Video tutorial
Overview
The Catalog Audit is your operational command center for catalog hygiene. While the Benchmark tool diagnoses your catalog's health, the Catalog Audit provides the tools to fix it. It allows you to filter your entire inventory by data quality and uses AI to automatically repair missing attributes, ensuring your products are discoverable by AI agents.
The Quality Breakdown
Upon entering the Catalog Audit, you are greeted with the Quality Breakdown bar. This is an interactive filter that segments your products based on their metadata completeness.
Green (Excellent): Products that are fully optimized.
Yellow/Red (Mediocre/Poor): Products missing key attributes.
Blue (Uncategorized): Products missing the "Taxonomy" category.
Prioritization Strategy
The most critical segment to address is Uncategorized. Products without a taxonomy category are virtually invisible to AI agents because the models cannot classify what the item actually is.
Filtering and Search
You can drill down into specific segments of your catalog to perform batch work.
Quality Filters: Click directly on the colored badges (e.g., clicking the "Uncategorized" badge) to instantly filter the list below to show only those items.
Status Filters: By default, the list shows Published (Active & Live) products. You can change this to view Draft or Archived products if you are working on pre-launch inventory.
Search: Use the search bar to find specific products by name or handle if you know exactly what you want to fix.
The "Fix Product" Workflow
This is the core feature of the Audit tool. It allows you to use AI to fill in missing data gaps without manual data entry.
Identify an Issue: Locate a product in the list with a low score (e.g., "Slides" with the issue No category assigned).
Launch Editor: Click the Fix Product button on the right side of the product row.
Review AI Suggestions:
The tool opens a detailed editor showing current product data alongside AI Suggestions.
Category Fixing: If the category is missing, the AI analyzes the product (e.g., title, description, image) and suggests the correct Shopify Taxonomy (e.g., "Slippers in Shoes").
Metafield Enrichment: The AI also suggests granular attributes such as Color, Size, Material (e.g., Rubber/Synthetic), and Style (e.g., Slide).
Applying Changes
You have two ways to apply the AI's recommendations:
Granular Selection: Click on individual suggestions to approve them one by one. This is useful if you want to verify specific attributes like "Target Gender" or "Shoe Width."
Accept All: If the AI's analysis looks accurate, click the Accept All button to instantly populate every missing field with the suggested data.
Final Step: Once the fields are populated, click Save in the top right corner. The product status will update, and it will move out of the "Uncategorized" or "Poor" segments in your dashboard.