Reading Time: 14 minutes



For decades traditional analytics have worked perfectly fine for the data-driven retail industry. However now, Artificial Intelligence (AI) and Machine Learning (ML) introduce an entirely new level of data processing which leads to deeper business insights. Data scientists could open a new world of possibilities to business owners extracting anomalies and correlations from hundreds of Artificial Intelligence/Machine Learning models.


AI in Retail: Statistics


CB Insights reports that in the period from 2013 to 2018 Artificial Intelligence startups raised $1.8 billion in 374 deals. Amazon can take credits for these impressive numbers because they made business leaders change their minds about Artificial Intelligence in the retail market – both physical and e-commerce strategies to stay ahead of the competition. At the moment over 28% of retailers are already deploying Artificial Intelligence/Machine Learning solutions, which is a sevenfold increase from 2016 where the number was only 4%. Take a look at the popularity of “AI in Retail” in Google Trends:



So how exactly you can utilize the benefits of AI in your retail business?


Using AI in the Retail Industry: SPD Group Use Case


Here, at SPD Group, we know how retail businesses could benefit from AI because we have practical experience. We developed the system for product suggestions based on tracking the customer’s location and actions in the store. It was aimed to boost sales for shop owners, improving customer experience and satisfaction by providing smart recommendations. 



This project had started as an idea to improve CRM in a supermarket. We had a goal to implement a customer identification system without any physical id cards and connect it to the existing CRM process. To achieve that we had to analyze video from cameras, identify customers in the frame, and track the customer position in the store to match it to the location of the products. In addition to this, the system should be able to alert staff when the customer is standing too long in one location so personnel could consult him if needed. All this valuable information is gathered to determine the products customer is preferring to create offerings for him in the future. 



Building this solution began with using existing security cameras in the store and setting up just a few additional cameras. We had used YOLO model with pretrained weights because of its effectiveness in identifying people. When the goal of multiple object tracking emerges, Tracklet Association method comes into play. This method processes and slightly improves information from YOLO, distinct similarities of the same visitor (this is called appearance embedding), forming tracklets and grouping them with the help of the network flow graph. To make it simple – this means that our system is able to interact with multiple customers now. We calculated the geometry of the cameras to determine their scope. Then, by implementing the perspective conversion, our system is now able to receive 2D coordinates on the location of certain customers. 



After this system was installed in the store, the owners obtained an entirely new level of insights, and with all of this information about customer’s preferences integrated with the CRM, they can predict demand for a particular product. More than that, owners come up with much more effective personal offers and promotional offers with adjusted price strategies for different groups of customers. Eliminating physical gift cards improved the shopping experience and customer satisfaction. Now, personnel can offer personalized discounts or ask about the experience with the last purchase making the customer feel even more welcomed. 



That’s was our own experience, but what about the global smart retail trends?

How is AI Being Used in Retail Today


So we are in 2019, Artificial Intelligence solutions still have plenty of room to grow. However, we can already present to you some AI in Retail use cases with proven business value. Here is how retailers could benefit from the technology.


Top 11 Uses of AI in Retail:


  1. The stores become cashier-free

The robotization of stores will result in reducing lines, lowering the number of human employees and will significantly save operational expenses. Amazon AI has already introduced checkout-free stores. The Amazon Go and Just Walk Out Shopping technology react if you took something from the shelf or put it back. When you walk out of the shop with products, the Amazon account will take money for your purchase. Amazon wants to make more shops driven by Artificial Intelligence like Amazon Go, where only six to twenty human staff members are needed.


  1. Chatbots to assist Customer Service

AI chatbots provide an even higher level of customer service, improve searching, send notifications about new collections and suggest similar products. If a customer already bought a black hoodie, a chatbot can suggest a snapback to match, and the look is complete. 80% of brands worldwide are already using or going to use chatbots in the near future. Tommy Hilfiger and Burberry have launched chatbots to help their customers navigate through their collections.



