IT in Retail

2024

A video surveillance system has been launched in Russian stores, which recognizes the thief before they commit theft

The Vidar company, together with Petrozavodsk State University, in early December 2024 presented a neural network video surveillance system capable of preventing theft in stores by analyzing the behavior of customers. The first tests showed an 80% reduction in losses. Read more here.

Robots used in Kh5 stores and warehouses. Photo

On November 21, 2024, X5 Group, which manages the retail chains Pyaterochka, Perekrestok, Chizhik, Okro, Victoria, Krasny Yar and Slata, announced the start of operation of mobile loader robots (FMR) at one of its sites. These machines are capable of performing physically complex and routine tasks on their own, reducing the burden on personnel and increasing the efficiency of operations. Read more here

Stores began to use smart carts that recognize goods themselves. You can pay them without a cash register

On November 12, 2024, the American company Instacart and the Australian retail chain Coles Supermarkets announced the introduction of Caper Carts smart shopping carts with artificial intelligence. The technology allows you to pay for selected goods, bypassing traditional cash desks, which increases convenience and saves time. Read more here.

Named the most important digital technologies for organizing sales

On August 28, 2024, Gartner unveiled a list of the most significant digital technologies for organizing sales. Various systems based on artificial intelligence will have a significant impact on the industry.

Digital Client Twin (DToC)

This is a dynamic virtual client view that businesses can use to model as well as predict consumer behavior. Customers can be individuals, groups of people, companies, organizations, etc. The DToC model helps providers better understand their customers and predict their behavior based on specific data and parameters. Digital twins increase business efficiency and enable personalized service delivery, which drives sales.

Metauniverses

The goal is to integrate a variety of products and services into a single user interface. Development metauniverses in the future will be ensured by the huge potential of the markets of forming technologies, as well as the growing demand for them. Gartner believes that metaverse will provide a new layer of integration between the physical and virtual worlds, opening up additional business opportunities.

Augmented and Virtual Reality (AR/VR) in the sales sector

These technologies will help companies shape a hybrid sales environment. Industries where physical product demonstrations are critical for revenue generation are implementing solutions that can mimic offline sites. AR/VR-solutions are in demand in areas such as retail, manufacturing, medical device sales, automotive, jewelry, real estate, etc.

Emotional AI

This is an AI that allows a computer to recognize and interpret human emotions and respond to them. A camera, microphone and various sensors can be used to collect data. The information obtained is analyzed using a special neural network, which allows you to determine emotions. In fact, emotional AI turns a person's behavioral characteristics into data that helps sales departments adapt to each specific customer and their current state. Machines will become more "humanized" as they can determine moods in many different contexts. Contact centers use algorithms based on voice analysis and natural language processing to identify emotions, followed by adaptation of the dialogue. In addition, computer vision-based AI is widely used in marketing research to assess user responses to certain products.

Generative AI for Sales

GeneAI can naturally respond to a conversation with a person, as well as serve as a tool for serving customers and personalizing their workflows. Chat bots, voice bots and GinI-based virtual assistants respond more accurately to consumer requests when first accessed. By submitting personalized offers and communicating with customers, more customer engagement can be achieved. In addition, with the help of Genia, it is possible to optimize business processes by quickly and efficiently analyzing huge amounts of information. The technology can be applied to all areas of the business, including marketing, customer service and sales.

Digital Trading Halls

These are special microsites created for a vendor and a specific group of buyers. Key features of the model include personalized interaction, two-way collaboration and content sharing, built-in video conferencing, consumer engagement analytics, and more.^[1]

Smart cameras, robots and neural networks. What digital technologies are used by Russian retailers to save more and earn money

Russian retailers are actively digitalizing, introducing various technologies in order to save and earn more, as well as increase customer loyalty. About how smart cameras, robots, neural networks and other technologies help retail networks, their representatives told TAdviser in June 2024.

