Appier has announced the campaign results of implementing its AI-based marketing personalization cloud for EF Shop, a Taiwanese fashion e-commerce brand, which has worked with Appier since 2019 to create a more engaging and personalized shopping experience.
EF Shop teamed up with Appier to initiate three-phases of digital transformation that uses AI to meet and exceed what shoppers may need and want during their customer journeys for a corresponding result enhancement.
Their experience indicated that AI product recommendation models that integrate text and images help improve website cross-selling capabilities, more accurately predict the potential needs of customers, create business opportunities and maximize the value of products.
The fashion e-commerce market is extremely competitive and fast-paced. To stand out, product recommendation has become one of the essential tools for many e-commerce. Common recommendation engines are mostly popular product recommendations based on sales, or similar product recommendations based on product categories.
However, in order to meet the diverse personalization needs of customers, e-commerce brands must provide a more diverse product recommendation mechanism in order to seize the ever-changing sales opportunities.
According to the report from McKinsey & Co, 35% of what customers purchased from Amazon and 75% of what they watched on Netflix are based on the algorithms of product recommendation. If brands can master the essentials of recommendation, they will have a greater chance to win over customers. The advanced recommendation algorithm may also be able to increase the visibility of all products by displaying more comprehensive products to customers, and e-commerce providers may be able to manage inventories more effectively.
Three-phase challenges of digital transformation
In general, the challenges of what EF Shop encountered during the three phases of digital transformation are:
● Challenge 1: Lacking an effective multichannel engagement approach
EF Shop, like most e-commerce stores, had multiple channels to reach potential customers but lacked an easy way to manage engagement on all channels, including website, social media, and email. This inability to juggle all marketing channels made it difficult for the brand to capture sales opportunities from all touchpoints.
● Challenge 2: Need for a recommendation engine
As EF Shop grew larger and stocked its platform with more items, it needed a way to better promote its products to shoppers. The recommendation section has always been a great place for shoppers to find relevant products that they may not have thought to look for. At this stage in their business, the client needed a recommendation engine for shoppers to better explore their site.
● Challenge 3: Time for more refined, personalized recommendations
In about a year’s time, EF Shop saw tremendous growth in its business from the basic recommendation feature it adopted in Phase 2.
To continue to cater to its shoppers and their evolving need for a diverse range of products and personalization in online shopping, EF Shop needed more advanced recommendations for both products and content.
Trilogy of implementing digital transformation
EF Shop looked forward to interacting more closely with customers online and providing its diversified product selection. Appier provided suitable solutions for different stages of EF Shop’s marketing processes, and gradually implemented a digital transformation trilogy for EF Shop.
● Phase 1: Multichannel marketing automation for easier engagement
EF Shop used AIQUA to automatically push out messages to customers on different channels such as its website, the LINE app, and email inboxes. In Phase 1, the client improved its web engagement that resulted in 4X the subscription rate and 3.6X the active subscriber count of the industry benchmark. With AIQUA, the client sent out 1-2 million marketing emails per month and effectively re-engaged dormant users to shop with the platform.
● Phase 2: Fundamental recommendation engine
EF Shop onboarded one of AIQUA’s recommendation engines that displays items that people who “viewed also viewed.” With this one recommendation scenario alone, which shows what people who looked at the item a shopper is currently looking at also viewed, the website was able to display personalized product recommendations to increase the probability of and speed up transaction. This section was strategically featured on product and shopping cart pages, where shoppers are likely to be thinking about what else to buy.
● Phase 3: Advanced recommendation engines with AI
EF Shop onboarded five other recommendation scenarios and, using Appier’s advanced hybrid algorithms, was able to serve shoppers with personalized recommendations. The engines used deep learning to analyze text and images from each viewed product to predict what other similar and relevant products a user may be interested in. The engines also used machine learning to analyze and predict user behavior to recommend products that will likely be viewed or purchased.
The AI-powered engine continues to learn as a shopper continues to interact with content and products, allowing the experience to become more and more personalized as time goes on. This mechanism accounted for a significant increase from the revenue ratio in Phase 2.