Amazon Introduces AI Tool to Reduce Clothing Product Returns

[Ebrun Original] Recently, Amazon has used artificial intelligence to introduce four methods to improve the online shopping experience for clothing products. These methods include AI personalized size recommendations, "Fit Insights" tools for brand sellers, standardized size charts using AI, and highlights of consumer size reviews, in order to reduce clothing product returns and increase the conversion rate of clothing products.

For a long time, the return rate of online clothing has been consistently high, causing considerable return costs, and clothing category sellers have also suffered from high return rates. Amena Ali, CEO of supply chain technology company Optoro, stated that in the United States, the amount of online retail returns has reached 743 billion U.S. dollars, and the cost of processing a return for a $100 product is as high as $30 or 30%, with the majority of the cost being significant discounts and shipping expenses.

Incorrect sizing and fit are important reasons for consumers to return clothing purchased online. Data from market research firm Coresight Research shows that the average return rate for clothing ordered online is 24.4%, which is 8% higher than the overall online return rate. Among them, nearly one-fourth of online clothing purchases are returned mainly due to improper fit.

In order to enhance the online clothing shopping experience and address fit issues, various e-commerce platforms and brands have successively adopted a variety of technologies such as AR augmented reality and AI artificial intelligence to create virtual fitting rooms online, thereby creating a new shopping experience for clothing products online.

Earlier, Google announced the introduction of generative AI technology into online shopping tools, allowing consumers to try on clothing in a virtual environment and recommend specific products based on consumer preferences. Through this tool, consumers can quickly generate images of clothing worn by models with different body types, skin colors, and hairstyles, making it easier for them to understand whether the clothing is suitable for their own body shape. At the same time, consumers can also search for similar products of the same size but with different prices, colors, and patterns while browsing.

New York fashion brand Khaite and French luxury fashion brand Balmain have both used the Bods virtual fitting technology on their own e-commerce websites, Bods technology draws on gaming technology to digitally dress up consumers' virtual images. The US outdoor sportswear giant VF Corporation and DTC brand Reformation have used 3DLook technology on their e-commerce websites to provide consumers with highly accurate size and size recommendations.

Recently, Amazon has also introduced artificial intelligence and machine learning models into the online shopping experience for clothing products, launching four methods to address sizing and fit issues when purchasing fashion products online.

1. AI Personalized Size Recommendations

Amazon has developed deep learning algorithms using artificial intelligence to recommend sizes to consumers based on detailed information about clothing products, reducing the time consumers spend looking for the correct size.

This feature considers the relationship between brand size systems, product reviews, and consumer preferences for fit, using algorithms to learn from millions of detailed product information (such as styles, size charts, and customer reviews) and tens of billions of anonymous consumer purchases, and combines this information in real-time to provide consumers with the most suitable size recommendations.

At the same time, this algorithm can continue to learn and adapt to changes in consumer size preferences. For example, if a consumer buys a specific size of children's pants this month, the algorithm will consider that they may need a larger size in the next few months.

In addition, this feature uses artificial intelligence to help consumers discover the most suitable alternative styles according to their preferences. Amazon's AI recommendations extract product data such as style, color, price, size, return rate, and consumer reviews from the massive product catalog on the site, in order to recommend other popular styles when consumers shop.

Amazon stated that the size recommendation system analyzes millions of data points every day and generates billions of size recommendations for millions of consumers in 19 regions around the world every month.

2. Brand Tool Fit Insights

The Fit Insights tool for brands provides insights by extracting and summarizing customer feedback on fit, style, and fabric, allowing brands to access comparisons of product return rates and low-return similar products, summaries of consumer feedback, and size chart analysis.

This tool and insights are available for use free of charge to U.S. clothing and footwear brands that have registered in the Amazon brand registry and have sold at least 100 items in the past 12 months.

Amazon stated that by using this data, brands can better understand consumer fit issues, improve the way they communicate sizes to consumers, and even incorporate feedback into future design and manufacturing processes.

3. AI Standardized Size Charts

By using artificial intelligence, Amazon converts data into standardized sizes, removes duplicate information, and automatically corrects missing or incorrect measurements to generate more accurate, standardized size charts.

At the same time, Amazon makes the size chart easier to understand, no longer displaying complete size tables in chart format, but instead attempting new methods to provide consumers with the most relevant size and size details, such as grouping sizes based on recommended size in order to easily find their appropriate size.

4. Fit Highlights from Consumer Reviews

Amazon creates review highlights based on the size recommended for each consumer, using common themes in the reviews to provide personalized size guidance to consumers.

Specifically, this feature tells consumers whether they should increase or decrease the size of a specific garment based on reviews from consumers who purchased the same size. Amazon uses artificial intelligence to extract detailed information from customer reviews, such as size accuracy, fit to specific body areas, and fabric elasticity. It then uses AI to summarize this information into easy-to-read review highlights, highlighting relevant information for each consumer without manually sorting through hundreds of product reviews.