last quarter.' This floods the chat, wastes tokens, and gives you unusable noise instead of an answer.
→
Be specific. Instead of listing everything, use get_event to check a specific event type, or use get_metrics to query an aggregate metric (e.g., 'What was the average time on page?').
Ignoring the data lifecycle
Assuming all data is available for the same time range, leading to confusion between event metadata and live metrics.
→
Always check get_project_info first to understand the data scope, or use list_log_sources to verify what data streams are currently active before querying.
When It Fits, When It Doesn't
Use this if you need to move beyond 'what happened' and get actionable summaries. This server is for people who need to know why the numbers look the way they do. You must be able to articulate a clear question about product behavior, conversion rates, or user groups. You shouldn't use it if you just need to look at a single, raw CSV export or if your question is purely about infrastructure configuration outside of product data.
When NOT to use it: If your goal is simply to read the raw database schema or if you need a dedicated, complex data warehouse join that spans multiple, unrelated systems (e.g., joining GrowingIO to Salesforce). For those cases, you need a dedicated ETL pipeline or a specialized data warehouse connector, not just a data access layer.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by GrowingIO. 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 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Manually pulling product insights is a nightmare of tabs and filters.
Every day, you run into the same problem: a core metric is off, but finding out why requires hopping between the dashboard, the event log, and the segment definition page. You click the 'Funnel' tab, then switch to 'Events' to see the drop-off reason, then jump to 'Segments' to isolate the cohort. This process takes 20 minutes and involves three different copies of the same data.
With the GrowingIO MCP Server, you keep it all in the chat. You ask, 'Why did the conversion rate drop for Power Users?' The agent runs `get_segment_users` to define the group, then runs `get_funnel` against them, and finally uses `get_metrics` to give you the final answer. You get the root cause, instantly.
GrowingIO MCP Server: Manage events, segments & metrics
Forget copying and pasting metadata definitions. Instead of spending time running basic `list_events` to see if you tracked the right variable, just ask the agent. It confirms if the event exists, what data it holds, and if it has been used in a conversion funnel.
It doesn't just give you data—it runs the analysis for you. You state the problem, the agent executes the necessary tool calls (`list_variables`, `get_metrics`, etc.), and delivers the answer, complete with context and actionable next steps.
Common Questions About GrowingIO MCP
How do I use the get_metrics tool with GrowingIO MCP Server? +
You ask the agent a natural language question about performance. For example: 'What was the average session time for the last 7 days?' The agent translates that into a get_metrics call and returns the result.
Can I use list_segments to find out who is using the product? +
Yes. You first use list_segments to see all available user groups. Then, you ask the agent to use get_segment_users to get the specific list of users you want to investigate.
Is get_funnel the right tool to check conversion rates? +
Yes. get_funnel is designed specifically for this. You just need to tell the agent the name of the conversion flow (e.g., 'Checkout Flow'), and it returns the detailed drop-off points and current conversion rate.
How do I check if an event is tracked in GrowingIO? +
You use list_events. This tool lists every behavioral event type available in your project, letting you confirm if the data point you need was ever recorded.
How do I use list_events to see what data is tracked in GrowingIO? +
The list_events tool shows all behavioral events tracked in your project. It lists the event names and provides basic metadata so you know exactly what data is available for analysis.
What if I need to find out details about a specific user segment using list_segments? +
The list_segments tool retrieves a list of all defined user segments. You can then reference these segment names when using other tools to analyze specific user cohorts.
Can I use get_project_info to verify my GrowingIO account setup? +
Yes, get_project_info fetches general metadata about your entire GrowingIO project. This confirms your connection details and overall project status.
Where can I find the data for advertising campaigns using list_ads? +
The list_ads tool lists all tracked advertising campaigns. This lets you review campaign names and identify which growth initiatives are feeding into your product data.
How do I find my GrowingIO Project ID and Client ID? +
Log in to your GrowingIO portal, go to [Project Configuration] -> [Project Info]. You will find your Project ID (ai) and Client ID (Public Key) there.
Where do I generate the API Token? +
In your GrowingIO portal, navigate to [Project Configuration] -> [API Token Management] to generate a long-term API Token for this server.
Can I query quantitative metrics through the agent? +
Yes. Use the get_metrics tool with a JSON query definition. This allows your agent to retrieve specific data points like DAU, retention rates, or custom conversion metrics.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
No2Bounce
Validate email addresses in bulk to reduce bounce rates and protect your sender reputation directly from your AI agent.
OmniEngage
Reach customers across email, SMS, push, and in-app with unified omnichannel marketing that personalizes every touchpoint.
Appfigures
Analyze app store performance with Appfigures — track downloads, revenue, reviews, and rankings via AI.
You might also like
USPTO API
Search US patents and trademarks — audit intellectual property via AI.
Knorish
Launch your online school and sell courses with a platform that bundles LMS, website builder, and payment processing together.
Authing
Cloud-native identity and access management platform — manage users, roles, and security logs via AI.