Zalando Co-CEO on Introducing Data Science in Fashion Retail

Banner State of Fashion Technology

This article first appeared in The State of Fashion: Technology, a detailed report co-published by BoF and McKinsey & Company.

Zalando is the largest European retailer only online, but there is another way he often describes himself: the most modern European technology company. Technology has been key to the company’s business since its founding in 2008 in Berlin. Today, it uses data to optimize everything from the way it buys products from brand partners to the way it delivers items to customers. It also uses technologies, including artificial intelligence, to provide customers with a more personalized experience on their website and in the app. The approach succeeded: in fiscal 2021, the total volume of goods on its platform rose 34 percent year-on-year to 14.3 billion euros ($ 15.7 billion), bringing in revenue of 10.4 billion euros.

Robert Gentz, co-founder and co-CEO, is helping Zalando move towards the following goal: by 2025, it expects annual sales of goods to reach 30 billion euros, as it aims to occupy more than 10 percent of the European fashion market. That is high ambition, and far from guaranteed because the competition is growing online. If Zalando wants to achieve that, it must continue to stand out, and technology will be vital in these efforts.

BoF: Personalization has been a major focus in Zaland for years and is a key part of the user experience it offers. Why is this so important to the company?

Robert Gentz: You have 1.4 million different items in Zaland. It’s a huge choice. And then you have 48 million customers. Using technology and data to bring the right customer to the goods, or the right goods to the customer, is important because, for these 1.4 million choices, how do you make sure she finds one item? That’s why we try to use technology to personalize it for customers as much as we can. It comes down to the problem of finding a partner: how do you assemble goods with customers?

BoF: What technologies do you use for this task?

RG: It’s AI. There is one program that works, the algorithmic fashion companion, which is based on items you have bought in the past. The algorithm combines matching items on [create] outfit, which we have learned through the way people combine [items]. When you look at click-through rates and purchase rates, the garments we produce hit what customers want. Thus, algorithms are continuously improved with feedback from customer data as well as feedback from people we produce internally.

BoF: What are some of the ways the user experience on the site or in the app is tailored to them?

RG: First of all, in onboarding you already have the opportunity to express the brands you like, your sizes. This already personalizes the page for you. As for the products and goods to the teaser you see, it is customized so that the Zalando shop looks different for each individual customer after they actually interact with us.

BoF: What metrics does Zalando look at to determine if these efforts are successful?

RG: Sometimes short-term indicators are not always the ones that lead to the right long-term answers. If you just want to optimize your click-through rates, then the items that might be the coolest have the highest click-through rates, but are probably not the ones that create the right offer, the right experience in the long run. What we mainly optimize is the long-term life value of the user, and the long-term value of the user life is generated using complex algorithms that [factor] how much time you spent on the site, how much you browse and what you buy – these are different sets [key performance indicators].

BoF: Discovering new products is a kind of value that a retailer can offer to customers, but if customers receive personalized recommendations based on previous behavior, does that limit their chances of discovering new items that they might like but are not like they? have you bought in the past? Is Zalando taking any steps to take this into account?

RG: Just looking at the past does not always give an answer to the question of the future. What we actually draw a lot of inspiration from is how the music industry is trying to solve the problem. You can’t do that with machines and past behaviors alone. You always have to mix new and modern fashion elements. Here fashion people help people from technology.

BoF: So there is still old-fashioned human curation in the process?

RG: Yes. In the end, it’s all about emotions. No one just wants to shop in a large automated warehouse. It’s about art as much as it’s about science.

BoF: Determining the right size and fit of a product remains one of the biggest hurdles customers face when shopping online. Zalando has invested a lot to help solve this problem. She bought a virtual wardrobe company in 2020, has a full-size and fitting department, and is establishing a technology center in Zurich dedicated to the task. How does Zalando use technology to solve or at least reduce these problems and what solutions is it exploring?

RG: What we are trying to achieve is that by 2030, you will probably not need a physical locker room. You have the same experience everywhere. What we do at this stage is mostly based on the data we receive from our customers to help them make better choices. It is largely based on returns – why you return a particular item – and customer feedback.

