As of 13 August 2026, AI can write Google Shopping product descriptions.

What to actually do

1. Hand it to a person

   The route this page recommends

   A person who owns the outcome does this end to end, worth it when the failure is dear.

2. Use a tool built for this

   Second choice

3. Do it yourself

   The distant third

   A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

   How to actually do it

   1. Open your product catalogue or feed export and gather each product ID, name, category, specifications, materials, dimensions, approved claims, variant details and intended use.
   2. Open your brand guidance and current Google Merchant Centre feed rules, then note the required character limit, prohibited claims, spelling preferences and tone.
   3. Paste the catalogue data, brand guidance and rules into the prompt, replacing the bracketed fields with your actual requirements.
   4. Ask the model to produce the table and separate factual claims from missing information instead of filling gaps.
   5. Compare every drafted description and claim against the current product catalogue, technical documents, packaging and approved claims list.
   6. Remove any unsupported wording, resolve every NEEDS MORE DATA item, and ask the model to revise only the affected descriptions.
   7. Export the checked descriptions into your feed, run your normal Merchant Centre diagnostics, and publish only after resolving feed errors or policy warnings.

   Prompt

   Write Google Shopping product descriptions from the catalogue data below.
   
   Use only facts stated in the data. Do not invent materials, dimensions, performance, certifications, delivery promises, discounts, stock information or customer benefits. Do not make medical, environmental or comparative claims unless they are explicitly supported by the supplied data. Keep the language clear, specific and suitable for a UK retailer. Avoid hype, keyword stuffing, repeated product names and calls to action. Preserve the product's intended use and important distinctions between variants.
   
   For each product, return a table with these columns: product ID, drafted description, factual claims used, missing information to confirm, and any wording that may need a policy or legal check. If a product lacks enough information for a reliable description, write "NEEDS MORE DATA" rather than guessing. Keep each description within [INSERT YOUR REQUIRED CHARACTER LIMIT] characters. Match this brand tone: [INSERT BRAND TONE]. Apply these additional catalogue rules: [INSERT BRAND OR CATEGORY RULES].
   
   Catalogue data:
   [PASTE PRODUCT DATA HERE]

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.