Read also about Machine Learning in Finance 



  1. In-store Assistance

Retailers also invest in technologies that help customers in the shopping process and also help staff in stores. Kroger Edge technology eliminates paper price tags in their stores, introducing smart shelf tags. It also provides video ads, nutritional info, and promotions on the displays. Lowebot, the autonomous in-store robot from Lowe, help customers to find what they need in the store in different languages. At the same time, it helps with inventory management thanks to real-time monitoring capabilities.


AI & ML solutions in the retail store



  1. Price Adjustment

Applications of AI for retail stores could help businesses to set prices for their products, visualizing likely outcomes of multiple pricing strategies. To be able to execute this, systems collect information about other products, promotional activities, sales figures, and additional data. Business leaders can present the best offers and get new customers and boost sales as a result. eBay and Kroger already apply Artificial Intelligence for their price optimization and stay flexible with the ability to adjust prices and promotions according to obtained information.


  1. Supply Chain Management and Logistics

Poor execution in this area leads to losses for retailers around the world in about $1.1 trillion every year. Leftovers and out-of-stock scenarios can be eliminated. AI in the retail supply chain can be used for restocking—calculating the demand for a particular product taking into account a history of sales, location, weather, trends, promotions, and other parameters. Morrisons has made a better situation with stock forecasting and replenishment in 491 stores with the help BlueYonder. It resulted in an up to 30% reduction of in-store shelf gaps.



Read more on Smart Supply Chain



  1. Machine Learning in Retail: Product Categorization is a great example of Machine Learning part of AI in the retail industry—it uses Machine Learning Models to classify over a million items from various sellers. Systems based on Machine Learning tag goods and sort them in different categories for customers who are seeking a particular type of product. Lalafo sellers can just upload the image of products they want to sell and Machine Learning retail software with computer vision would recognize it, classify and even suggest a price. This platform already processes more than 900 requests in a second, improving sales with relevant content leveraging Machine Learning Models.


  1. Visual Search

Visual Search systems powered by Artificial Intelligence allow customers to upload images and find similar products based on colors, shapes, and patterns. Image recognition technology from Cortexica promises close to 95% accuracy. Customers approved The Find Similar feature with 90% positive feedback. IR technology of American Eagle uses Visual Search, and not just helping people to get the same or resembling clothes but also suggests what can go well with it.


  1. Voice Search

Walmart, Tesco, Kohl’s, Costco, and many other big brands use Google or Amazon AI technology to provide customers with simple and quick voice search. Now customers can just ask Alexa for the desired item and its delivery status without typing anything. In fact, 27% of people worldwide use voice search on mobile, and 52% of them prefer it to mobile apps and websites for their convenience.


  1. Virtual Fitting Rooms

This is another awesome application of the technology we need to mention. It’s a great way for customers to save time and find the perfect outfit with all elements perfectly matching in a span of minutes! A virtual fitting kiosk from Me-Ality can scan you in 20 seconds and measure 200,000 points of your body in this period. Companies like Levi’s, Gap, Brooks Brothers, Old Navy, and others installed these scanners in their stores and receive a massive sales increase.


  1. Customer Satisfaction Tracking

Artificial Intelligence is capable to detect the mood of your customers during the shopping process. Walmart already introduced a facial recognition system for this function. The cameras are installed at the checkout, and if the customer is annoyed shop representative will talk to him. Mood tracking will definitely help to build stronger relationships with customers.


  1. Customer Behaviour Prediction

Personali and some other Artificial Intelligence platforms enable business owners to make use of behavioral economics and build individual approach to customers. Personali has an Intelligent Incentive platform that makes an analysis of customer’s psychology and emotions to increase purchases. The algorithm processes the customer’s emotional responses and behavior while previous shopping experiences and tries to come up with optimal pricing offers for a particular visitor.



Read more on AI in Customer Service



Machine Learning Predictions for Retail Business



How Retailers Could Benefit from AI and ML in Marketing? 


