Shopify MCP Server
Connect Claude, ChatGPT, or any AI tool to your Shopify store. Analyze orders, revenue, and product performance — and attribute sales back to Meta, Google, and TikTok ads — all with natural language.
What is the Shopify MCP?
The Pipeboard Shopify MCP (Model Context Protocol) Server connects AI assistants like Claude, ChatGPT, and custom agents to your Shopify store data. Pull orders, daily revenue, and product performance — then attribute every sale back to the ad platform that drove it — all through natural language.
Instead of switching between Shopify Admin and your ad dashboards, you can ask an AI to reconcile store revenue against Meta, Google, and TikTok spend and compute true ROAS in a single conversation.
Key features:
Getting Started
Set up in under 2 minutes
1. Sign up for a Pipeboard account:
pipeboard.co/auth/signup →2. Connect your Shopify store:
pipeboard.co/connections →3. Add the MCP server to Claude (or another MCP client):
https://shopify.mcp.pipeboard.co/Use this URL as a Streamable HTTP MCP server in your client. Authentication is handled automatically through your Pipeboard session.
Connect to Your AI Tool
The Shopify MCP works with any MCP-compatible client. Just add the server URL and start analyzing your store with natural language.
Claude Pro / Max
Recommended- 1.Open claude.ai/customize/connectors
- 2.Click "Add custom connector"
- 3.URL:
https://shopify.mcp.pipeboard.co/
Claude Code
CLI / VS CodeRun this command in your terminal:
claude mcp add --transport http pipeboard-shopify https://shopify.mcp.pipeboard.co/Then type /mcp in Claude Code to authenticate with Pipeboard.
Alternative: use an API token for instant setup (no browser auth needed):
claude mcp add --transport http pipeboard-shopify https://shopify.mcp.pipeboard.co/?token=pk_...Get your token at pipeboard.co/api-tokens
Claude Desktop JSON configuration:
Add this to your Claude Desktop MCP settings (Settings → MCP Servers → Add):
{
"mcpServers": {
"shopify": {
"url": "https://shopify.mcp.pipeboard.co/"
}
}
}More integrations:
Tips for getting started
Try these prompts
Revenue Overview
"Show my Shopify revenue summary for the last 7 days, broken down by day with conversion rate"
Ad Attribution
"Attribute my Shopify orders to Meta, Google, and TikTok for the last 30 days and tell me which platform drove the most revenue"
Product Performance
"Which products drove the most direct and assisted sales last month? Show me the top cross-sells"
True ROAS
"Compare my Meta ad spend against Shopify revenue attributed to Meta last week and calculate true ROAS"
Available Tools
The Shopify MCP provides read-only analytics tools organized by category. All tools are available through natural language when connected to an AI assistant.
Shop & Account
list_shopify_shopsList connected stores with domain, currency, timezone, and scope
get_shopify_shop_infoGet store details: name, domain, currency, plan, and country
Orders
get_shopify_ordersFetch orders by date range with line items, UTM data, and summary stats
Ad Attribution
get_shopify_orders_by_attributionAttribute orders to Meta, Google, and TikTok via customer journey (last-paid-click)
Product Performance
get_shopify_product_performanceProduct-level direct and assisted (cross-sell) sales by landing page attribution
Revenue & Analytics
get_shopify_revenue_summaryDaily Total Sales breakdown with sessions and conversion rate, reconciles with Admin
Cross-Platform Attribution & True ROAS
The standout capability of the Shopify MCP is tying store revenue back to the ad platform that drove each sale. It reads the same session data Shopify Admin shows under “Session details” (first visit, last visit, UTM parameters, referrer) and applies a last-paid-click rule within a configurable window.
Last-touchPrimary rule uses the last visit before purchase within the attribution window.
First-touchSee cross-platform journeys — e.g. Meta introduced, Google closed.
Configurable windowDefaults to 7 days; set 30 days to match Google click attribution.
Combine it with the Meta, Google, or TikTok Ads MCP to compute True ROAS = attributed Shopify revenue / ad spend per platform.
Best Practices
- Use list_shopify_shops first to grab the exact shop_domain other tools need
- Reconcile get_shopify_revenue_summary against Shopify Admin Total Sales line-by-line
- Set source_type="web" for a ROAS denominator (excludes POS and draft orders)
- Match the attribution_window_days to each platform — 7d for Meta, 30d for Google
- Check the orders_with_journey_data count to gauge how reliable attribution is
- Grant Protected Customer Data Access so customer_journey_summary populates fully
Frequently Asked Questions
Can the Shopify MCP edit my store or change orders?
No. The Pipeboard Shopify MCP is read-only analytics. It fetches orders, revenue, product performance, and attribution data — it never creates, updates, or cancels orders, products, or customers.
How does it attribute orders to Meta, Google, or TikTok ads?
It reads Shopify’s customer_journey_summary (the same data Admin shows under Session details) and applies a last-paid-click rule within a configurable window. Each order also returns first-touch and last-touch platforms so you can see the full journey.
Why are some orders missing attribution data?
Attribution relies on customer_journey_summary, which requires Protected Customer Data Access in the Shopify Partners Dashboard. When it is not populated, the MCP falls back to landing_site and referring_site — check the attribution_source field and the orders_with_journey_data count to see coverage.
Does the revenue summary match Shopify Admin?
Yes. get_shopify_revenue_summary returns the same fields as Admin’s Total Sales breakdown (gross sales, discounts, returns, shipping, taxes) so it reconciles line-by-line. Note that test orders are included since they live in the same Orders API — subtract them if you need an exact Admin Analytics match.
Which AI tools can I use it with?
Any MCP-compatible client — Claude Pro/Max, Claude Desktop, Claude Code, ChatGPT (Developer mode), Python with LangChain, n8n, and more. Just add the server URL above.
Support & Documentation
Need help?
