Products presented in ecommerce catalogs are logically grouped together for convenient browsing. These product groups are commonly called product categories. The set of product categories for a given electronic catalog is called the taxonomy of the catalog. The process of assigning products to categories is known as category mapping.
When data from a merchant's ecommerce catalog is published to a comparison shopping engine ("CSE") the CSE will then need to map the merchant's products to the taxonomy of the comparison shopping site. Most first-tier shopping sites (such as Shopping.com, NexTag, Shopzilla and Pricegrabber.com) require (or highly recommend) that the merchant map their products to the taxonomy of the CSE. Obviously this can be a challenge for the retailer.
This post, the eighth in our series of nine CSE 'best practices,' will talk about CSE category mapping and ways to make the process efficient and the outcome optimal.
Map Categories not Products
When creating a data feed for a CSE shopping site the merchant will want to associate each of their ecommerce product categories to a product category on the CSE site. Many merchant categories may be assigned to same CSE category though generally it's not the other way around. By mapping categories (instead of mapping individual products) a great deal of efficiency is gained... as all the products assigned to a given merchant category will be automatically mapped to the related CSE category.
A good category mapping tool set, such as the one provided with Mercent Retail, will make creating and maintaining this category-to-category association a snap for even the largest catalogs. That being said, it's important to take some time to map your categories thoughtfully. Remember, category mapping is all about merchandising. Making sure your products are categorized correctly on the comparison shopping channels will increase your clicks and increase your conversion rate.
Map the SKUs that matter... individually
For those products that really drive the majority of your business it's wise to review the shopping channel category mapping for those products on a SKU-by-SKU basis (or UPC-by-UPC basis). If sales results are not what is expected on a given CSE channel then work with your account manager at the shopping site to find out which categories give the most traffic and sales per product listing... and which categories are the highest performing for your given set of UPC codes or manufacturer part numbers.
Be wary of 'unified' category mapping solutions
Some data feed marketing solution providers approach category mapping by associating each merchant's product categories ( and/or products) to their unique category structure and then using their category structure to map to each CSE taxonomy. This is not optimal. It's like making a photocopy of a photocopy... you lose clarity in the reproduction. It's an unnecessary layer of abstraction that mostly benefits the vendor not the merchant, CSE or consumer.
The merchant's unique ecommerce categories (and all sub-categories) should have their own voice and find their right place on each shopping portal. Attempting to unify all merchants to a single vendor taxonomy forces all merchants using these vendors to have the same CSE facing categories which makes the ability to differentiate your category mapping (and product content related to categories) virtually impossible.
Be wary of 'black box' category mapping solutions
Beyond the pit falls mentioned above of using a unified vendor taxonomy, some data feed marketing vendors will further complicate this approach by relying heavily on software to associate a given product to their taxonomy. This automated mapping process generally focuses on string comparisons of key fields and uses relevance-based product matching on part number and product content data elements.
This type of solution doesn't provide the merchant with control of the process, transparency into the results or the ability to easily test among various category mapping alternatives. It's a 'black box' that may provide value... but it also may wind up making really poor merchandising decisions on key merchant categories, on high value products, on the most important comparison shopping engines.
Comments