We have many customers who order a very wide range of products and different brands. A customer returns an item, and another customer returns exactly the same item for the same reason, but kept a similar one. You get a data chart – a fit chart – and based on that we can make recommendations with existing customers with whom we have a deep relationship about whether the items fit or not. We have already managed to reduce size-related returns by 10 percent. The next iteration of this will be when we move more towards measuring the whole body and experimenting a lot more with 3D technology and body measurement technology.

BoF: Logistics is another complex area. How does Zalando use AI or other logistics management technologies?

RG: One of the biggest technology teams we have is working on convenience and logistics. An interesting problem is where to assign an item with [greatest] proximity to the customer in the warehouse network, which is very important to achieve sustainability and delivery time by avoiding individual shipments. Where you have size and brand and other items, it becomes very granular. This is a very big problem of data and algorithms.

BoF: Are there any features of Zaland’s organizational structure that allow it to better integrate technology and data? Even companies that want to make the most of technology are not always ready for it. For example, departments may be isolated so that they do not look at the same data to make decisions.

RG: One of the great things we are at least trying to do is bring together multifunctional teams as much as we can. We have about 2,500 software engineers working in Zaland in different teams. When we have big projects, we try to put different disciplines on the table and for all of them to see this problem.

BoF: One of the big challenges that companies face is to make sure that all the data they rely on is clean, and then they have to be able to extract valuable insights from it. How is Zalando coping with these challenges?

RG: I wouldn’t say we’re perfect at this, but we’re very focused on it. We place ownership of certain amounts of data that we produce in terms of who is responsible for it and we have constant discussions about how to get better data. It is a culture of data purity.

BoF: AR and VR have received more attention as everyone talks about the metaverse. Are there new technologies or applications that Zalando sees could have a big impact in the future?

RG: Returning to the real problems of size and fit, this augmented reality space could be a good catalyst for making real discoveries in terms of solving the virtual testing experience for customers and getting concrete answers whether an item suits you personally or not, before physically holding it in hand. It’s something we’re pretty passionate about, that this part of the metaverse could actually help us solve big problems in terms of size and fit and sustainability. When it comes to the purely virtual world and objects that live only virtually, we are still researching.

BoF: Although e-commerce has grown, stores are still where they sell the most. In 2018, Zalando launched its Connected Retail platform to offer stocks from physical stores. How is Connected Retail progressing and how does technology enable that program?

RG: It is obvious that during the pandemic, this increased considerably, so that there are now about 7,000 stores that trade on Connected Retail. This is a big part of the affiliate program. How technology can help [is that] we actually provide [partners] with interface. It does not require any integration into the store. It requires matching the inventory that the store has with the database so that customers can order from it, and it also requires a certain interface regarding the physical aspects of logistics. In the future, what becomes much more interesting is when we will be able to combine it with our efforts in local delivery [to] allow customers who want to order inventory that is nearby.

BoF: Zalando says he wants to have a net positive impact – that is, to run the company “in a way that gives back to society and the environment more than we take.” That’s a big goal and something most of the fashion industry is thinking about. What role can technology play?

RG: I think many of the fashion challenges in terms of sustainability – in terms of size and fit, overproduction, resource allocation, personalization and so on – are basically a problem of data and collaboration. As fashion brands become smarter in terms of data in terms of their own supply chain – they don’t have to be more technically savvy, but I think they’re more informed – and working together, we can all produce a fashion ecosystem that makes more sense and consumes less resources.

What we try to do ourselves is work with brands very early in the design process to understand how data can help the whole process. Less resources are spent, at least for us in terms of delivery and returns. This creates more profit funds for everyone, and this can be reinvested. But in general, what is very clear to me is that in the end it is about data, it is about cooperation, data exchange. Many of the problems we see in terms of overproduction, in terms of misproduction or non-design for circularity, can be solved in the long run.

This interview is edited and concise.

Banner State of Fashion Technology

Leave a Reply

Your email address will not be published.