Runs in your Google Sheet

Stop rebuilding spreadsheets.
Claude writes them for you.

Claude reads your live Shopify, Meta, Google and Klaviyo data and builds the report, model or forecast straight into a Google Sheet, formulas and all. Point it at Polar's semantic layer and every number, down to contribution margin, matches your bank account, not a guess. Build it once, schedule it, and it refreshes on its own.

/ ⏎
Connect Polar to Claude Book a demo
01 · What you need
Polar Analytics

Polar Analytics account

Connect Shopify, Meta, Google, Klaviyo, Amazon and more. The Polar MCP gives Claude one clean, attributed feed with contribution margin defined once, the source of truth your sheet is built on.

Connect your data →

Claude Cowork or Desktop

Claude reads your sources and writes straight into the sheet, formulas and all. Enable connectors, and use Claude Cowork if you want it to rerun on a schedule.

Get Claude →

A blank Google Sheet

Create an empty sheet and name it, for example Weekly Report. Claude builds the tabs, formulas and formatting into it. It stays a Google Sheet you fully own.

Create a sheet →
02 · Three phases to a self-updating sheet
1

Connect your data in Polar ~5 minutes

Build the semantic layer Claude will model from.
How to
1
In Polar, connect Shopify, your ad platforms, Klaviyo, Amazon and GA4. Polar's semantic layer turns raw feeds into governed metrics: blended ROAS, true CAC, contribution margin, defined once and consistent everywhere.
2
Confirm your 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: a model is only as honest as the numbers feeding it. Ask for "contribution margin" off raw feeds and Claude builds a beautiful shell, then guesses the inputs. Polar hands it real COGS, shipping, fees and attributed spend, so the margin line is one you'd put in front of your board.
Connect your data in Polar
2

Connect Claude to Polar and your sheet ~2 minutes

Enable the Polar MCP and the Google Sheets connector.
How to
1
In Claude, open Connectors and enable the Polar MCP, then sign in to Polar.
2
Enable the Google Sheets connector too, so Claude can write directly into your file.
3
Open your blank sheet (for example Weekly Report) and keep it handy. Claude reads Polar and writes into that sheet.
Why Polar: instead of juggling five or six raw connectors and joining them on every run, Claude reads one clean feed. Cheaper per run, and the numbers agree.
No Polar? You can wire up Shopify, Meta, Google and Klaviyo MCPs directly. It works, it just guesses at the baseline, and a confident number on a wrong baseline is how you overspend with data on your side.
Open Connectors in Claude
3

Prompt it once, then schedule it the payoff

Describe the report once. It writes itself every Monday.
What this unlocks

Build once. The sheet does the rest.

Describe the report, model or forecast in plain English. Claude reads the live, attributed numbers from Polar and writes the tabs, live formulas and a plain-English "what changed" summary straight into your sheet. Save it as a recurring Claude Cowork task and it reruns on its own, every Monday, before you're in.

"Using the Polar MCP and my "Weekly Report" sheet, build a report of net revenue, orders, new vs repeat, spend by channel, blended CAC and contribution margin, this week vs last with % change. Add a plain-English summary, flag moves over 10%, use live formulas."
"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 things you can build:

PROMPTMonday report: revenue, spend, blended CAC, new vs repeat, WoW.
PROMPTContribution margin by channel and product, live formulas.
PROMPTWhat happens to CAC and margin if I raise Meta spend 20%?
PROMPTWhere should we move budget next month? Write the call.
PROMPTDraft our board pack: actuals vs plan, variances explained.
PROMPTAdd a cohort tab: 30/60/90-day repeat rate by first product.
Get the full playbook
Get started

Build it once. Your Google Sheet updates itself.

Connect your data in Polar, point Claude at your sheet, and describe the report once. Then schedule it and stop rebuilding spreadsheets, start deciding with them.

Get started by booking a demo Connect your data in Polar ↗

What you can do

Ask in plain English and let AI analyze your live data — no formulas, no exports.