M.Video relies on artificial intelligence. For example, it analyzes the behavior of customers when choosing a product and other indicators, and then forms a set of goods on shelves in stores for different groups of customers, regions and depending on the season. 

𝓲

Unsplash

Another project at M.Video is video analytics of visitors. This retailer's own development allows each employee to receive real-time notification that there is a customer on the sales floor to approach.

Neural networks in retail are used to analyze data, respond to reviews on marketplaces, recognize texts and pictures, check the assortment, identify fraudulent actions, in marketing and other tasks. At the same time, Dmitry Smirnov, director of business development in retail and FMCG at CROC, notes that the immaturity of artificial intelligence and gaps in legislation restrain the development of these technologies in the industry: no more than 30% of retailers use AI.

X5 Group told TAdviser that the effect of the introduction of artificial intelligence in the company in 2023 reached 5 billion rubles. The group uses these technologies to:

• price generation;
• promotion;
• assortment management;
• for store redevelopment: smart cameras help calculate the optimal distance between stands depending on the store format and the type of products on the shelves;
• minimizing queues;
• control of freshness of products or filling of shelves;
• definition of an object on a scale.

According to the retailer, by using an AI inventory planning system, the Kh5 managed to reduce the number of decommissioned products by 2%, increase the network's revenue by 1%, and also optimize the inventory of goods throughout the supply chain by several percent.

X5 also pays attention to use, electronic price tags which significantly save employee time and reduce paper costs. On average, one supermarket "" Intersection spends about 12 hours of working time every day to completely replace paper price tags and 4 hours to verify them.

Stores Dixy"" also use electronic price tags. They are equipped with a color screen based on electronic ink technology (e-ink) and extended functionality (color LEDs, the ability to work with a large temperature range, etc.).

In Magnit, in addition to the "classic" directions of using AI and machine learning (predicting sales, facilitating routine retail operations), they began to develop new solutions in this area. Among them are smart search and content generation technologies, computer vision, virtual assistants for staff, hyper-personalized solutions for communication with customers. For example, the use of computer vision on the sales floor helps to ensure control over the use of carts and baskets in order to determine and maintain their optimal number for each store, reduce losses and increase the level of customer service. 

Magnit also uses smart scales in its stores that recognize goods with 98% accuracy (customers do not need to remember the product number for weighing).

In some Russian stores you can find robots. A robot cat has appeared in Perekrestok in the north of Moscow, which delivers promotional goods and communicates with store visitors. In May 2024, Lenta involved a cleaning robot in the Super Lenta store on Novocherkassk Boulevard in Moscow, and now the retailer plans to introduce more than 100 broken robots.

𝓲

Tape

Also, more and more often you can have barista robots that cook and give out coffee, ice cream or other goods in parks or shopping centers. Robotization is one of the trends that has a high potential for development and implementation in other markets, believes Igor Vaag, deputy general director of Atol.

According to the Association of Retail Companies (AKORT), warehouse robotization, in particular, helps to reduce the cost of staff at distribution centers by up to 20%.

In 2024, Gloria Jeans launched a project with AI fitting rooms that allow you to "try on" a thing in 5-7 seconds. The buyer uploads his photo to a VR device in the store or takes it using a camera in a special "magic" mirror.

𝓲

Gloria Jeans

Major retailers are introducing contactless look payment technology through a facial recognition system. It repeatedly speeds up the payment process: with the help of a look at it, it takes 3 seconds instead of 34 seconds when paying in cash and 15 seconds when paying cashless.

Roman Sokolov, head of business development at RNT Group (part of the Reksoft group), in a conversation with TAdviser, stressed that retail shows great interest in AI, but in practice, less than 30% of companies from this industry use such technologies.

Sometimes it seems to business that it is cheaper and easier to entrust work to a person, companies lack information about real work cases, "he said.

Leroy Merlin also shared their successful examples of using digital technologies. According to the press service of the network of construction stores and home goods, thanks to the use of machine learning in logistics, the number of manual conversions of goods acceptance from suppliers decreased by 50% and the cost of assembling customer orders decreased by 25%. The retailer also uses neural networks to create visual and text content on the site (product cards began to be filled out twice as quickly).

