Polar Analytics MCP Server

Query unified Shopify, Meta Ads, Google Ads, Klaviyo, and 40+ ecommerce data sources through Polar's governed semantic layer with certified metrics like ROAS, LTV, CAC.

Analytics by Polar Analytics OAuth2 (Claude/ChatGPT directory) or Bearer Token (n8n, Make, API) active

Connect (Remote Server) Docs

Overview

Polar Analytics is an ecommerce business intelligence platform that unifies data from 40+ sources (Shopify, Meta Ads, Google Ads, TikTok, Klaviyo, Recharge, Stripe, Amazon, and more) into a governed semantic layer. The Polar MCP server exposes this semantic layer to AI agents like Claude, ChatGPT, and Cursor, letting them answer ecommerce questions using pre-defined, certified metrics instead of raw SQL.

The server is provider-hosted at https://api.polaranalytics.com/mcp and exposes a small set of high-level tools for generating reports, listing dashboards, fetching dimensions, and creating shareable deep links back into Polar. Because the AI calls governed metric operations (such as blended ROAS with Shapley-based attribution, cohort LTV, CAC, AOV, contribution margin, and MER) rather than writing ad hoc queries, the same question always returns the same calculation, which reduces hallucination risk.

It is notable because it pairs the MCP transport with a managed semantic layer specifically built for DTC and ecommerce operators. Setup is point-and-click through Claude's connector directory or via a Bearer API key for automation platforms like n8n and Make, with no warehouse provisioning required.

Tools

Tool	Description
`get_context`	Initializes a conversation and returns workspace metadata, available metrics, and dimensions.
`generate_report`	Generates a custom analytics report with rows, a totals row, and a deep link into Polar.
`list_dashboards`	Lists the dashboards available in the connected Polar workspace.
`get_dashboard_details`	Fetches the configuration and underlying data for a specific dashboard.
`list_dimensions`	Returns compatible breakdowns, filters, and dimension settings for a given set of metrics.
`get_view_details`	Retrieves details about a saved view (geographic region, custom segment, etc.).
`generate_report_link`	Creates a shareable Polar deep link for a report in table, bar, or pie mode.
`rate_report`	Submits user feedback (1-10) on the quality of a generated report.

Setup Guide

Prerequisites

A Polar Analytics workspace with at least one synced data source (initial sync can take up to 24 hours).
For Claude or ChatGPT: a Personal or Team plan that supports custom connectors.
For automation tools (n8n, Make, custom clients): a Polar MCP API key generated at https://app.polaranalytics.com/mcp.

Option 1: Claude Desktop / Claude.ai (OAuth)

Open Claude's connector Directory and find the Polar Analytics connector.
Click Connect (Personal) or Add to your team (Team plan).
In Claude, go to Customize > Connectors and press Connect.
Log in to your Polar workspace when prompted to complete OAuth.

Option 2: Generic MCP client (Bearer Token)

Generate an API key at https://app.polaranalytics.com/mcp, then configure your MCP client to use the streamable HTTP transport:

{
  "mcpServers": {
    "polar-analytics": {
      "url": "https://api.polaranalytics.com/mcp",
      "transport": "http",
      "headers": {
        "Authorization": "Bearer YOUR_POLAR_MCP_API_KEY"
      }
    }
  }
}

Option 3: n8n / Make

Server URL: https://api.polaranalytics.com/mcp
Transport: HTTP Streamable
Auth: Bearer Auth, paste your Polar MCP API key as the token.

Use Cases

Ask conversational questions about blended ROAS, CAC, LTV, AOV, MER, and contribution margin across Shopify, Meta Ads, Google Ads, TikTok, and Klaviyo without writing SQL.
Auto-generate weekly or monthly executive summaries of ecommerce performance with consistent, certified metric definitions.
Diagnose anomalies (for example "Why did CAC spike last week?") by breaking metrics down by channel, campaign, region, or product.
Summarize existing Polar dashboards and reports inside Claude or ChatGPT and share back a deep link for the team.
Power n8n or Make automations that pull governed metrics into Slack alerts, Notion docs, or client-facing reports.

Example Prompts

"Run an analysis of net sales by Top 5 Cities (using Billing Address), ranking by year from 2020 to 2025 full year."
"What is the ad spend for the UK region in August 2025?"
"Give me blended ROAS, CAC, and MER for last month and compare to the previous month."
"Summarize the 'Acquisition' dashboard and send me a deep link to the underlying report."
"Break down LTV by acquisition channel for customers acquired in Q1, cohort view."

Pros

Official, provider-hosted MCP server with OAuth installation directly from the Claude and ChatGPT connector directories.
Backed by a governed semantic layer, so the same question always returns the same certified metric (ROAS, LTV, CAC, MER, etc.).
Unifies 40+ ecommerce sources including Shopify, Meta Ads, Google Ads, TikTok, Klaviyo, Recharge, Stripe, and Amazon.
Works with Claude, ChatGPT, Cursor, n8n, and Make using either OAuth or a Bearer API key.

Limitations

Requires an active paid Polar Analytics workspace; not usable standalone.
Read-only analytics surface; no tools for writing data back to Shopify or ad platforms.
Initial workspace sync can take up to 24 hours before the MCP is useful.

Alternatives

Triple Whale Moby AI for similar DTC ecommerce analytics, though without a public MCP endpoint at the same maturity.
Shopify's official MCP server combined with a Meta Ads or Google Ads MCP for source-level access without a semantic layer.
Cube.dev MCP for teams that want to build their own governed semantic layer on top of a custom warehouse.