Summarize anything

Turn thousands of rows into a three-line summary of what matters.

Google Sheets

Find trends and outliers

Spot spikes and drops without building a single report.

Google Sheets

Clean and reformat data

Dedupe, split and standardize values with one request.

Google Sheets

Answer business questions

Which channel is growing fastest? Answered in seconds.

Google Sheets

Compare periods

Week-over-week or month-over-month, the math is done for you.

Google Sheets

Turn data into a report

Ask once and get a full written breakdown, ready to share.

Google Sheets
Google Sheets
+
+

One tool is a slice. Polar gives your AI the whole store.

A single source only holds its own data. Connect Polar and your AI reads your live Shopify, Meta, Google and Klaviyo data, answering questions no single tool could hold: blended CAC, contribution margin and LTV.

The foundation

What Polar adds under the hood

Not another connector. The reliable data layer that makes AI answers trustworthy.

Built on a semantic layer

Built on a semantic layer

Raw data becomes clean, reliable metrics like blended CAC, contribution margin, and LTV. Defined once and reused everywhere, so you are never charged per credit to recompute an answer.

Learn More →
Works with your whole stack

Works with your whole stack

Natively integrated with 45+ connectors like Shopify, Amazon, Meta, Google, and NetSuite. One click, no engineering, each brand on its own dedicated Snowflake database.

Learn More →
Connected to every AI you use

Connected to every AI you use

One MCP endpoint plugs your live data into ChatGPT, Claude, Manus, Perplexity, Lovable, Slack, and Notion, with all the trust your numbers deserve.

Learn More →
Your strategy, your rules

Your strategy, your rules

Reliable because it is yours. Every metric, threshold, and agent action is editable and stays up for human review. No black boxes, no runaway automations.

Learn More →

Do more with Polar + AI

Real questions ecommerce teams ask once their whole store is connected through Polar.

Where am I wasting spend this week?
$4,210
wasted
3 campaigns below target ROAS flagged

Paid media health check

Catch overspend across Meta and Google before the week closes.

What's my blended ROAS by channel?
3.4x
blended
Meta, Google and TikTok combined

Blended ROAS, live

True return per channel, ad cost and COGS included.

Which campaigns can I scale?
2
ready to scale
Prospecting 4.1x, Brand 6.7x ROAS

Scaling opportunities

Find the winners worth more budget today.

Net margin by product after ads and COGS?
42%
top SKU
Serum 42%, Kit 31%, Refill 9%

True net margin

Real profit per SKU, not just revenue.

Contribution margin trend?
+6 pts
up
Trending up over the last 90 days

Contribution margin

Track the metric that actually pays the bills.

Which SKUs sell below break-even?
2
below
Refill and sample kit losing money

Break-even check

Stop bleeding on products that lose money.

LTV by acquisition cohort?
$180
Mar cohort
Improving month over month

LTV by cohort

See which months bring your best customers.

What's my repeat purchase rate?
38%
+4 pts
vs the previous 90 days

Repeat rate

Track loyalty moving in the right direction.

Which customers are at churn risk?
601
at risk
Lapsed 60d+ and declining AOV

Churn risk

Win them back before they are gone.

Power up your business

See how ecommerce brands run their entire store from a chat: you asking, Polar answering.

▶

See it on your own data

Connect your store to Polar and ask your AI anything, across every channel, in plain English.

Book a demo

Frequently asked questions

How does this connector work with Polar?
Polar pulls this tool's live data into a governed semantic layer, so your AI answers questions across it in plain English — no exports, no formulas.
How do I connect it?
Connect the source in Polar in one click, then ask your AI through Polar's MCP. Setup takes minutes.
Where does Polar come in?
This tool is one source. Polar unifies all your store data so AI can answer across your entire business, not just one tool.
Do I need to know SQL or formulas?
No. You ask in natural language and the AI handles the analysis on Polar's clean data.