Enrichment

What is Enrichment used for?

Enrichment adds information from a supplemental file to existing Smart Products—for example warehouse stock, supplier attributes or additional content missing from the primary source. The workflow is shared by all data sources.

File enrichment is not AI enrichment. Nifelo uses the supplied values and does not generate product data on this page.

Matching products

An enrichment file does not create products. For every row, Nifelo finds an existing product in the same source by matching a Nifelo identifier—sku by default—to the corresponding file column. Remaining columns are merged when the values match. Unmatched rows are skipped.

Use a unique, stable product or variant identifier. Check capitalization, leading zeros, spaces and whether the file contains parent or variant SKUs.

Files and upload

Upload CSV, JSON or XML up to 20 MB. Extension and actual content must agree. Use one product or variant per row, consistent field names and at least one reliable identifier column.

After upload, open Map, select the Nifelo Smart Field, select its matching file column and decide whether to apply the file automatically after future imports. Start processing to see total rows and matched products.

A low match count can mean missing or differently formatted identifiers, duplicates, a wrong column, parent-versus-variant SKUs or products not imported yet. Verify samples on the Products page.

Fields and ownership

Use existing Smart Field names such as brand, material, color or stock. Unknown columns may be stored as supplemental data but are not automatically useful for validation or channels.

Merging happens per field:

• a customer value with local priority is protected;
• protected AI or knowledge values are not silently replaced;
• a filled rule-derived value remains, while an empty one may be supplemented;
• other source values can be supplemented or replaced by the enrichment value.

The filename is recorded as field origin and can be inspected under Products → Enrichment.

Automatic application

With Apply automatically, the stored file is processed after every later primary import using the saved identifier mapping. Nifelo uses the last uploaded file; it does not refresh it from an external URL. New products are enriched only if their identifiers occur in that file. Leave this off for one-time or quickly aging data.

Editing and deletion

Edit changes the identifier, column and automatic setting and can process the file again. Test carefully: an accidental matching value can enrich the wrong product.

Deleting removes the upload and stops future automatic use.

Deletion is not an undo operation. Values already written to Smart Products are not automatically restored.

Restore values through the authoritative source, a controlled replacement file or available field history, then import and validate as needed.

Recommended workflow

1. Import the current primary source.
2. Prepare a clean file with one stable identifier per row.
3. Use recognizable Smart Field names.
4. Upload and map both identifier columns.
5. Enable automatic use only for a maintained dataset.
6. Process and compare total and matched rows.
7. Review matched and unmatched samples and field origin.
8. Check Data Validation, Coverage & analysis and the target channel before publishing.