AKORT clarifies that retail chains use generative AI to prepare texts of cards describing the goods in applications and create images, reducing the work time of employees to prepare materials by 10-15%.

Bilaine announced a 157% increase in demand for retail solutions developed by the company using big data analysis and artificial intelligence in 2023. Konstantin Romanov, director of artificial intelligence and digital products at Beeline, noted that interactive content using mobile applications or AR glasses in advertising increases conversion by 94%, which indicates a possible boom in augmented reality technologies in retail.

Auchan Tech told TAdviser that the use of artificial intelligence helps the group to debug code 15-20% faster, eliminate vulnerabilities and speed up testing processes. Auchan also uses generative networks in its marketing and promotional activities. For example, using neural networks, the retailer creates its own product catalogs.

At the same time, a huge part of the digitalization of retail is not visible to customers. According to Vyacheslav Predybailo, director of industry solutions at Evotor, retailers are investing in the development of such internal business processes as the integration of online and offline sales, cybersecurity, import substitution, replacement of an aging fleet of equipment and software.

2023

Sales of smart retail equipment in the world increased by 13% and exceeded $20 billion

In 2023, the global market for smart retail equipment reached $20.08 billion. For comparison, a year earlier, costs in this area were estimated at $17.84 billion. Thus, growth was recorded at 13%, as stated in the materials of Market Research Future, published at the end of October 2024.

Analysts take into account sales of solutions such as POS systems (software and hardware complexes designed for settlements with buyers and automation of the cashier's workplace), smart shelves, digital signage, self-service terminals, as well as systems for mobile payments. One of the industry's main drivers is the integration of Internet of Things (IoT) technologies. Retailers are increasingly using IoT devices to improve operational efficiency, improve customer experience, and optimize supply chain management . Smart equipment allows you to track inventory in real time, as well as collect data on customer behavior and preferences. This allows you to improve the efficiency of operations and quickly adapt to a changing situation. In addition, IoT devices facilitate communication between stores and consumers.

Another stimulating factor is the achievements in the field of analytics and artificial intelligence. Retailers use these technologies to process vast amounts of data and gain insights into customer behavior and industry trends. Smart equipment with AI functions allows you to personalize marketing strategies, optimize inventory management and increase the efficiency of activities in general. Using the capabilities of predictive analytics, stores can more accurately predict demand, thereby minimizing overstocking or lack of demanded products. In addition, chatbots and AI-based virtual assistants become an integral part of round-the-clock customer support, ensuring that requests and problems are resolved as quickly as possible.

In 2023, POS systems made the largest contribution to total sales - about $8.5 billion. Another $4 billion was provided by digital signage, $3.2 billion by smart shelves. The segment of self-service terminals is estimated at $2.7 billion, while mobile payment systems brought in $1.68 billion. Geographically, the leader in the global smart retail equipment market is North America with sales of $8 billion at the end of 2023. This is followed by Europe, where the result of $6 billion was recorded. The Asia-Pacific region closes the top three with an indicator of about $4.5 billion. South America's contribution is estimated at about $1.3 billion, while the Middle East and Africa accounted for less than $0.5 billion. Among the leading players in the industry are:

• Square;
• Oracle;
• Samsung;
• Intel;
• SAP;
• Honeywell;
• Microsoft;
• Amazon;
• Apple;
• IBM;
• Panasonic;
• Alibaba;
• Google;
• Nvidia;
• Cisco.

