By Maísa Benatti, CEO, AIUTA 

When I joined Farfetch’s customer service team in Lisbon, I worked the night shift, answering customer service calls for clients in Asia and Australia. Later I moved to a different team where I started answering calls from clients in the US and Europe. I wanted to understand the problems customers were actually experiencing, and answering their calls was a very direct way to learn.

Sizing came up constantly.

Customers shopping across countries and brands were trying to answer what should have been a straightforward question: which size should I buy? Their specific needs vary by country, some clients were looking for precise measurements others for more straightforward conversions but the underlying need was the same.

Helping them could mean comparing charts, researching other websites, and translating one brand’s sizing into something familiar. A customer might know exactly what fits them at home and still have very little confidence ordering from a brand in another country.

That experience led me into product management for size and fit, and into a problem I would continue working on at Farfetch and Amazon. It also shaped how I think about the shopping experience we are building at AIUTA.

The confusion is older than e-commerce

A 1941 U.S. Department of Agriculture report described a problem that still feels familiar: garments with the same size label could vary greatly between manufacturers, and few fit without alteration. The report also documented the cost of that inconsistency, including alterations, returns, wasted time and lost goodwill.

The story of sizing began well before online shopping. Before ready-to-wear, clothing was made to measure. In the 1800s, military uniforms helped drive large-scale sizing, using body measurements to create categories that made mass production possible.

As department stores and ready-to-wear grew in the early 20th century, women’s sizing systems emerged. Manufacturers developed their own size charts, and inconsistency between brands remained a challenge.

Clothing sizes were developed to help manufacturers produce garments for groups of people. Over time, different countries, brands, and categories developed their own approaches.

The 1941 study aimed to give garment and pattern makers a more logical basis for sizing. It measured thousands of women and found that one measurement, such as bust, could not predict vertical measurements like height. It proposed combining height and weight to classify body types. The study was an important step, though it did not represent every body type.

Later, brands also changed size labels over time. A size 8 in one decade could become a size 4 in another, with marketing influencing some of those decisions. That added another layer of confusion for shoppers.

E-commerce amplified the existing problem. Customers could no longer try garments on before buying, and size and fit became a major challenge for online fashion retailers.

A size label compresses a great deal of information into a number or a letter. Much of the detail a shopper needs disappears in that process.

Two garments labelled “medium” can have different shoulder widths, lengths, proportions, and intended silhouettes. Even within a single brand, a fitted shirt and an oversized shirt are designed to sit differently on the body.

Behind those differences are patterns and grading rules: the decisions about how a garment changes from one size to the next. A larger size involves adjustments across different parts of the garment. Those adjustments are not necessarily proportional, and brands make different choices about them.

A conversion chart can translate labels. Understanding how a particular garment will fit a particular person requires much more information.

Fit also involves preference

Even accurate body and garment measurements leave questions unanswered.

Fabric stretches, drapes, gathers, and moves. A garment behaves differently depending on its construction and the body wearing it. Two people with similar measurements may prefer completely different fits. One wants a close silhouette. Another wants room to move. Someone else deliberately sizes up to create a particular look.

This is why I think about shopping confidence through two related questions:

Will this look good on me?

Which size will give me the fit I want?

A shopper may ask both at once, but answering them requires different information.

What virtual try-on can help us understand

Body scanning and fit recommendation tools emerged to help shoppers navigate size and fit online. More recently, AI-powered virtual try-on has made it possible to explore how a garment may look across different body shapes. The focus is beginning to expand from size recommendations to visual confidence.

Virtual try-on can make the visual decision much more personal. It can help someone explore a silhouette, colour, or combination of clothes using their own image.

For that experience to be useful, preserving the shopper’s body shape and the garment’s defining details matters. The result needs to give them a meaningful basis for a decision.

Predicting precise physical fit adds another set of requirements: reliable dimensions, garment construction, fabric behaviour, and the wearer’s preferences. A convincing visualisation alone cannot establish that a waistband will feel comfortable or a sleeve will allow the movement someone needs.

I think being clear about these capabilities helps us build better products. We can improve the information available to shoppers today while continuing to work on the harder questions.

A better experience starts with less guesswork

My years working on size and fit taught me to pay attention to the effort we ask of customers. Every chart they have to interpret, measurement they have to take, or unfamiliar sizing system they have to decode adds work to a purchase they were excited to make.

There is room for better garment information, better recommendations, and more useful visual experiences. They can work together to help someone understand what they are buying. Our longer-term ambition includes more size-accurate virtual try-on, realistic garment simulation and more personalized sizing at scale. When I think back to those customer service calls, the ambition is still very practical: give people enough clarity to make a choice they feel good about when the clothes arrive.