MELISSA WESTON ZALANDO Personalisation At Zalando @Melissa_Weston_ @Zalando 17/09/2019
OUR STORY – NOW AND THEN
WE WENT FROM START- UP… @Melissa_Weston_ @Zalando 3
…TO GROWN -UP @Melissa_Weston_ @Zalando @Melissa_Weston_ @Zalando 4
HOW OUR ASSORTMENT GREW ASSORTMENT 2008 2010 2011 2018 @Melissa_Weston_ @Zalando 5
WE BRING FASHION TO PEOPLE IN 17 COUNTRIES 2008-2009 2010 2011 2012-2013 2018 @Melissa_Weston_ @Zalando 6
FROM SCREAMING FOR ATTENTION… @Melissa_Weston_ @Zalando 7
…TO INSPIRING PEOPLE AND CREATING FASHION EXPERIENCES @Melissa_Weston_ @Zalando 8
OUR VISION: THE STARTING POINT FOR FASHION
PLATFORM STRATEGY BRANDS CONSUMERS ENABLER FASHION, TECH, OPERATIONS = the three core competencies are the basis of our platform strategy 10
SUSTAINABLY STRONG GROWTH 11 @Melissa_Weston_ @Zalando
ZALANDO AT A GLANCE > 300 visits ~ 5.4 billion EUR per million month revenue 2018 > 400,000 > 28 product choices ~ 14,000 > 80% 17 million > 2,000 employees in of visits via brands countries Europe active customers mobile devices @Melissa_Weston_ @Zalando 12
WE LOVE FASHION @Melissa_Weston_ @Zalando 13
WE OFFER A SUCCESSFUL AND CURATED ASSORTMENT > 400,000 HIGHLY EXPERIENCED articles from category management > 2,000 international brands 11 private labels CURATED SHOPPING LOCALIZATION with Zalon of the assortment @Melissa_Weston_ @Zalando 14
WE DRESS CODE 15
WE ARE CONSTANTLY INNOVATING TECHNOLOGY help our brand to HOME-BREWED, WIN ONLINE CUTTING-EDGE & SCALABLE technology solutions 7 international tech locations HQs in Berlin > 2,000 employees at @Melissa_Weston_ @Zalando 16
TORU A new employee called Toru is starting in Zalando 's warehouse in Erfurt, Germany. You might consider this nothing special, but Toru is an autonomous picking robot that can not only move whole boxes, but even individual objects. @Melissa_Weston_ @Zalando 17
PERSONALIZATION AND BIG DATA We use Big Data to better describe and identify different components of fashion across different brands so that they can be properly indexed and a more accurate recommendation made to the consumer ● Algorithmic Fashion Companion (AFC) ● Sizing (Based on size-related returns and machine learning) ● Homepage (Compositional relevance and recommendations) ● Big Data & Personalization (Dublin Tech Hub) 18 @Melissa_Weston_ @Zalando
ALGORITHMIC FASHION COMPANION (AFC) ● AFC: is a digital, scalable outfit recommendation tool, providing customers with unlimited outfit suggestions. Customers browsing Zalando are shown recommended outfits based on anchor items in their wishlist and anchor items which they have at home (previously bought on Zalando and not returned). ● Anchor items can be any article of clothing around which an outfit is then built. The AFC is an algorithm which makes use of machine learning. ● The algorithm was taught how to put together outfits by studying 200k manually-created outfits from Zalon and Shop the Look. 19 @Melissa_Weston_ @Zalando
SIZING ● Sizing: The sizing team is one of our best examples when it comes to personalization. Customers are shown recommendations on sizing in one of two ways, either by suggesting a specific size for them, or by suggestion that the brand often produces, say, shoes one size bigger or smaller than the customer might be used to. ● These size recommendations are based on two different data sets: firstly, size-related returns, in which we record whether a customer found an item of clothing to be too large or two small. ● The second data set is generated by a team of fitting models, who try on different items and compare them to their ‘regular’ size. Machine learning algorithms are then able to predict which items from certain brands are likely to have sizing issues. 20 @Melissa_Weston_ @Zalando
HOMEPAGE PERSONALIZATION ● Homepage: On the homepage, we follow two types of personalization. ● The first is what we call compositional relevance: it sorts the content entities on the home page according to your browsing behaviour, preferences and purchase history. For example if you buy a lot of items on sale, a content entity displaying sale items will be shown first. ● The second form of personalization is in the form of recommendations: what gets shown in these content entities is further personalized according to your tastes. 21 @Melissa_Weston_ @Zalando
PERSONALIZATION & MOBILE ● Personalization & Mobile: Effective personalization must be a seamless experience across all devices: ● “That’s one of the reasons w e do invest in making it clear that, if they [our customers] login, it w ill help to make the experience better for them and, if they start liking brands, then w e can save that and capture data. Ultimately w e need a 360 degree channel alignment” - Daniel Schneider. 22 @Melissa_Weston_ @Zalando
BIG DATA & PERSONALIZATION ● Big Data & Personalization: In order to analyze the big data needed for personalization, a Tech hub was opened in Dublin in April 2015. ● One aim is to develop fashion insights and secure an advanced level of personalization. The reasoning behind this was that when we receive data on different fashion products from our suppliers, more often than not it is highly inconsistent across brands. We need to be able to better describe and identify different components of fashion so that they can be properly indexed. ● Another key purpose which the Dublin office fulfils is creating a machine learning platform to allow more teams within Zalando to make use of advanced artificial intelligence applications. 23 @Melissa_Weston_ @Zalando
WE FOSTER THE INNOVATIVE POWER OF OUR EMPLOYEES HACK WEEK TECH ACADEMY INNOVATION LAB SELF- ORGANIZED GUILDS USER LAB @Melissa_Weston_ @Zalando 24
WE LIVE OPERATIONAL EXCELLENCE @Melissa_Weston_ @Zalando 25
WE OPERATE FASHION INNOVATIVE content creation 9 fulfillment centers > 20 customer care in 17 13 payment methods adapted to each markets market languages @Melissa_Weston_ @Zalando 26
ZALANDO.CO.UK 27 @Melissa_Weston_ @Zalando
THE STARTING POINT FOR FASHION 28
THANK YOU MELISSA WESTON NORTHERN EUROPE MARKETING LEAD UK & IE MELISSA.WESTON@ZALANDO.DE @Melissa_Weston_ @Zalando 17.09.219
DISCLAIMER This presentation and its contents are strictly confidential. It may not, in whole or in part, be reproduced, redistributed, published or passed on to any other person by the recipient. The information in this presentation has not been independently verified. No representation or warranty, express or implied, is made as to the accuracy or completeness of the presentation and the information contained herein and no reliance should be placed on such information. No responsibility is accepted for any liability for any loss howsoever arising, directly or indirectly, from this presentation or its contents. 30
Recommend
More recommend