Overall, the global smart retail equipment industry has the potential for sustainable expansion, the study said. This is due to the constant development of technology and a change in consumer behavior, which opens up wide opportunities to improve retail operations and establish new ways of interacting with customers. At the end of 2024, revenue in the market under consideration is estimated at $22.61 billion. Market Research Future analysts believe that in the future, the CAGR (average annual growth rate in complex percentages) will be 12.58%. As a result, by 2032, costs globally will reach $58.3 billion.^[2]

Digitalization of retail: new priorities and trends. TAdviser Review and Rating 

Russian retail optimizes business processes, replaces foreign software, develops online trading. IT companies, in turn, are increasing expertise and continue to form a new portfolio of domestic products that help retail chains achieve significant changes in business. Read about how the retail digitalization market developed in 2022 and 2023 in a new TAdviser review.

Which major digital technologies help companies sell better

On May 16, 2023, Gartner named key digital technologies that experts say will help businesses increase consumer engagement and increase sales of products and services.

Among the main factors contributing to growth are the interaction of sellers and buyers in real time, the development of interactive services, the introduction of artificial intelligence, etc. The use of advanced digital instruments is especially important in light of the current macroeconomic environment, declining purchasing power and global market uncertainties.

The main factors contributing to growth include the interaction of sellers and buyers in real time

Generative AI-based e-mail services

Gartner believes that by 2025, about 30% of outgoing messages from large organizations will be formed using neural networks. Generative AI will help in the creation of personalized letters with offers from sellers that take into account the interests of individual consumers. Such messages will be created using CRM platform data (customer relationship management system).

Visual interaction

Such tools will help companies establish effective interaction with consumers. The parties will be able to discuss complex projects in real time with their visual presentation. Visual collaboration will take the process of communication between sellers and buyers to a new level.

Virtual Reality (VR)

VR-based solutions will allow customers to familiarize themselves with a particular product in the metaverse and get a complete understanding of it. After that, it will be possible to make an informed decision on the need for the actual purchase of the goods.

Digital Trading Rooms (DSR)

This concept provides for the formation of a digital space in which the company's product range is presented and related information is available. Such sales floors optimize the interaction of a small group of target buyers with the supplier. In addition, they can act as a platform for discussing projects.

Collaboration Solutions

Products like Microsoft Teams and Slack provide a dialog environment for effective interaction. Companies can use the technology to create a variety of consumer channels, including real-time answers to questions.

Conversation Analytics

Using AI tools to analyze conversations with buyers, companies will be able to more effectively build a dialogue, as well as provide the most relevant information and provide recommendations. Gartner predicts that by 2025, 75% B2B organizations will complement traditional sales schemes with neural network-based solutions.

Products like Microsoft Teams and Slack provide a dialog environment for effective interaction

Visual models

This approach allows customers to interact with the visual presentation of a real product. You can see how the change in certain parameters will affect the characteristics of the product in question. This simplifies the process of configuring and defining optimal requirements.

Interactive demos

The method allows consumers to interact directly with digital products, including generated data. This allows you to organize virtual tours to explore the capabilities and functions of a specific product.

Narrative automation

Technology based on AI algorithms allows you to automatically conduct any research regarding a particular company - for example, collect and analyze financial indicators. The data obtained can then be used to improve sales efficiency.^[3]

2021: Stores start using electronic price tags linked to product expiration dates

In mid-April 2021, it became known that Iper grocery stores in Milan (Italy) began to use electronic price tags tied to the shelf life of products. The longer perishable products are on the shelf, the cheaper they become, since the electronic price tag is automatically adjusted using an algorithm. The store was one of the first to roll out the technology of Israeli startup Wasteless, which aims to reduce food waste in supermarkets by selling food before it has to be discarded. Read more here.

2019

10 Retail Technology Trends for 2020

At the end of November 2019, Forbes presented 10 main technological trends in retail for 2020.

Extended reality

Virtual and augmented reality offer retailers several ways to improve customer experience. Extended reality, from viewing catalogs to "fitting," is already used by many retailers. In the future, extended reality will allow consumers to share the experience of buying in virtual reality with friends.

