Categorization

Map product types to the correct Google Product Category and check which product fields that category requires.

What is categorization used for?

Categorization maps product types from a data source to the official Google Product Category taxonomy. Nifelo uses this mapping to understand the product group and which additional fields are important or required. One mapping applies to all products with the same source category or product_type.

This guide explains the general workflow. Also open Categorization under your source type and follow the source-specific guide where instructions differ.

Requirement: a useful product type

Categorization uses the Smart Field product_type. If it is unavailable, Nifelo cannot collect source categories. First verify under Smart Fields that the correct source field populates product_type.

Use meaningful values such as Apparel > Women > Dresses. Avoid brands, campaigns, or internal codes without product meaning. Correct structural categories in the source whenever possible.

Overview and history

Nifelo lists each unique active product_type, its selected or suggested Google category, the origin of the choice, confirmation status, AI confidence, and missing-field warnings.

Apply history can reuse previously confirmed knowledge. Review historical results when the meaning of a product type has changed.

AI categorization

AI can assess unmapped product types using product type, title, description, and an image as context. It proposes an official Google category and may show confidence or reasoning.

  • Always review the complete category path.
  • Confirm only when it applies to every product in the source group.
  • Choose manually or improve the source data when uncertain.

An AI suggestion is not final human approval. Confirmed choices may be reused later.

Credits and wallet usage

AI scans consume categorization credits. The screen shows monthly limits, usage, and available wallet credits. Automatic wallet permission is configured separately under AI management.

Apply history and review existing suggestions before starting another scan. Scan only categories that still need a result.

Choose a category manually

Search or browse the taxonomy, inspect the full path, and choose the most specific category that applies to the entire source group. If a group contains different product kinds, split it in the source first.

Confirm suggestions

Confirmation means you have reviewed the suggestion. Revoke confirmation if it is wrong, then select another category or improve the source and reassess it.

Required and recommended fields

Google categories can require fields such as brand, GTIN/EAN, color, size, gender, age group, material, or condition. A warning does not necessarily mean the category is wrong; product data may be incomplete. Complete it through Smart Fields, source data, or normalization.

Uncategorized products

Products without valid categorization may be excluded from processes that require category context, and sales channels may require or strongly recommend a Google category. Resolve Action required items before large-scale enrichment or exports.

Save or reset

Saving stores the mapping inside Nifelo and does not change the original category. Reset removes current mappings and may delete manual confirmations. Reset only when deliberately starting over.

Recommended first check

  1. Verify product_type under Smart Fields.
  2. Check that source categories are clear and specific.
  3. Apply history.
  4. Scan only unmapped types.
  5. Review confidence and full paths.
  6. Confirm correct categories and choose uncertain ones manually.
  7. Resolve missing-field warnings.
  8. Save and inspect products before enrichment, normalization, or export.