Skip to main content

Shopify MCP Server

Connect Shopify and ask your AI tool why net sales trail gross sales, which variants get sent back most, and whether first-time buyers come back for a second order. It reads orders, line items, refunds, discount codes and first and last visit UTMs, so you can skip exporting another orders CSV.

  • No credit card required
  • Set up in 2 minutes
  • Read-only access
Shopify logoTwo Minute Reports logo
Works with
  • ChatGPT
  • Adidas logo
  • Accenture logo
  • Cloudflare logo
  • Canva logo
  • Paytm logo
  • Doodle logo
  • Proton logo
  • SAP logo
  • UEFA logo
  • Virgin Australia logo
  • AppSumo logo
  • Walmart logo
  • Nestlé logo
  • Salesforce logo
  • Spotify logo
  • Airbnb logo
  • Samsung logo
  • Ford logo
  • Vodafone logo
  • Whirlpool logo
  • Flipkart logo
  • boAt logo
MCP SETUP

Set Up the Two Minute Reports Shopify MCP in 3 Simple Steps

Add Two Minute Reports to your AI tool once, then ask about your Shopify whenever you need an answer.

Create Your Free Account

Sign up for Two Minute Reports and connect your Shopify account. Every account you manage is added for you.
Connect free

Add Two Minute Reports to Your AI Tool

Find Two Minute Reports in your AI tool's connector list and click Connect. Or add it as a Custom MCP Server with this URL:

Start Asking Questions

Type your question the way you would ask a teammate. Your AI tool checks your Shopify data and replies in seconds.
See example questions
Use cases

What the Shopify MCP Finds When Net Sales Lag Behind Gross Sales

One prompt that covers refunds, discount codes, repeat customers, UTM sources and low stock on your best-margin SKUs, with a reply you could forward to the store owner.

Shopify

Compare October with September in my Shopify store. Which variants lost most to refunds, did discount codes cut AOV, did returning customers order less, which UTM sources brought first orders, and which high-margin SKUs are nearly gone?

AreaWhat I foundAction
RefundsTwo sweater variants make up 41% of the $8,900 refundedAdd size guide
Discount codesWELCOME15 was on 31% of orders, with AOV $18 lowerCap the code
Repeat buyersReturning customers placed 38% of orders, down from 44%Plan win-back email
AttributionFirst visit credits Instagram with 590 orders, last visit with 412Compare UTM models
InventoryThree SKUs at a 52% gross margin have under 20 units leftReorder this week

Store owner summary: Gross sales grew 9%, but refunds on two sweater variants and a heavily used WELCOME15 code held net sales growth to 3%. Start with the size guide and a cap on the code before the holiday sale.

Illustrative example

Two Minute Reports Tracks Every Shopify Order From First Visit to Refund

The connector reads 161 Shopify fields, from order and line item details to payments, inventory and the UTMs on each customer's first and last visit.

Sales and margin

Total sales, gross sales, discount amount, refund amount, net sales, shipping, tax, tips, duties, gross profit, gross margin and cost of goods sold.

Orders and fulfillment

Orders, average order value, ordered, returned and net quantity, order tags, payment status, fulfillment status, discount code, sales channel, POS location name and cancelled orders.

Products and variants

Product title, variant title, variant SKU, product type, vendor, category, tags, status, price, comparison price and cost per item, plus line item name, price and SKU.

Customers

Customer count, new customers, returning customers, customer order count, customer average order amount, customer tags, account age in days, last order date, country and city.

First and last visit

Source, source type, referrer URL, landing page URL, referral code and UTM source, medium, campaign, content and term, for both the first visit and the visit that converted.

Payments and inventory

Gross, refunded and net payments, transactions, transaction kind, payment method, accelerated checkout, inventory quantity and inventory value.

See all Shopify metrics and dimensions

Shopify's Developer and Shopping Servers Versus a Shopify MCP Built for Store Reporting

Shopify's own MCP servers serve developers and shoppers: Dev MCP answers questions about its APIs and docs, while the UCP and Customer Accounts servers handle catalog search, carts and a signed-in shopper's own orders. Community servers such as shopify-mcp on npm call the Admin API with your app credentials and can change products and orders.