One of the main problems of the marketing department of any retail enterprise can be formulated in the question: “Who are our customers?”. How to collect customer data correctly, what data will be fundamental, how to process the collected data, how to apply it correctly? This is not a complete list of issues that every enterprise faces in determining its customers and further working with them. Artificial intelligence marketing solutions are not only able to answer all these questions but also can come up with winning marketing campaigns and strategies.



In January 2019, the American manufacturer of hardware and software IBM published The coming AI revolution in retail and consumer products report. It is based on a survey of 1,900 retail companies in 23 countries around the world. The study focuses on how automation can help shops reduce the impact of the human factor and improve customer service. According to the IBM study, by 2021, more than 70% of retailers and consumer goods manufacturers will use intelligent automation tools across all value chains. We will take a look at how intelligent AI changes the marketing side of retail and what changes are waiting for buyers very soon.



The Complete Transformation of Marketing Ideas in the Retail Industry

According to the above-mentioned survey, 79% of company representatives were confident that artificial intelligence technologies would revolutionize marketing, making it more strategic. The focus will be on the narrower segmentation of the target audience, product innovations and the identification of new consumer incentives. 86% also believe that the efficiency of marketing teams increases with the use of AI tools, and 82% think that they have a positive effect on interaction with the end-user, especially by automating time-consuming tasks.



All these conclusions are united by one main point: marketers can no longer live and work without Big Data retail analytics.



Big Data Retail: Processing Immense Volumes of Marketing Data 

The only thing that AI has not learned yet is how to sell products without having physical customers. Therefore, in order to use artificial intelligence in marketing in terms of working with information, you need clients and a lot of customer data. This data includes information on consumer behavior and changing preferences, social trends, and attitudes that can be traced in social media, and the diverse interests of users based on queries in a search engine. Artificial intelligence marketing tools are now available for any enterprise and retail as one of the most dynamic industries should be one of the pioneers of these technologies.



Read more on Artificial Intelligence in Customer Service



Unlimited Predictive Analytics Possibilities

Now the companies that show the dynamics of profits above average, remain flexible and quickly respond to changes in demand and the market situation. If earlier it was possible to get sales off the ground by hanging a sale sign on a store door, today it will not work. Now you need to understand the mood and behavior of consumers, and not focus only on the actions of competitors. In such matters, artificial intelligence has great potential for retail.


The algorithm is trained on the same principle as the person – on the experience of past transactions (and errors, including). The only difference is that the machine learning algorithm makes it much faster, never gets tired and does not forget anything. Living through each transaction, it captures non-linear connections between a set of factors.


It is necessary not only to understand how the price change for this product will affect its sales. It is important to separate seasonality, the actions of competitors from changes in consumer sentiment and promotional activities, and then track the relationship between positions and how a change in the set of prices will affect all indicators. AI-driven technologies are able to cope with these and other marketing tasks.


Maximum Personalization

All marketing from ancient times to yesterday was built around the product. And even when marketers started talking about customer focus, it was still a story about a product. They have always tried to make the product more satisfying, more suitable for customers. As a result, the main sales strategy was to convince the customer that he needs this product.


Today, these marketing strategies do not work. Artificially intelligent solutions make it possible to collect a huge amount of customer data – and, if processed correctly, we will be able to predict with high probability which product or service configuration will suit a particular customer. A person is not able to process such a large amount of data, but artificial intelligence AI can do it. They can process a huge amount of input information, build relationships, make a forecast and form an individual proposal.


The Most Effective Communication with the Consumer

As we have already said, the main trend in the development of communications between the brand and the buyer is to create personalized content for each client and individual product offerings. The application of artificial intelligence in marketing can help create a personalized experience, both in an online channel and in a physical store. For this, today it is possible to use chatbots, as well as providing the buyer with the possibility of shopping using voice, rather than typing on the keyboard. Consumers will soon be talking to robots and other electronic devices in real-time while shopping. This will become a reality thanks to the success in the processing of ordinary speech by artificial intelligence. The quality of interaction with the customer grows continuously, all interaction characteristics are improved, and new opportunities for personalized contact and supply are also emerging.


