Fit is about more than wearability – it also means catering to our customers’ personal style.
One Size Does Not Fit All – and even if it could, would we want it to? We love variety: We want to buy our cars in red, black, white, blue, and argue passionately over what to watch on Netflix, not to mention which snacks are best. Popcorn: salted or sweet?
Fashion is at the forefront of personal expression, and our job at Zalando is to make fashion as accessible, personal and relevant as possible for all our customers. Sizing is a huge part of personalization and one of the greatest challenges faced by retailers, especially when selling online. That’s why Zalando has a team dedicated to making size make sense. Comprised of data scientists, researchers, product developers, business analysts, business integration experts, category buyers and more, the Sizing Organization Team at Zalando understands that sizing is about more than “Does It Fit?”.
“The impact of size and fit extends beyond mere wearability; these factors are also central to personal style. Helping customers understand the items within the context of our vast catalog that might appeal to them is a win for brands, customers, and the fashion ecosystem as a whole and therefore will be crucial for personalization at Zalando”, says Stacia Carr, Head of Engineering and Sizing at Zalando.
The team first began working on sizing recommendations for footwear a year and a half ago, making use of actual fitting models in Berlin – as well as our own employees – in combination with machine learning and the vast quantities of available data. Shoes were the ideal subject for the sizing recommendation pilot as, compared with textiles, their composition is much simpler. Always hungry for the next challenge, however, 2017 saw Zalando’s sizing team focus on two of the most difficult categories in terms of volume and complexity respectively: men’s shirts and women’s dresses. “Making a correct size recommendation for the right dress can be very tricky. We have to consider bust, hips, length, sleeves, etc.”, says Carr, adding undeterred: “It’s a motivating new boundary for us to push – now and beyond.”
Shoes were the ideal subject
for the sizing recommendation pilot.
Their composition compared
to textiles is much simpler.
The sizing team made use of actual fitting
models in Berlin, as well as our own employees...
...and used their findings in combination
with machine learning and data science.
As for the future of sizing, Zalando likes to keep its eye on the horizon and continues innovating and reimagining fashion with a strong focus on its customers. With so many new approaches to the topic, including the use of computer vision and 3D scanning, there is no shortage of opportunities for us to improve the online shopping experience.
Making a correct size recommendation for the right dress can be very tricky. It’s a motivating new boundary for us to push - now and
HEAD OF ENGINEERING AND SIZING