AI enrichment uses prompts from your Prompt Library and context from your Knowledge Library to generate content that matches your brand voice and product context.
Enrichment types
Titles
Generate compelling, SEO-friendly product names from basic product data.
Descriptions
Create detailed product descriptions that highlight features, benefits, and use cases.
Short Descriptions
Write concise summaries for product listings, category pages, and search results.
SEO Title & Meta Description
Generate optimized SEO titles and meta descriptions for search engine visibility.
Attributes
Extract and populate product specifications and technical details from existing content.
AI Categorization
Automatically assign products to the right categories based on their content.
Translation
Translate product content into 35+ languages simultaneously.
Web Research
Search the web to find and enrich product data using product identifiers.
How to enrich products
Select products
Check individual products, or use filters and select all matching results. You can also apply enrichment to your entire catalog.
Choose an enrichment type
Pick what you want to generate — titles, descriptions, short descriptions, SEO meta, attributes, AI categorization, translation, or web research.
Select a prompt (for content types)
For titles, descriptions, short descriptions, SEO meta, and attributes: browse and select a prompt from your Prompt Library. Use the search bar to find prompts by name or description. Your default prompt is pre-selected if you’ve set one.
Start enrichment
Click Start Enrichment. Processing runs in the background — monitor progress in the Process Tracker.
Content enrichment
Titles, descriptions, short descriptions, SEO meta, and attributes all follow the same flow: select products → pick a type → choose a prompt → start. The AI uses the prompt you select combined with:- The product’s existing data (name, attributes, category, images)
- Your Knowledge Library context (brand voice, target audience, industry knowledge)
- Content Logic variables if referenced in the prompt
What each type generates
| Type | Output | Stored in |
|---|---|---|
| Titles | Product name | Name field |
| Descriptions | Full product description | Description field |
| Short Descriptions | Concise summary (100-150 words) | Short Description field |
| SEO Title & Meta Description | SEO-optimized title and meta description | Meta Title, Meta Description fields |
| Attributes | Extracted specifications and features | Custom attribute values |
Prompt selection
Each enrichment type has its own set of prompts. Prompts are created and managed in the Prompt Library and fine-tuned in the Prompt Lab. When selecting a prompt, you’ll see:- Prompt title and description
- Tone of voice — Professional, Casual, SEO-focused, Sophisticated, and 10+ more options
- Purpose/style — Standard, Short, Detailed, Benefits-focused, Technical, Marketplace, Luxury, FAQ-style, and more
- Industry — 21 industry presets from Electronics to Fashion to Industrial
AI Categorization
AI Categorization automatically assigns products to categories based on their content. Instead of selecting a prompt, you configure a confidence threshold and category count.Configuration
| Option | Description | Default |
|---|---|---|
| Confidence threshold | Minimum AI confidence required to assign a category (0–1.0 slider). Higher = more accurate but fewer matches | 0.75 |
| Category count mode | Auto (AI determines optimal count) or Manual (fixed max, 1–10) | Auto |
Set the confidence threshold
Adjust the slider. A higher threshold means only high-confidence category assignments are made.
Choose category count
Leave on Auto for AI-determined categorization, or switch to Manual and set a max.
Translation
Translation converts product content into one or more target languages simultaneously. WISEPIM supports 93 languages including English, Dutch, German, French, Spanish, Italian, Portuguese, Polish, Scandinavian languages, Asian languages, and more. For full details, see Translating Products.Web Research
Web Research uses AI to search the web for product data based on identifiers like EAN, UPC, or MPN. It cross-references multiple sources for accuracy and structures the results into your catalog. You choose how products are identified for the search:| Identifier | Description |
|---|---|
| EAN / Barcode | European Article Number — globally unique, most reliable (recommended) |
| UPC | Universal Product Code — 12-digit US barcode |
| Model Number | Manufacturer model identifier |
| SKU | Your internal product code |
| Brand + Name | Brand and product name combination |
| Brand + MPN | Brand and manufacturer part number |
Single product enrichment
You can also enrich individual fields directly from the product detail page. Each enrichable field has an Enrich button that opens the same prompt selection flow, scoped to that specific field and product. Enrichable fields:- Product name (titles)
- Description
- Short description
- SEO title & meta description
- Product attributes
- Categories (AI categorization)
Reviewing enriched content
AI-generated content is written to the enriched/reviewed fields — the original content is preserved separately so you can always compare. After enrichment:- Review the output — Check the generated content for accuracy, brand consistency, and completeness
- Edit if needed — Adjust tone, add details, or correct any factual issues
- Mark as reviewed — Change the product’s review status to “Reviewed” using the status toggle, or bulk-update the status across multiple products
Using Content Logic with enrichment
Content Logic variables can be referenced in your prompts to give the AI dynamic, per-product context. This means the AI receives different instructions or context depending on each product’s brand, category, or attributes. Example prompt using variables:- Default variables — built-in product data
- Global variables — company-wide values
- Brand variables — per-brand content
- Category variables — per-category content
- Conditional blocks — if/else logic
Automating enrichment
All eight enrichment types are available as automation actions — trigger them on a schedule, after imports, when products are updated, or when quality scores change.Tips
- Start small — Test a prompt on 5-10 products before running it on your full catalog. Review the output and adjust the prompt if needed.
- Fill your Knowledge Library — The more context the AI has about your brand, audience, and products, the better the output. See Knowledge Library.
- Use specific prompts — A prompt for “B2B industrial fastener descriptions” will outperform a generic “write a description” prompt every time. Create prompts per category or product type in the Prompt Library.
- Combine with Prompt Lab — Use the Prompt Lab to test and iterate on prompts with real product data before running bulk enrichment.
- Monitor in Process Tracker — Large enrichment batches run in the background. Check progress in the Process Tracker.
- Layer with bulk editing — After enrichment, use Bulk Editing to append standard content blocks, compliance text, or ordering info using Content Logic variables.