Category Mapping allows AI Workflows to automatically assign products to categories using AI.

Instead of manually categorizing products, AI can analyze product content and determine the most relevant category structure automatically.

This is especially useful for:

Category Mapping analyzes product information such as:

and matches products to the most relevant categories inside your catalog structure.

Examples:

Incorrect or missing categories can negatively impact:

Manual categorization becomes difficult at scale, especially when importing large supplier catalogs.

Category Mapping helps automate this process.

Category Mapping is commonly used for:

A supplier imports products using inconsistent category names such as:

AI can automatically map these products into:

This creates a cleaner and more consistent catalog structure.

A Category Mapping action consists of:

Products are processed individually through the mapping pipeline.

Inside a workflow:

You will then enter the Category Mapping configuration screen.

The configuration screen allows you to define how products should be categorized.

Common settings include:

You can select:

Example:

Different models may affect:

AI determines the most relevant category based on product information.

Examples of analyzed data:

The AI then suggests the best matching category inside your catalog structure.

Prompts define how AI should categorize products.

Simple example:
"Assign the product to the most relevant webshop category."

Advanced prompts may include:

Example:
"Assign products to the most relevant ecommerce category based on the product description and attributes. Prefer the deepest matching category when possible."

Category Mapping often performs better after Attribute Extraction.

Example:

This creates stronger category consistency.

The scope determines which products the action processes.

This is often controlled through:

Focused scopes improve:

Before running large category mapping actions, you can test categorization on selected products.

Testing helps validate:

Testing is strongly recommended before large scale execution.

Test results display:

You can:

Reasoning explains why the AI selected a specific category.

Example:
"The product was categorized as Dry Cat Food because the description references adult cat kibble and dry nutrition."

Reasoning improves transparency and moderation control.

Confidence scores indicate how certain the AI is about the suggested category.

Higher confidence often means:

Lower confidence may indicate:

After testing:

Products are then processed individually through the mapping pipeline.

Results become visible inside the Results tab.

The Results tab displays:

This provides visibility into mapping execution.

Depending on the workflow configuration:

Once approved, mapped categories synchronize back into the product data.

AI performs better with:

Messy category trees reduce mapping quality.

Extracted attributes improve categorization significantly.

Examples:

can all improve mapping accuracy.

Different suppliers often structure products differently.

Creating supplier specific workflows improves consistency and moderation quality.

Low confidence results may require:

Confidence scoring helps prioritize moderation.

Example:
A webshop imports supplier products without usable categories.

Workflow:

This creates a fully automated product onboarding pipeline.