Glammer Society

A size 4 in one brand may fit like a size 2 in another brand. Sizing inconsistencies remains a constant struggle for shoppers in search of the right fit.

On January 8, 2024, Amazon Fashion shared how they can standardize sizing through tools using large language models (LLM), generative AI, and machine learning. These tools include Fit Review Highlights and Fit Insights. Available online, shoppers can find their best fit and reduce returns. US apparel and shoe brands in the Amazon Brand Registry can use the tool for free that sold at least 100 units in the last 12 months.

Amazon noticed a glaring problem: lack of adequate fit leads to increased returns. Especially online, customers cannot try clothes on. This leads to people buying clothes in multiple sizes, to see which fits best, and returning the rest. This process is time-consuming and painful for buyers and causes sellers to amass costs. USPS Delivers points out, “For fashion apparel, physical stores experienced a return rate of about 3%, compared with about 25% for online sales. Highly fashionable items, for which the fit was highly important, saw an even higher return rate.” This data illustrates that online returns are eight times greater than returns made in physical stores.

For customers: Fit Review Highlights 

Amazon believes personalized and adaptive size recommendations can fix fit issues by considering the sizing relationships between brands, product reviews, and the fit preferences of the customer.

Amazon mentions, “We create a review highlight for each customer based on their recommended size using common themes across reviews. The goal is to make it easier for our diverse customer base to get personalized size guidance so they can determine how a style will fit on them.”

The algorithm anonymously clusters related customers and products with a similar fit. Then, the tool combines known product details with customer purchases and the sizes that were not returned by similar customers. Using AI and LLM, Amazon removes duplicate information and auto-corrects missing or incorrect measurements to improve accuracy.

For brands and sellers: Fit Insights Tool

The Fit Insights Tool provides sellers with customer feedback data to improve design, manufacturing, and communication of sizing to their buyers. For example, the tool can help brands inform customers if apparel “runs big” or “true to size.” LLM extracts and aggregates data from customer feedback relating to fit, style, and fabric. From the tool, insights include “return health,” which compares the product return rate against similar products with low returns, provides a size chart analysis, and a customer feedback summary.

Many large companies, such as Amazon, incorporate AI into their operations, services, and products. In the future, shopping will include AI to create a personalized and unique experience for buyers. The analytics derived from this technology will allow sellers to understand and cater to the needs of their customers.

Featured Photo: Edward Berthelot/Getty Images