Does the AI change anything in Shopify?

No. This workflow is read-only: the AI reads products, stock levels and orders, and produces analysis. You place the orders.

How accurate are the run-out predictions?

They are based on your recent sales velocity, so they are as good as your sales history. Treat the first runs as a starting point and tune the lookback window to match your seasonality.

AI Use Cases

Quick winBusinessSaves 1 day per reorder cycle

Shopify stock forecasting

Ask AI to watch your Shopify stock: a live low-stock view plus sales trend analysis that predicts when to reorder each product.

The problem

Stock checks in Shopify mean exporting reports, stitching spreadsheets together and eyeballing sales trends product by product. By the time the reorder list is ready, it is already out of date, and slow sellers quietly tie up cash while bestsellers run out.

The workflow

1
Connect Shopify
Add the Shopify connector to your AI platform so it can read products, inventory levels and order history.
2
Build the low-stock view
Ask for every product below its reorder point, grouped by supplier, with current stock and days of cover remaining.
3
Analyse sales velocity
Have the AI compare recent sales rates against stock on hand to predict the run-out date for each product.
4
Get the reorder list
Ask for a reorder list ordered by urgency, with suggested quantities based on lead time and sales trend.
5
Repeat on a schedule
Save the prompt and run it weekly, or each morning in busy season, and compare against the last run.

Start here

Once your connectors are set up, this prompt kicks the whole workflow off. Adjust the details to match your business.

Paste this into your AI tool

Using my Shopify store data, list every product likely to run out of stock in the next 21 days based on the last 60 days of sales. For each one show: current stock, average daily sales, predicted run-out date and a suggested reorder quantity assuming a <your supplier lead time> day lead time. Order by urgency and group by supplier.

The outcome

A reorder list with predicted run-out dates in minutes, instead of a day of exports and spreadsheet work.

Time saved: 1 day per reorder cycle

UK fashion retailer: saved the stock team a day on the first run. (Anonymised; client approval pending.)

What it needs

The connectors this workflow uses, and how it runs on each AI platform.

Connectors

Shopify

Works on

Last verified 18 October 2026

Copilot (free)	Not possible	The consumer app cannot add custom MCP servers.
Copilot Pro	Not possible	Drafted, to verify.
Microsoft 365 Copilot	Build required	Via declarative agents; MCP tool calls consume Copilot Credits.
Copilot Studio agents	Build required	Needs a Copilot Studio licence.
Claude Free	Works natively	Limited to 1 custom connector.
Claude Pro/Max	Works natively
Claude Team/Enterprise	Admin setup needed	Owners add custom connectors.
Claude Code	Works natively
Gemini Enterprise	Admin setup needed	Custom MCP blocked by default org policy; not on Business edition.

Who it's for

Retail & eCommerce Operations

The reusable skill

The reusable recipe: give AI read access to inventory and order history, define what "low stock" means for your business (reorder point, lead time, days of cover), then ask for exceptions rather than reports. The same pattern works for any system that holds stock levels and a sales ledger, not just Shopify.