AI Revolution: How artificial intelligence is changing the world of e-commerce and how to deal with it

Natalie S, communications expert
Natalie S, communications expert
Dec 21, 2023
12 minutes
According to IBM's Global AI Adoption Index, more than 44% of companies are already integrating AI into their current applications and processes. If you're not among them, it's time to do something about being left behind.
In this article, we describe how artificial intelligence can help an e-commerce company. We look at what problems arise when implementing AI and how to solve them. We decide who should invest in in-house development and who should use off-the-shelf tools.
According to McKinsey & Company research, AI can improve customer experience on the one hand and create data and security issues on the other. So let's understand what role technology plays in online retail today.
At we do outsourced development, so we have the opportunity to immerse ourselves in the client's business and find the right solution together with them. We have cases of using AI for fashion e-commerce projects and for controlling access to the territory of shopping and business centers. In addition, we pay a lot of attention to the use of AI in food tech, which we talked about in the article AI in foodtech: plans, ideas, features.
But for now, let's get back to AI's capabilities, challenges and solutions in e-commerce.

What challenges does AI solve for businesses?

Personalisation and product recommendations

Let's break it down using Amazon as an example.
AI collects data on customers' shopping habits to suggest products that would be impossible to refuse. This includes information about purchases, views, additions to cart, time spent on site, ratings and reviews.
Analyze collected data using ML and predictive analytics algorithms. Identify patterns and trends in customer actions. For example, they can identify that customers who have purchased a TV often buy crisps.
Based on the data collected, Amazon's artificial intelligence creates unique profiles for each customer. These include information about preferences, interests, budget, and the type of device used to access the platform.

Improving the user experience

Artificial intelligence is able to provide customer support 24/7 using chatbots.
Sephora has successfully integrated AI into chatbots to improve customer interactions. What does it look like?
Counseling. Sephora's chatbots can recommend beauty products. The AI analyzes information about skin type, customer preferences and beauty features to suggest the most appropriate products.
Make-up and skincare tips. Chatbots guide you step-by-step through skincare and make-up with product information and application techniques.
Matching shades and products. By analyzing a client's color type and preferences, chatbots help them decide on the right shade of cosmetics or skin care product.
Answers to frequently asked questions. For example, about products, promotions, shop locations and opening hours.
Customer support: AI in chatbots solves many standard customer queries, easing the burden on the support team.
Sephora uses data collected through chatbots to better understand customer needs and anticipate preferences. This helps to fine-tune services and better advertise their services and products, making targeting personalized and interactive.

Demand forecasting

Artificial intelligence analyzes data on purchases, customer behavior and external factors to predict demand for goods.
Zalando, one of the largest fashion e-commerce companies in Europe, is using AI to anticipate demand for products. How does it work?
Zalando collects and analyzes a huge amount of data about purchases, product views, returns and other parameters. AI helps predict seasonal and fashion trends to prepare in advance and have current products in stock. Based on demand data and forecasts, Zalando optimizes the assortment by deciding which items to order more and which to order less. This avoids overstocking and reduces the risk of unsaleability.
Zalando also uses AI to determine the effectiveness of promotions and marketing campaigns. They can tailor promotions and discounts based on demand forecasts.

Fighting fraud

Alibaba, China's largest e-commerce company, is actively using artificial intelligence to combat fraud on its Taobao and Tmall platforms.
Alibaba's AI systems continuously analyze millions of transactions and user actions. They identify anomalies and inconsistencies that could indicate fraudulent activity. For example, they look for counterfeit products, fake reviews, and more.
For example, to identify counterfeit goods, AI analyzes photos and descriptions.
By analyzing customer behavior, AI can detect the use of stolen data.
Alibaba is also introducing biometric authentication to secure transactions and identify users through facial scanning and fingerprint recognition.

Opinion forecasting and social media analysis

Artificial intelligence tracks mentions of a brand and its products on social media to identify trends, determine what customers think, and respond quickly to feedback.
ASOS, an online fashion retailer, is introducing artificial intelligence (AI) to analyze social media and predict customer opinions.
Chatbots and algorithms search for mentions of ASOS products and brand, categorize them by sentiment (positive, negative, neutral) and analyze the content to identify key themes and trends.
It allows the company to:
Respond quickly to customer feedback and comments.
Understand which products or collections generate the most interest from the public.
Predict what fashion trends will be in demand in the future.

