Google Gemini
Answers, not guesses

Ask Gemini about your store.
Answered from your live data.

Gemini doesn't have your commerce data, so it guesses. Polar fixes that. Connect once and ask about your live Shopify, Meta, Google, TikTok, Klaviyo and Amazon data in plain English, right in the Gemini CLI. Every answer runs through Polar's governed semantic layer, so it matches your bank account, not a guess.

/
Setup Guide

Connect Gemini to your data, step by step.

Connect your data in Polar, add Polar as a connector in Gemini, then just ask. A few minutes, no code, no API keys to manage.
Polar Analytics

Polar Analytics account

Connect Shopify, Meta, Google, Klaviyo and more. The Polar MCP exposes clean, attributed live data, the source of truth Gemini reads from.

Connect your data →
Google Gemini

Google Gemini CLI

Polar connects to Gemini over MCP. Add it to the Gemini CLI and ask in plain English — Gemini reads your live, governed Polar data and answers.

Connect Polar to Gemini →
Polar MCP

One endpoint to add

Add Polar's MCP endpoint and sign in to Polar. No API key, no setup. Your live data lands in Gemini.

Get the Polar MCP →
1

Connect your data in Polar ~5 minutes

Build the semantic layer Gemini will read from.
How to
1
In Polar, connect Shopify, your ad platforms, Klaviyo, Amazon and GA4. Polar's semantic layer turns raw exports into governed metrics: blended ROAS, true CAC, contribution margin, defined once and consistent everywhere.
2
Make sure your data sources are connected, then wait for the first sync. The initial historical sync can take up to 24 hours for a new Polar account.
Why it matters: the answer is only as good as the data behind it. Polar's layer means the numbers Gemini returns in a Sheet or Doc match your bank account, not a cookie estimate or an LLM's guess.
Connect your data in Polar
2

Connect Polar to the Gemini CLI ~2 minutes

Add your Polar API key. No OAuth setup.
How to — in the Gemini CLI
1
Install and open the Gemini CLI. With Node 18+ installed, run npm install -g @google/gemini-cli, then launch it with gemini.
gemini · ~
$ gemini Gemini CLI v0.49.0 Authenticated with gemini-api-key /auth Tips for getting started: 1. Create GEMINI.md files to customize your interactions. 2. /help for more information 3. Ask coding questions, edit code or run commands 4. Be specific for the best results > Type your message or @path/to/file…
Step 1The Gemini CLI open and ready in your terminal.
2
Get your Polar API key. In Polar, open Settings → API keys and generate one. This is what the Gemini CLI uses to read your governed data — no OAuth setup required.
app.polaranalytics.com/settings/api-keys
API keys
Generate a key so the Gemini CLI can read your governed, attributed Polar data.
••••••••••••••••••••••••••4a2f Copy
+ Create API key
Step 2Generate a Polar API key in Settings, then copy it.
3
Add Polar to the Gemini CLI with your key. One command — the -s user scope keeps Polar available in every session, not just the current folder:
gemini mcp add -s user --transport http \
  --header "Authorization: Bearer <YOUR_POLAR_API_KEY>" \
  polar https://api.polaranalytics.com/mcp
Prefer config? Add this to your user settings.json (~/.gemini/settings.json) instead:
{
  "mcpServers": {
    "polar": {
      "httpUrl": "https://api.polaranalytics.com/mcp",
      "headers": { "Authorization": "Bearer <YOUR_POLAR_API_KEY>" }
    }
  }
}
gemini · ~
> gemini mcp add -s user --transport http --header "Authorization: Bearer ••••" polar https://api.polaranalytics.com/mcp ✓ Shell gemini mcp add -s user --transport http --header … polar MCP server "polar" added to user settings. (http) The polar MCP server was added successfully. Restart Gemini, then run /mcp.
Step 3Register Polar with your API key — MCP server "polar" added.
4
Verify, then ask. Run /mcp to see polar connected with its tools listed, then just prompt Gemini referencing your live Polar data.
gemini · ~
> /mcp Configured MCP servers: polar - Ready (30 tools, 8 prompts) Tools: - mcp_polar_get_metrics - mcp_polar_generate_report - mcp_polar_list_dashboards - mcp_polar_get_dashboard_details - mcp_polar_get_data_integrity_report - mcp_polar_send_file_to_slack …and 24 more Prompts: - Executive Summary - Profitability Deep-dive - Media Buying Health …and 5 more
Step 4polar connected and ready — now just ask Gemini using your live data.
Native MCP fit: Polar exposes one standard MCP endpoint — the same one already live in Claude's connector directory. In the Gemini CLI you authenticate with a Polar API key (Bearer token), the same key-based connection used in Perplexity, Lovable and Manus. No custom build.
Get your Polar API key
3

Ask anything, from your live data the payoff

Ask Gemini in the CLI. Polar powers every answer.
What this unlocks

A real answer, not a hallucination.

Ask Gemini in plain English, right in the CLI. Polar's connector pulls the live, attributed numbers from your semantic layer and Gemini reasons over them: blended ROAS, true CAC, contribution margin, LTV cohorts, all pre-computed. It builds the tables, reports and exports you ask for — and because it is Gemini, it can blend those governed numbers with live web context, your data plus the market, in one answer.

"Polar's semantic layer is way cleaner structure, way fewer hallucinations. The most delightful MCP experience." How RipnDip uses Polar's MCP vs the alternatives.

A few of the questions you can ask:

ASKBuild a table of true CAC by channel, last 90 days.
ASKWhy did blended ROAS drop last week?
ASKWrite this week's executive summary.
ASKHow does my ROAS compare to my category right now?
ASKWhich SKUs will stock out before their lead time?
ASKDraft the weekly KPI email for the leadership team.
See the prompt library
Get started

Only Polar gives Gemini clean, attributed data to answer from.

Connect your data in Polar, add the Polar MCP to the Gemini CLI, then just ask. A real answer in seconds, from your governed data.

Your Prompt Library

60 prompts Gemini can run on your live data

Filter by what you need or who it is for, copy the prompt, paste it into Gemini with Polar connected. Every answer comes from your governed, attributed data, right inside Google Sheets, Docs, Gmail and Looker Studio.
Type
Audience
60 of 60
Three Pieces, One Flow

You ask. Polar answers. Gemini explains.

Gemini is the interface you already use. Polar's semantic layer turns your question into a governed, accurate query your definitions of ROAS, CAC and margin, on clean attributed data.
01 · Ask

Gemini

In the Gemini CLI, in plain English

02 · Translate

Polar MCP

Your governed metrics

03 · Answer

Your live data

Clean, attributed, straight from your governed layer

The Foundation

Gemini didn't have your data. Now it does.

Out of the box, Gemini guesses, paste in a CSV and hope it reads it right. Polar's semantic layer feeds it your live, attributed numbers directly, so every answer is financial-grade: matches your bank account, never a hallucination. No SQL, no exports, no data team.
Connect Polar to Gemini
Get started · connect once · ask anything

Gemini didn't have your numbers. Now it does.

Connect Polar to the Gemini CLI, then just ask. Every answer runs through Polar's governed semantic layer: financial-grade, no SQL, no exports.

Phase 1 ~5 minConnect your data in PolarConnect Shopify and your stack. Polar's semantic layer turns it into governed metrics, then wait for the first sync.
Phase 2 ~2 minConnect Polar to the Gemini CLIAdd Polar's MCP to the Gemini CLI with your Polar API key. No setup.
Phase 3 the payoffAsk anything, from your live dataAsk in the Gemini CLI. Polar pulls the live, attributed numbers and Gemini answers.