PPCDATAFEED

Help center · Rules and optimization

AI-Optimized Feed: rewrite the whole catalog for Google

A separate data source built from an existing one - titles, descriptions, attributes and categories improved by AI, with a quality score.

The AI-Optimized Feed is a data source type. It reads products from another data source, optimizes them and keeps the result as its own catalog - the original stays untouched. Part of the AI Optimizer module.

AI-Optimized Feed wizard, step 2 Field Mapping: Required fields mapped 8/8 - id, title, description, link, image_link, price, availability, brand; Recommended - additional_image_link, gtin → barcode; Available fields on the right
  1. Go to Data SourcesCreate New Data SourceAI-Optimized Feed.
  2. Pick the source data source and check the Field Mapping - required, recommended and optional Shopping fields are matched automatically; fix anything wrong.
  3. Switch on the Optimization options you want: Optimize titles, Optimize descriptions, Generate product highlights, Add missing attributes, Fill Google product category, Normalize attribute values, Translate feed content (with the target language).
  4. Click Generate preview - a sample of 10 products before and after, with the Feed Quality Score for both.
  5. Happy with the preview? Save. The full catalog is processed in the background - for large catalogs this takes hours the first time.

Good to know

  • Only products whose relevant content changed are optimized again on later updates; prices and stock are refreshed every time.
  • Google product category is chosen from a shortlist of Google’s official taxonomy - never invented.
  • The Feed Quality Score (0-100) measures required fields, recommended attributes and what the AI improved. It is deterministic, not an AI opinion.

Still stuck? Write to info@ppcdatafeed.com - we answer ourselves and can set it up with you.