Inspiring and empowering customers with AI-powered experiences

A look behind the scenes at how Zalando is using technology to make the customer experience more entertaining

June 6, 2024
Technology
Three mobile screens showing the Zalando trend spotter functionality; 1: Landingpage: Buzzworthy Berlin, Trending this week: overview different trending items of clothing; 2 detail page of metallic ballerina pumps; 3 showing info on how it is currently trending - behind the buzz: this item has X percent increase in likes and add-to-bag

For large parts of the e-commerce industry, the digital discovery process for lifestyle and fashion products still prioritises convenience over inspiration, engagement and fun. We are using the latest technology – including generative AI - to solve this challenge and offer customers an entertaining one-stop shopping experience.

Tian Su talking in front of a purple backdrop with the word Shoptalk

Doing this will give our customers more reasons to visit us more frequently and spend quality time with us. Two key initiatives on this journey are the Zalando Assistant and the technology behind it, as well as Trend Spotter, a new fashion trend discovery experience that we launched recently. We sat down with Tian Su, VP Personalisation & Recommendation, to learn more about both projects and take a look at the future of customer experience at Zalando.

We have heard of the Zalando Assistant before, so let’s start with Trend Spotter. Can you tell us more about this feature?

Trend Spotter is a new experience that shows which fashion trends are emerging on Zalando across six major European fashion capitals. We are uniquely positioned in that we connect thousands of brands with millions of customers across Europe. Because of this, we are able to highlight emerging fashion and lifestyle trends in Berlin, Paris, Milan, Antwerp, Stockholm and Copenhagen every week - all powered by our own data.

It’s important to note that Trend Spotter isn’t based on sales, so not on what people are actually buying already. It’s all about people’s desire to own an item, so we use a combination of different factors, such as products being searched for, liked or added to people’s carts.

Lots of the items in the Trend Spotter will have an article, written by our fashion expert, explaining why they have been included and whether it is related to a trend, be it a global or a local one. We are bringing our tech and fashion expertise together to give customers a new way to discover items they might like based on the latest trends in the iconic fashion capitals of Europe.

That sounds cool but… is there more to come?

It’s important to understand that these innovative products and new technologies are just the beginning. We have only begun to tap into what Trend Spotter or the Zalando Assistant can offer our customers. Zalando is evolving: instead of only offering outstanding convenience and a purely transactional experience, we also want to deliver an experience that inspires and entertains our customers. With Trend Spotter, we can help designers, brands, creators and users to be aware of emerging fashion trends, remain relevant and create or wear styles that resonate with current tastes.

Mobile screen showing the Zalando trend spotter functionality; different trending items of clothing are being shown, each of them has info on how it is currently trending on the website - behind the buzz: this item has X percent increase in likes and add-to-bag

Can you give us a sneak peak of what this might look like for Trend Spotter?

The technology behind Trend Spotter allows us to see what customers want to own, so we’re talking about emerging trends. These can be triggered by many things, including culturally relevant events and personalities. For example, when the Barbie movie came out last year, we saw a huge spike in searches for what we now know as “Barbiecore”. Celebrities also drive customer behaviour on Zalando; we see this whenever Taylor Swift releases a new album. This is very useful information for us and our partners. We can see what customers want to own in the near future and use that data to adapt our offering in real time to meet these emerging demands. But that’s only one of many possibilities.

Now that we have learned more about this new product, what about the one we already know? We launched the beta version of the Zalando Assistant about half a year ago. What has happened since then?

Since we launched the Zalando Assistant, over 500,000 customers have already engaged in  conversation with it. With each conversation lasting about four back and forth discussions  on average, this has given us a lot of input to work with already. We really want to understand how customers are using our assistant and how we can make it even better for them. We have seen that the searches people make with the assistant are three times the length of searches in the search bar, so customers really use it for complex searches, describing occasions, feelings and detailed expectations about the appearance of the item they’re looking for.

You mentioned that you’re improving the assistant based on customer feedback. Can you share more about those improvements?

We are improving it in a number of different areas. One example is that the early version of the assistant was not always able to refer back to a certain point of the conversation reliably; so when a customer would say “I like the second item you showed me, can you show it to me in different colours?”, the assistant didn’t know which of the options it presented was item number two - it could be the second item of the most recent recommendation or the second item in the whole conversation - and was not able to always give a satisfactory answer to such questions. This is one of the early improvements we did based on customer feedback.

We also saw that many customers were asking which trends and items are currently “hot” in the fashion world, so we recently integrated Trend Spotter into the Zalando Assistant. Now customers can simply click on “Explore this week's trends!” and see this week’s Trend Spotter. That’s another example of how we are enhancing the experience based on what our customers tell us to do.

Nice! You were probably expecting this question, but could you also give us a sneak peak at the future of the Zalando Assistant before we finish?

Sure! We want to roll out the Zalando Assistant to more countries because it’s only available to customers in Germany, Austria, the UK and Ireland at the moment. In the long run, we want all our customers to experience this new way of navigating and exploring which is always available at their fingertips and assist them proactively. We are also starting to integrate the technology behind the assistant into other areas on Zalando, like the search bar. In the future, customers will be able to enter longer and more complex queries using the search bar (e.g. “knee-length red dress with stripes”) and get much better results back because the Zalando Assistant technology has a deeper understanding of what items look like than the traditional search feature. As you can see, there are a lot of things yet to come, so stay tuned!

We will! Thank you Tian!

Tian Su and Dirk Cohausz talking on stage in front of a purple backdrop with the word Shoptalk

Zalando SE Boiler Plate EN

About Zalando

Founded in Berlin in 2008, Zalando is building the leading pan-European ecosystem for fashion and lifestyle e-commerce around two growth vectors: Business-to-Consumer (B2C) and Business-to-Business (B2B). In B2C, we offer an inspiring and quality multi-brand shopping experience for fashion and lifestyle products to about 50 million active customers in 25 markets. In B2B, we are using our logistic infrastructure, software and service capabilities to help brands and retailers run and scale their entire e-commerce business, on or off Zalando. As an ecosystem, Zalando aims to enable positive change for the fashion and lifestyle industry.