Taxonomy for Proton PIM: Best Practices & Guide
A well-structured taxonomy is the backbone of any Product Information Management (PIM) system. It determines how your products are organized, searched for, and presented to customers.
What is Taxonomy?
In the context of PIM, Taxonomy refers to the hierarchical classification of your products. It’s the "tree" structure that organizes your catalog.
- Categories: The branches of the tree (e.g., Hand Tools > Hammers > Sledgehammers).
- Attributes: The data points that describe the products within those categories (e.g., Weight, Handle Material, Head Weight).
Core vs. Extended Attributes
Proton's PIM structure distinguishes between two types of data fields to help you manage complexity:
1. Core Attributes (Global)
These are fields that apply to every product in your catalog, regardless of category.
- Examples: Product Name, SKU, Description, Manufacturer, Price, Image URL.
- Why they matter: These ensure every item has a baseline level of data for identification and basic searching.
2. Extended Attributes (Category-Specific)
These are specific technical specifications that only apply to certain categories.
- Examples:
- For Light Bulbs: Lumens, Base Type, Wattage.
- For Safety Gloves: Material, ANSI Cut Level, Size.
- Why they matter: This is where the richness of product data lives. It allows customers to filter by specific needs (e.g., "Show me all ANSI Level 4 gloves").
How Proton Handles Taxonomy
Unlike rigid systems that force you into a pre-set structure, Proton is designed for flexibility and AI-driven efficiency.
- Bring Your Own Taxonomy: We can ingest your existing category structure. If you already have a taxonomy that works for your ERP or website, we map to it rather than replacing it.
- AI-Driven Enrichment: Proton's AI agents scan your product data to identify missing attributes. For example, if a product description says "12-inch steel handle," the AI can extract Handle Length: 12" and Handle Material: Steel into structured fields automatically.
- Flexible Mapping: You can map your internal PIM taxonomy to different output channels. If "Customer A" calls them "Power Tools" and "Customer B" calls them "Electric Tools," the PIM can manage these distinctions for different e-commerce or catalog exports.
Best Practices for Building Your Taxonomy
1. Design for the Customer, Not the ERP
Your ERP codes might make sense to your warehouse team, but your taxonomy should match how customers search.
- Bad: Dept 04 - Fasteners
- Good: Fasteners > Screws > Wood Screws
2. Standardize Your Units
Ensure consistency in how data is stored.
- Don't mix 12 in, 12", and 1 ft. Choose one standard (e.g., inches) for the attribute value so filters work correctly on your website.
3. Don't Over-Categorize
Too many layers make navigation difficult. Aim for a depth of 3-4 levels max (e.g., Category > Sub-Category > Product Type). If you need more detail, use Attributes as filters instead of creating a new sub-category.
4. Start with High-Impact Categories
You don't need to perfect your entire catalog on Day 1. Focus on the top 20% of categories that drive 80% of your revenue. Get the taxonomy and attributes right for those first.
Common Questions
Q: Can a product live in multiple categories? A: Yes, products can often be "mapped" to multiple nodes in a taxonomy tree depending on the end-channel (e.g., a "Safety Vest" might live in both "Safety Gear" and "Construction Apparel").
Q: What if I don't have a taxonomy? A: Proton can help you build a baseline taxonomy using industry standards or by clustering your existing products to see natural groupings.