Selling on Multiple Marketplaces: How to Handle Different Taxonomies

The Problem: Every Marketplace Has Its Own Categories

If you sell on Google Shopping and Amazon, you already know this pain. Google uses the Google Product Taxonomy with 6,000+ categories. Amazon uses Browse Tree Guides (BTGs) that vary by marketplace and product type. Otto has Warengruppen. eBay has its own category IDs.

A single product — say, a Bluetooth speaker — might need to be categorized as:

• Google: Electronics > Audio > Audio Players & Recorders > MP3 Players & Portable Audio
• Amazon: Electronics > Portable Audio & Video > Portable Bluetooth Speakers
• Otto: Technik > Audio > Lautsprecher > Bluetooth-Lautsprecher
• eBay: Consumer Electronics > Portable Audio & Headphones > Portable Speakers

Same product, four different category paths, four different taxonomy structures.

Why Marketplaces Use Different Systems

Each marketplace built its category system around its own product catalog and customer behavior:

• Google designed their taxonomy for advertising — it's optimized for matching products to search queries across all product types.
• Amazon organizes by department and shopping behavior — their browse tree is designed for how customers navigate to products.
• Otto uses a German-market-focused system that reflects their traditional catalog structure and local shopping patterns.
• eBay has a flat category system with attributes, optimized for auction and fixed-price listings.

These aren't minor variations. The hierarchies are fundamentally different — different root categories, different levels of depth, different naming conventions, and different levels of specificity.

What This Means in Practice

For a multi-channel seller with 1,000 products, you potentially need:

• 1,000 Google category assignments
• 1,000 Amazon BTG mappings
• 1,000 Otto Warengruppen assignments
• 1,000 eBay category IDs

That's 4,000 category assignments to maintain. When a marketplace updates their taxonomy (Google does this several times a year), you need to review and potentially remap affected products.

Manual mapping at this scale is unsustainable. Most teams either:

1. Use overly broad categories (hurting discoverability)
2. Only categorize properly for their primary channel (leaving other channels underoptimized)
3. Spend excessive time on spreadsheet-based mapping that could go toward growing the business

Using Custom Taxonomies for Multi-Channel Categorization

The key to managing multiple taxonomies efficiently is to treat each marketplace's category system as a separate categorization job:

Step 1: Obtain the marketplace taxonomy. Download the official category list for each marketplace. Amazon provides BTGs through Seller Central. Otto provides Warengruppen to registered sellers. Google's taxonomy is publicly available.

Step 2: Convert to a standard format. Format each taxonomy as a plain text file with one category path per line, using > as the hierarchy separator:

Technik > Audio > Lautsprecher > Bluetooth-Lautsprecher
Technik > Audio > Kopfhörer > Over-Ear-Kopfhörer
Technik > Computer > Laptops

Step 3: Run separate matching jobs. Upload your product CSV once, but run it against each marketplace's taxonomy file separately. This gives you marketplace-specific category assignments for every product.

Step 4: Export and integrate. Download the results for each marketplace and add them to the corresponding product feed. Most feed management tools (Channable, DataFeedWatch, Productsup) accept category data as a CSV column.

Tips for Multi-Marketplace Categorization

Keep a master product list. Use one canonical CSV with your product IDs, titles, and descriptions. Run this same file against each taxonomy to ensure consistency.

Prioritize your biggest channels first. If 80% of your revenue comes from Google Shopping, get those categories right first. Then expand to secondary channels.

Re-run when taxonomies update. When Google publishes a new taxonomy version or Amazon restructures a browse tree, re-run your products against the updated taxonomy to catch any changes.

Check edge cases manually. AI categorization handles the bulk accurately, but products at the boundaries between categories (a "laptop desk" — is it a desk or a laptop accessory?) benefit from a quick human review.