Garments first undergo a sampling process using standardised, yet impractical, production models. Once approved, the same measurements are then categorised into 5 or 6 sizes, typically labelled as XXL, XL, L, M, S, etc., through a standard algorithm. However, the complexity arises when different brands utilise diverse models with unique size measurements, creating confusion for consumers. This sizing inconsistency becomes particularly evident when customers engage in online clothing purchases, potentially dissuading them despite the convenience that online shopping offers. The challenge lies in the lack of uniformity across brands, affecting the reliability of size expectations for consumers.
This leads to return of the merchandise resulting in loss of business (revenue) and also the customer. Currently, as per various reports, the return rate of purchased clothing is as high as 35 per cent, mostly due to bad sizing and fits.
A recent study by Tessuti found that for men’s online apparel companies, return rates are most commonly caused by clothing that fits too small, the source of 23 per cent of returns. For womenswear brands, clothing is most likely to be sent back because it’s too big, which is the cause 22 per cent of the time. Likewise, the children’s clothing category provides similar figures to the return rates in men’s apparel, with 31 per cent of returns due to items being too small and 16 per cent due to being too big.
AI has a solution for this, as, in addition to its immense potential for design and its capacity to optimise the manufacturing of clothing, AI has the capability to address a persistent issue in the fashion industry: uneven clothing size.
Making personalisation accessible
According to research by Epsilon, 80 per cent of US online shoppers are more likely to make a purchase if retailers offer personalisation. These findings are consistent with a recent survey by Klarna, which found that 65 per cent of respondents want more personalisation in their shopping experiences going forward, with 23 per cent saying they would be willing to rely on AI to tell them which clothes would best suit their bodies and styles.
The most common reasons given by customers for returning items were incorrect sizing or fit (93 per cent) and subpar product quality (81 per cent).
Indian womenswear brand, FS Life (FableStreet), solves the sizing issues head-on with AI algorithm and data analysis. Being an online-first brand, Ayushi Gudwani, Founder, FS Life, said that to achieve desired results, data analysis is mandatory. “The data points from 1000+ women in India have helped FS Life solve the sizing issue with our products. We use AI and data analytics to predict our next set of bestsellers across verticals and that creates a lean and efficient supply chain model with near-to-perfect sizing.”
AI innovator and sizing consultant Shopify, the e-commerce platform that enables merchants to start, grow and manage a business and True Fit, a leading AI platform that decodes size and fit for consumers and apparel and footwear retailers, were the leaders to collaborate to change the conventional sizing chart. Retailers and brands working with Shopify can now download True Fit’s AI programme, Fashion Genome, the company’s machine learning dataset that is the largest in the world.
Fashion retailer Levi Strauss & Co. offers a ‘Virtual Stylist’ function through the usage of this technology. Levi’s Virtual Stylist, an AI powered programme available on Levi.com and Facebook Messenger, uses 3D body scanning and computer vision technology to generate a digital avatar of the customer’s body and recommends the best size and fit for different Levi’s products depending on the customer’s measurements
and preferences.
Customers who are happy with their purchases are not only more likely to make additional purchases from the store, but they are also less inclined to return their clothing. The practice of a customer purchasing many sizes online with the aim of returning the ones that don’t fit can be avoided by having the customer’s measurements and sizes pre-stored in the database of the brand they have selected.
Inclusivity via technology
Companies are integrating AI into their e-commerce models in an effort to increase the appreciation and inclusion of diverse body types and sizes. Customers can enter their individual measurements to create a realistic model that precisely depicts their body type.
The technology determines client body measurements and generates detailed body models at scale by analysing millions of data points from in-depth studies and human body scans. On the basis of this information, it then suggests sizes that are best-fit.
Virtual try-on features are being added to mobile apps by online fashion shops more and more. One of the biggest names in the industry, Asos, was amongst the first to accomplish this and Walmart now offers this tool on its website. Utilising augmented reality technology, the virtual try-on feature projects the chosen item onto the user’s body, producing a realistic and captivating visual experience. With the use of this technology, online shopping might be even more enjoyable, returns can be lower and consumer happiness can rise.
Big names in fashion are starting to notice this. Girlfriend Collective, an eco-friendly activewear brand, has partnered with Bold Metrics, an AI sizing intelligent start-up, to implement the Find My Size service, a sizing programme powered by Bold Metrics’ AI-driven Smart Size Chart to find the best size for consumers’ preferred fit quickly and accurately. Bold Metrics said that ever since introducing its AI-powered solution, it has increased conversion rates and average order values while reducing return rates for companies by up to 32 per cent. Additionally, Levi’s declared that in order to improve online purchasing and diversify its model selection, it would be incorporating AI-powered models into its e-commerce strategy.
Satisfying the sustainability itch
According to a report by the British Fashion Council’s Institute of Positive Fashion (IPF), 23 million returned clothing items were disposed of or burned in the UK in 2022, producing about 7,50,000 tonnes of CO2 emissions. The most common reasons given by customers for returning items were incorrect sizing or fit (93 per cent) and subpar product quality (81 per cent).
Globally, the typical return rate for online purchases is between 20 per cent and 30 percent, according to Richpanel, a platform for online customer service. Using the Hohenstein Institute’s technology resources for textile research, Sizekick, a Munich-based firm providing sizing solutions for apparel online shops, created two new AI products: BodyFinder and BodyScanner. BodyFinder uses AI to suggest realistic body shapes and the BodyScanner allows users to scan their own bodies using a smartphone – all they need to do is turn around once in front of the camera. According to Sizekick, simple solutions like this can enable more environmentally friendly online shopping and drastically cut related CO2 emissions.
Enhancing Made-To-Measure
AI-powered made-to-measure services are also becoming popular among customers who wish to go beyond the personalised purchasing experience.
With the debut of its Mobile Tailor 2.0, a software-as-a-service (SaaS) product that enables businesses to digitally measure customers, 3DLook, an AI-powered body scanning and virtual try-on solution, unveiled a revolutionary device that gives companies the means to measure consumers’ bodies digitally with precision. The system takes a customer’s front and side shots and instantly creates 3D models and precise body measurements that businesses may use.
The solution intends to lower remake and alteration costs through the use of sophisticated computer vision pipelines, body measurement and size recommendation algorithms. According to the company, their solutions can minimise product returns for traditional fashion e-commerce enterprises by 6 per cent, eliminate the human measurement procedure and increase consumer loyalty by up to 67 percent.
Even while AI is still in its infancy, it has the potential to upend and challenge the current e-commerce landscape and demonstrate that there is no one size fits all when it comes to fashion.
Fashion BrandApparel IndustryArtificial Intelligence