Read more on Machine Learning in Finance


CRM Marketing in Combination with Marketing AI

AI tools enable marketers to be more effective in using CRM systems. Today it is possible to integrate AI marketing solutions with a CRM system and share valuable data. For example, analytical CRM, developed by the American company SAS, took into account several hundred factors for each audience group. The average check, the frequency of visits to the store, favorite categories of goods, their price range, the time of visiting supermarkets mattered. Based on this data, the system predicted how customers from a particular group would respond to a specific promotional offer.


Microsoft Dynamics 365 AI is a great example of a cloud-based business solution that leverages ERP and CRM functionality along with modern BI, usage of Office 365, and optional functionality for specific use cases. It enables sales representatives to build stronger relationships with customers, make more intelligent business decisions, and close sales targets faster.


After that, CRM chose a communication channel and suggested what to write in the message. As a result, the retailer managed to increase the effectiveness of marketing campaigns by 5% and reduce the cost of communication with customers by almost half (40%).


By integrating these intelligence marketing innovations into your business, you will receive real-time recommendations, taking into account a number of parameters, including purchase history and income. It is also possible to make predictions for targeting leads who are more likely to make purchases.


E-Mail Marketing That Really Works

Your database contains information about all the people on the mailing list, all content ever sent to them, each click that reflects interaction with the brand, each product, its life cycle, etc. Can you create a perfect email campaign corresponding to the interests of each of your consumers, so as to maximize your profits? At the moment, there is not a marketing platform capable of this, but for AI this is not the problem at all.


AI will help to select only those offers, design of the letter and style of the message which most suits specific users. In addition, the AI also collects data on the best time to send your email. Analyzing previous user experience, AI will help determine the most appropriate day and time for sending letters for each subscriber. Accordingly, the sender will be able to establish the most effective communication with their customers. So, if emails are an important part of your content marketing strategy, direct your efforts towards adopting data-driven technologies.



Impact of AI in Retail


According to a retail executives survey by Capgemini at AI in Retail Conference, the application of the technology in retail could save up to $340 billion each year for the industry by 2020. The estimates are that 80% of savings will come from AI improvement of supply chain management and returns. As far as customer-facing functions, respondents believe that chatbots and self-checkout services will be the most beneficial for retailers. The global market for Artificial Intelligence Retail Business is expected to grow to $5,034.0 million by 2022.


Top AI Investments Predictions for 2019: 


  1. Global spending on AI is expected to reach $35.8 billion by the end of 2019, a 44% increase compared to 2018, according to International Data Corporation. 
  2. Retail will lead to global spend on AI systems in 2019. With the $5.9 billion investments on automated customer service agents, shopping advisers and product recommendation platforms
  3. Automated customer service will be the recipient of the biggest amount of investments in 2019 at $4.5 billion  
  4. Sales process recommendation and automation will see $2.7 billion in investments
  5. Automated threat intelligence and prevention will receive $2.7 billion
  6. More than $2 billion investments will share automated prevention maintenance, diagnosis and treatment systems, fraud analysis and investigation, intelligent process automation and program advisors and recommendation systems



The Future


According to a retail executives survey by Capgemini at AI in Retail Conference, the application of the technology in retail could save up to $340 billion each year for the industry by 2020. The estimates are that 80% of savings will come from AI improvement of supply chain management and returns. As far as customer-facing functions, respondents believe that chatbots and self-checkout services will be the most beneficial for retailers. The global market for AI in retail is expected to grow to $ 5,034.0 million by 2022.




Artificial Intelligence and Machine Learning-based solutions can help your retail business grow. Stay relevant and surpass the competitors on the market! Automated processes, better insights for your business and stronger relationships with customers will result in revenue increase. Artificial Intelligence retail solutions like chatbots, visual search, or voice search can dramatically transform your bottom line. If you ready to reach the full potential of your business with AI solutions for retail or have any questions on this subject feel free to contact us at SPD-Group!