FeatureShopify's MCP serversOpen-source serversTwo Minute Reports
Who maintains itShopify (Dev MCP is open source under MIT, part of the Shopify AI Toolkit)Independent developersTwo Minute Reports, hosted for you
What it is built forDev MCP: docs, GraphQL schema and code checks. UCP and Customer Accounts: catalog, carts and a shopper's own ordersReading and editing products, orders, customers and inventory through the Admin APIStore reporting on sales, orders, customers and attribution
Sales reports for the merchantNot includedRaw order records your AI has to total itselfNet sales, AOV, gross margin and new vs returning customers as ready metrics
What you needDev MCP runs locally with npx; Customer Accounts needs OAuth with PKCE and a custom domainAdmin API credentials from your own app and a place to run itOAuth sign-in and one URL
Access to your storeDev MCP does not touch store dataOften read and write, depending on the scopes you grantRead-only on your Shopify store
Other data sourcesShopify onlyUsually Shopify onlyShopify plus 30+ other sources, with all client stores in one workspace
Dashboards and client reportsNot includedNot includedDashboards and scheduled white-label reports, included in every TMR plan

Your Data Is Safe

Security you can explain to your clients, too.

We don't store your advertising data on our servers. Two Minute Reports connects to your data sources using OAuth - your credentials never touch our system. Your AI only reads what you've explicitly connected.

OAuth Authentication

Connects securely to your data sources. No passwords stored. Revoke access anytime from your platform settings.

Read-Only Access

Your AI can only read your data. It cannot create, edit, pause, or delete anything. Your account stays fully in your control.

No Data Training

Your advertising data is never used to train AI models. What's yours stays yours - full stop.

Frequently Asked Questions

Yes. Refund amount sits next to gross sales, discounts and net sales, and returned quantity sits next to ordered and net quantity. Group them by product title, variant title or variant SKU to see which sizes or colors come back most. For exact field names, check the full list of Shopify metrics and dimensions.

Ask for orders, average order value and discount amount grouped by discount code, then compare them with orders that used no code. A code that shows up on many orders but pulls AOV down may be worth capping or limiting to first orders. Order tags help you split influencer or wholesale codes from public ones.

Yes. The channel dimension separates orders by sales channel, and POS location name and location name show which shop handled each order. Ask for net sales and orders by channel and location to see how your retail stores compare with the online store over the same weeks.

Dev MCP helps developers search Shopify docs, explore the GraphQL schema and check code, without touching store data. The UCP and Customer Accounts servers let shopping assistants search a catalog, build carts and show a signed-in shopper their own orders. None returns merchant sales reports. To use Dev MCP next to ours, see Two Minute Reports in Claude Code.

Partly. Inventory quantity and inventory value show what is on hand, and ordered quantity by variant SKU and date shows how fast each one sells. Ask your AI to divide stock by average daily units sold over the last 30 days. Seasonal swings are not factored in, so treat the result as an estimate.

Yes. Each order carries first visit (FV) and last visit (LV) fields: source, source type, referrer URL, landing page URL, referral code and all five UTM tags. Ask for orders grouped by FV: UTM campaign, then by LV: UTM campaign. The gap shows which campaigns start journeys and which ones finish them.

Yes. Gross profit, gross margin and cost of goods sold are available, along with cost per item, product vendor and product type. Ask for gross margin by vendor for the last quarter. Cost per item comes from your product records, so variants with no cost entered will make margins look higher than they are.

Yes. New customers, returning customers, customer order count and customer average order amount are metrics, and Is Returning customer, account age in days and last order date help you build segments. Ask which first visit UTM source brings buyers who place a second order within 90 days.

Yes, if Facebook Ads is connected in the same workspace. Shopify holds no ad spend, so your AI takes spend from the Facebook Ads MCP and orders from Shopify where the last visit UTM source matches your Meta tags. To review the ad side in depth, run the Meta Ads audit Claude Skill.

Take back your time. Wow clients.

Get Two Minute Reports.

Start your free trial today

Built by marketers. Built for marketers.