Logistics optimisation

Reducing time costs and lowering the cost of delivering products is another challenge that suppliers and retailers are tackling with AI.
Leading the way is "" (Jingdong), China's largest online e-commerce platform.
First of all, applies AI algorithms to optimize delivery routes. Based on order data, they select the most efficient routes and take into account a variety of factors such as road conditions.
Next, the AI system manages inventory in the company's warehouses. It optimizes the distribution of products, plans deliveries and monitors stock levels. also uses data analytics and machine learning to predict demand for goods. An important part of their logistics is automation. Robots and autonomous systems sort and pack thousands of items.
The company is also introducing intelligent vehicles, such as autonomous delivery vehicles and drones. They use AI for navigation and safety.

Virtual fitting rooms and AR shopping

With AR (augmented reality) and VR (virtual reality), e-commerce companies are giving customers the opportunity to "try on" products, such as clothing and accessories.
Let's look at the example of L'Oréal:
The company has a mobile app called "L'Oréal Makeup Genius" that allows you to virtually try on make-up products. Using the smartphone camera, the app scans the user's face and allows them to "apply" lipstick, shadows, etc.
L'Oréal also provides virtual tools for matching hair color shades. Users can change the color of their hairstyle on the screen and choose the best option.
In addition to virtual fittings, customers can virtually view the product and even place it in their environment using the camera of their mobile device.
The company is also exploring technologies for mixed reality (Mixed Reality).

What happens if you delegate tasks to artificial intelligence?

The application of artificial intelligence in e-commerce leads to multiple improvements in customer experience and optimisation of business processes. However, there are also challenges that limit the utilization of its potential.
One of these is the challenge of understanding the context and intent of customers. For example, recommendation algorithms may have difficulty trying to accurately identify customer interests. Let's say someone is browsing products as a gift, and the recommendation system misinterprets this as the customer's own preferences.
It is also important to consider the limitations of natural language processing (NLP). For example, the skills of automated chatbots to perceive and analyze complex queries are not yet well developed. Therefore, the interaction with customers is not at the same high level as when communicating with a live person from the support team. More often than not, this causes customer dissatisfaction and results in lost sales.
In addition, many are concerned about the security of their own data that companies collect for AI analysis. Collecting and processing information about customers and their purchases puts privacy and security at risk.
Some are concerned that AI could cause ethical dilemmas, such as using algorithms to manipulate consumers.
We should not forget that training and maintaining AI models is quite a costly process. After all, you need access to large amounts of data and experts to develop and update models.
A complete shift to the use of AI could lead to the loss of the human element in maintenance. And it's also important to have mechanisms for feedback and accountability in case of errors or problems caused by AI. Who is responsible for its actions if something goes wrong?
It's also very important for businesses to remember that the rules and laws governing the use of AI can vary from country to country.

How to implement AI in e-commerce painlessly?

Various strategies and techniques can be applied to address the challenges associated with the use of artificial intelligence (AI) in e-commerce.
For example, to ensure data quality, you can:
Apply cleansing and structuring to the collection of information
Develop monitoring systems
Train models on more data (e.g., team up with other companies).
Privacy and data security will help ensure:
Compliance with relevant data protection laws and standards such as GDPR in Europe.
Develop mechanisms to encrypt and anonymise data.
Regular assessment and auditing of system security.
To address ethical issues, they recommend:
Introduce ethical standards and principles into the development and use of AI
Ensure transparency in the operation of the algorithms
Evaluate the impact of AI solutions on different customer groups and their needs
And to get rid of customer dissatisfaction, we need to give customers a choice between AI-enabled service and human service.
In general, it is important to create a balance between AI-assisted automation and human maintenance, to conduct continuous training of personnel to work with artificial intelligence, and to collect feedback. It is also important to clearly define responsible persons and procedures for solving problems arising from AI actions.
You should think about how you will implement AI at the stage of developing a business strategy that takes into account the costs and benefits of AI, staff training, etc.

So what's the next step?

Artificial intelligence is a real revolution in e-commerce, and it's worth understanding who should implement it and who can benefit from ready-made tools based on it.
Which companies should be adopting AI right now?
If you have the resources and access to big data, it is definitely possible to build your own AI solutions. This is especially useful for large companies with lots of customer data.
If it's an industry where every customer is unique, as in the fashion or catering industries, AI can help improve personalisation and boost sales.
If you need to stay ahead of the competition.
But there are other companies that can benefit from AI-powered SaaS solutions:
Small or medium-sized company without huge resources
An industry that already has a wide range of out-of-the-box solutions such as analytics, marketing and inventory management.
If you still have questions about AI, you can start with another article, that can provide you with the information about how artificial intelligence is being integrated into the lives of creators and beyond.