← All use cases

Use case B8 · Business

Gemstone jewelry founder connects Claude to Shopify and Search Console to audit and fix the store

Anosh Bin Suhail, founder, Orah Jewels ·

The short version

What they did

Uses Claude with connected tools to find and fix broken meta descriptions, diagnose a slow-loading article from session recordings, run an SEO audit, and publish long gemstone guides to Shopify.

Result (as claimed)

“What normally takes a developer a day of back-and-forth, I watched happen in about 40 minutes while I was having tea.” On the audit: “The audit found 18 articles with SEO issues.”

Prompt

I run a gemstone jewelry store on Shopify. Help me review our blog’s search listings. The store is [store], and the period is [last 90 days or dates]. Use [connected Shopify and Search Console data, or attached exports of page data, titles and descriptions].

1. Match each article’s search data to its page title and meta description, the short summary used in search listings.
2. Flag missing descriptions, titles over 60 characters and descriptions over 160 characters as review points, not automatic failures.
3. Return a table with article, search clicks and impressions, current text, issue and suggested replacement.

Write replacements in [brand tone] using only the article’s facts. Don’t invent gem properties, prices or ranking promises. Mark unmatched pages and unavailable data. Make no store changes until I approve.

Do it yourself

These steps are my suggested way in, not Anosh Bin Suhail’s setup.

Level: Advanced when the AI connects to your store and makes changes. Pasting exports into a chat is Beginner.

What you need

  • Admin access to your store and Google Search Console.
  • A chat assistant that can read spreadsheets, such as Claude or ChatGPT.
  • For the full version: a desktop AI app connected to your store and analytics through connectors (MCP).

Start simple

  1. Export 90 days of page data from Search Console.
  2. Export your blog posts’ titles and meta descriptions from your store, or copy them into a sheet.
  3. Upload both to a chat with the Prompt.
  4. Ask for a table of problems and a suggested fix for each.
  5. Make the fixes yourself in your store, a few at a time.

Take it further

  • Connect the AI app to your store and Search Console so it can pull the data itself.
  • Let it draft the changes, then approve each one before it publishes.
  • Re-run the audit monthly and compare clicks before and after.

Check your work

  • Spot-check five fixes in the live store and in Google’s results.
  • Give it read-only access first. Only allow changes after you trust it, and back up your content.
  • Read long guides it writes before they go live. Check gem facts, care advice and prices.

Claimed. Quoted from the source but not yet checked against it. Prompt adapted for this library, inspired by the case. Business size: 50 workers. Results are quoted as the source claimed them and never improved. Spot an error or a broken link? Email charlie@cbarmorecpa.com.

More in Marketing

New cases go out every Saturday in Inside AI.