> ## Documentation Index
> Fetch the complete documentation index at: https://docs.wisepim.com/llms.txt
> Use this file to discover all available pages before exploring further.
# AI Categorization
> Automatically assign products to your existing category tree with AI, with a confidence threshold you control, so nothing lands in the wrong place.
AI Categorization reads each product's content and assigns it to the right categories in **your existing category tree**. It does not invent categories, it places products into the structure you already have, and it only commits an assignment when its confidence clears the threshold you set.
**What it enables:** categorize a freshly imported catalog in minutes instead of days, keep categories consistent as you add products, and fix the uncategorized "Other" bucket that quietly hurts navigation and merchandising.
AI Categorization is **promptless**, it needs no prompt from your library. It costs **1 credit per product**. It assigns your existing categories and never creates new ones.
## When to use it
* **After an import** where products arrive with no category, or a flat supplier category that doesn't match yours.
* **To clear the "Uncategorized" bucket** that builds up over time.
* **When category assignment is inconsistent**, similar products scattered across different branches.
* **As an [Automation](/en/automations/overview) step**, so new products are categorized the moment they're created.
## How to run it
Check products individually, or filter (for example, "uncategorized") and select all matching results.
Click **Enrich with AI** in the toolbar.
Pick **AI Categorization** as the enrichment type.
Adjust the confidence threshold and how many categories each product can receive (see below).
Click **Start Enrichment** and track it in the [Process Tracker](/en/other-features/process-tracker).
## Configuration options
### Confidence threshold
The AI scores how sure it is about each category match. It only assigns a category when that score clears your threshold.
* **Default: 75%.** A balanced setting that assigns confidently in most catalogs.
* **Range: 0–100%, in 5% steps.** Raise it for accuracy, lower it for coverage.
* **Higher threshold** = fewer assignments, but the ones made are more reliable. Products it isn't sure about are left for you to place manually.
* **Lower threshold** = broader coverage, but review the edge cases.
Start at the **default 75%** for a first pass. Then lower the threshold to 60–65% and run again on whatever is still uncategorized, you get the safe, confident assignments first, and only review the genuinely ambiguous products.
### Category count per product
* **Auto-assign (recommended):** the AI decides how many categories fit each product, some belong in one, some in several.
* **Manual maximum:** turn auto-assign off and set a cap of **1–10** categories per product. Use this when your merchandising rules limit how many categories a product may appear in.
## What good looks like
* **The uncategorized count drops sharply** after the first run at the default threshold.
* **Assignments match your merchandising logic**, products appear where shoppers would look for them.
* **A small, reviewable remainder** stays uncategorized, the genuinely ambiguous products, which you place by hand or by lowering the threshold.
## Act on what you find
Your threshold may be too high for this catalog, or your category names may be ambiguous. Lower the threshold to 60–65% and re-run on the remainder, and consider clarifying vague category names. **Outcome:** higher coverage without sacrificing the confident first-pass assignments.
A misplacement usually traces back to thin product content, the AI had little to go on. Enrich the [title](/en/ai/titles), [description](/en/ai/descriptions), and [attributes](/en/ai/attributes) first, then re-categorize. **Outcome:** accurate placement driven by richer product data.
Turn off auto-assign and set a manual maximum (for example, 2–3). **Outcome:** category counts that match your merchandising rules.
Add AI Categorization as a step in an [Automation](/en/automations/overview) triggered on product creation or import. **Outcome:** new products are categorized the moment they enter the catalog, with no manual run.
## Measure the impact
Better categorization improves navigation and the relevance of category-level merchandising. Review structure and category health in [Category Insights](/en/analytics/category-insights), and watch how category pages perform in [Search & Discovery](/en/analytics/search-discovery).
## Related
Analyze category structure, health, and revenue.
Extract the specs that help the AI categorize accurately.
Categorize new products automatically on import.
Overview of all 14 enrichment types.