AI Industry

AI Development and Trends in E-Commerce

The traditional retail industry is undergoing a significant reinvention and upgrade as more and more brick and mortar stores boost business by adopting e-commerce platforms powered by cutting-edge tech.

The traditional retail industry is undergoing a significant reinvention and upgrade as more and more brick and mortar stores boost business by adopting e-commerce platforms powered by cutting-edge tech. The recent rapid development and deployment of AI technologies such as machine learning, computer vision and reinforcement learning have enabled new e-commerce products and solutions for various scenarios and strengthened the retail value chain.

E-commerce Market Size & AI Opportunities

E-commerce business typically activates across multiple online sales platforms (e.g. Alibaba’s Taobao and Tmall, Amazon,; or on a brand’s own official web stores (e.g. Tesla, Nike, Casper). Thanks to recent advancements in AI and digital technologies, operating costs for e-commerce have been reduced, enabling more retailers to realize e-commerce transformations. The 2018 global retail e-commerce market amounted to US$2.8 trillion and is expected to grow 75 percent to US$4.9 trillion by 2021. China is the world’s largest e-commerce market with the most B2C sales and biggest consumer base. According to an AliRearch (阿里研究院) survey of more than 1,000 e-commerce retailers in China, over 80 percent of the merchants have adopted and frequently use AI tools in their businesses.

Source: AliResearch

AI Technologies Boosting the E-commerce Business

Current AI-driven e-commerce strategies are mainly supported by computer vision, natural language processing (NLP) and reinforcement learning technologies.
 Computer Vision: By leveraging computer vision, retailers are able to create product listings with high-quality images, enabling customers to better understand the details of a product or service. Computer vision technology also has prospects in areas such as algorithm-driven product poster design and product recommendations based on customers’ visual style preferences. 
 NLP: E-commerce platforms use NLP in their searching and sorting algorithms for keyword analysis and attribute extraction from product descriptions to improve the shopping experience with better product matching. Dynamic NLP-driven sorting systems can also more efficiently guide customers to trending products. 
 Reinforcement Learning: Many large e-commerce platforms utilize reinforcement learning technology and big data in user behavior prediction, to optimize product rankings on search results and boost their e-commerce conversion rates.

AI Implementations in the E-Commerce Value Chain

Product Searching: Product searching is one of the most frequently used and important features for e-commerce platforms. Customers are able to find products matching their interests through keywords, where product matching relies on NLP technologies; and visual “search by image,” which leverages computer vision. E-commerce platforms also utilize reinforcement learning technologies to optimize their ranking algorithms and deliver better search results.
 Personalized Product Recommendation: In addition to searching, e-commerce platforms also use machine learning and NLP techniques to engage consumers and make personalized product recommendations based on their shopping trends and browsing history.
 Dynamic Pricing: Many e-commerce platforms use dynamic pricing tools powered by big data and machine learning algorithms to make real-time price adjustments or predict future prices based on supply and demand projections.
 Fraud Risk Management: E-commerce retailers utilize machine learning technologies to identify potential fraudulent credit card transactions to prevent and control risks in real time and ensure secure online payments.

Representative Use Cases for E-commerce AI Implementation

Alibaba: Alibaba launched its self-developed image search engine “Pailitao” in 2014. Since then, Pailitao has been widely implemented on Alibaba e-commerce platforms “Taobao” and “AliExpres” to automatically return visually similar pictures to those uploaded by consumers, helping for example to coordinate clothing, accessories and other products.
 Pinterest: In 2017 US social media platform Pinterest launched its visual search tool “Lens” as an idea discovery engine on its mobile app. The company partnered with retail giant Target to integrate Pinterest’s Lens into the e-commerce platform to help customers find items of interest through images.
 Stitch Fix: As a personal styling service provider, Stitch Fix delivers highly personalized clothing recommendations with styling algorithms developed using machine learning technology and individual customer preferences and purchase history data.
 Amazon: Amazon has long implemented dynamic pricing algorithms for product sales on its e-commerce platform, frequently adjusting product prices based on factors such as inventory and demand trends.

Limitations of AI Application in E-commerce

Cold Start Problem: Due to data scarcity, retailers operating a new business on an e-commerce platform may not be able to take advantage of advanced AI-based features like recommendation system and dynamic pricing that rely on big data and analytics. 
 Algorithm Scalability Issue: Reinforcement learning technology can encounter performance bottlenecks on e-commerce platforms. Algorithms often struggle with scaling problems and can face challenges effectively and efficiently searching through very large decision spaces.
 Long Tail Effect: E-commerce recommendation algorithms may present only a small number of the most popular items to customers, and fail to recommend rare “long-tail” products that could be more appealing to niche consumers, as such products can for example lack sufficient ratings data.

Future Trend of AI Applications in E-commerce

Data-driven Customization: Personalized product searching and recommendation algorithms based on individual user information (e.g., social behavior, occupation, preferences, etc.) will become increasingly common on e-commerce platforms.
 Human in the Loop: Business operating models that combine AI with human expertise will become increasingly prevalent. For example, the company Stitch Fix has adopted a hybrid approach that always keeps humans in the loop to support for example product recommendations.
 Connecting Offline Channels with Online E-commerce Retailing: More retailers, especially fashion merchants, have provided “search by image” services to enable consumers to locate and make online purchases of an item they see for example at a fashion show or in a print magazine.

Source: Synced China

Localization: Tingting Cao | Editor: Michael Sarazen

6 comments on “AI Development and Trends in E-Commerce

  1. Pingback: AI Development and Trends in E-Commerce – Synced – Trendz

  2. Pingback: AI Development and Trends in E-Commerce – AI Bits – News, Code, Discussions, Community

  3. That is a great way to handle e-commerce. It is a pity not many e-commerce stores are leveraging chatbots. They should consider starting using it

  4. Thanks for sharing this valuable article, Michael.

    With advances in disruptive technologies such as artificial intelligence and machine learning, almost every industry seems to be revolutionizing at a fanatic pace. You have explained it very well how AI can be integrated with Ecommerce.

    Keep sharing.

  5. Hi, thanks! I am looking forward to improved tools and plugins like magento 2 reward system that will use AI. Also, an improved understanding of the data on users who come to the site would be important for e-commerce.

  6. The main function of advertising is to convince a potential client of the need to purchase it. She acts as a source of information and a factor of psychological impact on potential consumers, boost conversion rates. In other words, advertising affects demand, can manage it.

Leave a Reply

Your email address will not be published.

%d bloggers like this: