Product Classification

Product classification is the process of defining what a product actually is, so the right data model applies to it. It answers the question: “What kind of thing is this?” — and that decision determines which attributes are required, which are optional, and how the product behaves across the catalog. Classification is upstream of almost every other product data decision.

How Classification Works

When a new product enters the catalog, it must be classified before anything else happens. A cable gets classified as a cable, which means the system knows it needs attributes like length, voltage, conductor material, insulation type, and colour. A valve gets classified differently, triggering a different set of required fields. A consumable gets classified differently from a capital item — which affects pricing rules, inventory behavior, and reorder logic.

Classification is the decision that drives everything downstream:

• Which attributes appear on the product page
• Which fields are required vs. optional in the PIM
• Which category it belongs to in the taxonomy
• Which schema template the product uses
• How the product behaves in search and filtering

Get classification right and the rest of the data model falls into place. Get it wrong and you end up with cables missing voltage ratings and valves without pressure specifications — which means broken filters, failed searches, and customer frustration.

Classification vs. Taxonomy

These two concepts are frequently confused. The difference matters:

• Classification is the “what” — defining what a product is so the right data model applies
• Taxonomy is the “where” — the hierarchy that determines where customers find it

A product can be correctly classified (you know it’s a cable, so the right attributes are required) but sitting in the wrong taxonomy location (filed under “General Components” instead of “Wiring & Connectivity”). Fixing taxonomy doesn’t fix classification — and vice versa.

In practice, classification often informs taxonomy placement, but they are separate decisions managed by separate rules.

Why Classification Is the Critical Upstream Decision

Poor classification is one of the most expensive product data problems in B2B eCommerce because it’s hard to find and hard to fix at scale. When a product is misclassified, the wrong attributes are attached to it. Customers searching for “12V 20A cables” won’t find it because the voltage and amperage fields are blank — they were never required because the product was tagged as a connector instead of a cable.

The cost compounds. Every downstream system — search, filtering, recommendations, AI-generated descriptions, channel feeds — inherits the classification error. Fixing it later means re-attributing potentially thousands of products.

In B2B, this matters more than in B2C because buyers expect to search by specification. They’re not browsing. They need a specific part that meets a technical requirement. If classification is wrong, the product might as well not exist in the catalog.