Content Enrichment allows AI Workflows to automatically generate, improve or optimize product content using AI.

Instead of manually rewriting product information, AI can create structured, optimized and scalable content directly inside your workflow pipeline.

Content Enrichment is commonly used for:

Content Enrichment analyzes existing product information and generates improved content based on your prompts and configuration.

The AI can use:

to create richer and more consistent product content.

Many product catalogs suffer from:

This can negatively impact:

Content Enrichment helps automate these improvements at scale.

Content Enrichment is commonly used for:

A supplier provides this description:

"Dry cat food with salmon for adult cats."

AI can enrich this into:

"Premium salmon dry cat food for adult cats, specially developed to support healthy digestion and daily nutrition. Ideal for cats with sensitive dietary needs."

This enriched content can then be used for:

A Content Enrichment action consists of:

Products are processed individually through the enrichment pipeline.

Inside a workflow:

You will then enter the Content Enrichment configuration screen.

The configuration screen allows you to define how AI should generate content.

Common settings include:

You can select:

Example:

Different models may affect:

You must define which fields AI should generate or overwrite.

Examples:

Only selected fields are enriched.

Prompts define how the AI should generate content.

Prompts can range from:

Example:
"Generate an SEO optimized shopping description using a professional tone of voice."

More advanced prompts may include:

Prompt quality directly affects output quality.

Content Enrichment can use data generated by previous workflow actions.

Example:

This allows workflows to create richer and more contextual content automatically.

The scope determines which products the enrichment action processes.

This is often controlled through:

Focused scopes improve:

Before running large enrichment actions, you can test generation on selected products.

Testing helps validate:

Testing is strongly recommended before large scale execution.

Test results display:

You can:

Reasoning explains why the AI generated specific content.

Example:
"The description emphasizes salmon based nutrition and adult cat dietary support because these themes were identified in the source data."

Reasoning improves transparency and moderation control.

Confidence scores indicate how certain the AI is about the generated output.

Higher confidence often means:

Lower confidence may indicate:

After testing:

Products are then processed individually through the enrichment pipeline.

Results become visible inside the Results tab.

The Results tab displays:

This gives visibility into enrichment execution.

Depending on the workflow configuration:

Once approved, enriched content synchronizes back into the product data.

Different categories require different writing styles.

Example:

Category specific prompts improve output quality significantly.

Weak example:
"Generate a product description."

Better example:
"Generate a Google Shopping optimized product description using a professional and informative tone focused on pet nutrition."

Specific prompts create stronger results.

Combining Attribute Extraction with Content Enrichment creates much richer content.

Example:

can all improve generated descriptions automatically.

Always validate:

before processing large catalogs.

Example:
A webshop imports cat food products with duplicate supplier content.

Workflow:

This creates a fully automated enrichment pipeline.