Current Challenges

When Rules Fail

Sometimes things slip through the cracks. Rule-based categorization can put products in a completely unrelated category. FeedAi™ helps protect your products from these mistakes.

Current Challenges

When Rules Fail

Sometimes things slip through the cracks. Rule-based categorization can put products in a completely unrelated category. FeedAi™ helps protect your products from these mistakes.

Real World Examples

Though ruled-based product categorization has a 91.8% accuracy, mistakes are common. Below are examples of real products that have been erroneously placed in a category. Our revolutionary FeedAi™ technology can significantly lower the number of these categorization failures.

Coated Jeans

Title = Women’s Mid-Rise Coated Skinny Jeans

Put into ‘Apparel & Accessories > Clothing > Outerwear > Coats & Jackets’ because it contains the word ‘coat’

Product vs. incorrect category

Taxonomy categorization error

Zebra Purse

Title = Zebra Striped Purse 

Put into ‘Apparel & Accessories > Clothing > Underwear & Socks > Bras’ because it contains the letters ‘bra’

Product vs. incorrect category

Taxonomy categorization using AI

Short Sleeve Shirt

Title = Men’s Short Sleeve Shirt in Midnight

Put into ‘Apparel & Accessories > Clothing > Shorts’ because it contains ‘short’

Product vs. incorrect category

eCommerce product categorization using AI

Dress Boot

Title = Jimmy Choo High Heel Dress Boot

Put into ‘Apparel & Accessories > Clothing > Dresses’ because it contains the word ‘dress’

Product vs. incorrect category

product categorization mistakes

Paper Printer

Title = Canon L100 Laser Printer, Plain Paper Print

Put into ‘Office Supplies > General Office Supplies > Paper Products > Printer & Copier Paper’ because it contains the word ‘paper’

Product vs. incorrect category

AI and eCommerce for accurate taxonomy

See FeedAi™ in Action