← All use cases

Use case B29 · Business

Retail operator turns supplier-invoice photos into Shopify product records with Claude

Ryan Hamilton ·

The short version

What they did

Hamilton connects Claude to Shopify and gives it invoice photos. His saved routine creates product records with descriptions, images, prices, costs, quantities and UPCs. He describes this as work for his small retail store; the post also mentions his separate work at Plus-Plus USA. The date comes from post metadata.

Result (as claimed)

“Was it perfect every time? No. Did it save me HOURS AND HOURS of tedious work? Absolutely.” He does not give an aggregate time log or error rate.

Prompt

I run a small retail store and want to turn supplier invoices into draft Shopify product data. The invoice photos or PDFs are [attach], our product import template is [attach], and our existing product list and approved pricing rules are [attach or say unavailable].

1. Extract one row per invoice line into the template, including SKU, UPC, title, cost and quantity where readable.
2. Add proposed retail prices only if my supplied rules or documents support them. Flag unreadable fields, possible duplicates and missing images.
3. Return a reviewable import table and state the invoice-line and output-row counts.

Use plain field labels. Preserve leading zeros in codes and leave uncertain digits blank with a note. Don’t invent product details, images or prices. Don’t create, publish or change store records until I approve the table.

Do it yourself

These steps are my suggested way in, not Ryan Hamilton’s setup.

Level: Beginner for building an import table in a chat. Letting the assistant write to your store directly is Intermediate.

What you need

  • A chat assistant that can read photos, such as Claude or ChatGPT.
  • Clear photos or PDFs of supplier invoices.
  • Your store’s product import template, exported from your store admin.

Start simple

  1. Take a sharp, flat photo of one invoice.
  2. Upload it with your import template and the prompt above.
  3. Check every row: cost, quantity and UPC against the invoice.
  4. Set retail prices yourself, then import the file as drafts.
  5. Review the drafts in your store before you publish them.

Take it further

  • Connect the assistant to your store, if your platform offers a connector, so it creates draft products for you.
  • Save your rules for titles, descriptions and pricing in a project so each invoice gets the same treatment.
  • Ask it to flag products you already carry, so you update stock instead of making duplicates.

Check your work

  • Compare the UPC and cost on every line with the invoice. Small digit errors are common.
  • Count the rows. The table should match the number of lines on the invoice.
  • Keep new products as drafts until you’ve checked them. Don’t give it permission to change live prices or delete products.

Source checked. The link, author, date and quote were confirmed against the source. The result itself has not been reproduced. Prompt adapted for this library, inspired by the case. Business size: Tiny retail store (headcount not stated). Results are quoted as the source claimed them and never improved. Spot an error or a broken link? Email charlie@cbarmorecpa.com.

More in Operations

New cases go out every Saturday in Inside AI.