Predictive Analytics (Big Data)

Retail organisations have never had trouble collecting large amounts of data, but their analysis has been difficult. Predictive analytics is changing the market: retailers can now actively influence the future by analyzing consumer behavior and trends from the past. By successfully analyzing the data, retailers can personalize the shopping process, meet the needs of consumers depending on where they are, increase efficiency and reduce costs.

Forbes presented 10 main technological trends in retail for 2020

Preguessing

Any Google user expects to find what they need at the moment, allowing the search engine to predict their interests and post results in accordance with them. Retailers, which are guided by the same principle of anticipating desires, have a huge advantage. Thanks to direct digital access to consumers and powerful analytical capabilities, enterprises can predict what the client may need and rank search results accordingly.

Recommendations

Another powerful technology tool for retailers is recommendation mechanisms that help shoppers find things they never knew they had. The more data in algorithms to evaluate behavior behavior, the better the recommendations. Thanks to such systems, retailers receive double benefits: from increased sales and from increased customer satisfaction.

Automate order execution

It is estimated that by the end of 2025, more than 580,000 autonomous mobile robots will be used in warehouses, which will complete customer orders. Automation can significantly reduce order processing time. Many retailers are trying to increase the speed and flexibility of operations in this way to compete with Amazon.

Facial recognition

Facial recognition technology helps retailers actively combat shoplifting. In addition, such technology can provide a personalized approach to the buyer. Of course, in this case, legal questions arise regarding privacy, but this is unlikely to stop retailers.

Facial recognition technology helps retailers actively fight shoplifting

Inventory Management

Automated warehouses allow you to optimize inventory control. Intelligent robotics and computer systems securely store data on all products that are available online in real time. This accurate data helps retailers manage commodity flows and make reliable forecasts.

Robotic shops

Robots can answer questions and help customers search for items, as well as email out special offers or coupons. Such robots are already in use, for example, Lowe's homeware store has LoweBot, which helps serve customers.

Customer Chatbots

In retail to trade , they are quickly being implemented chat boats to serve customers. Messaging has become the preferred channel for consumer contact: it helps customers both before and after a sale, improving sales and helping customers solve problems without human intervention.

Internet of Things (IoT) and intelligent equipment.

With the amount of data available and the ability to process it, a personalized approach to the customer is provided, and intelligent equipment, such as sensors built into cabinets, are changing the approach to inventory.^[4]

How a mobile app can help retailers win a customer war

Improving business efficiency states and using technology is the main task that our country sets itself. The active introduction of new IT solutions has become one of the main trends for modern ones. At retailers the same time, keeping an eye on innovation drives the need to meet changing customer needs. What technologies give the maximum effect? What innovations and ideas do you need to pay attention to? Answering these questions, Omega-R presented an application that allows you to automatically build high-quality and long-term relationships between partners (B2B) and customers (B2C). He spoke more about this during his speech on May 29, 2019 at TAdviser SummIT, Rustam Fatykhov project manager. Omega-R

What technology traditional retailers are using to keep up with online rivals

In late May 2019, the CEO of Pensa Systems, a manufacturer of drones for retail, Richard Schwartz, explained what technologies traditional retailers use to keep up with online competitors.

Recently, retailers have been looking for better ways to automate in-store activities at no significant cost. Historically, the supply chain (from warehouses to distribution centers and retail stores) has been fairly controlled, but now it is actively changing and shrinking thanks to online trading. Traditional stores are in some cases more convenient for the customer, but they require time-consuming maintenance. Non-aggressive and affordable robotics is one way to redistribute labor in a traditional store that allows more attention to be paid to the customer and away from more boring and tedious activities such as sourcing goods from stock.

Retailers are counting on drones to deliver goods and in stores

Drones used in retail are just the tip of the iceberg of a vast system involving AI and machine learning. The drone has a low cost, is easy to prepare for work and is able to move almost instantly to any part of the store. Moving along storage shelves, the drone is able to "view" objects from different points of view, replacing up to 100 fixed cameras. Such mechanisms allow for more accurate recognition of goods than conventional stationary cameras or even ground robots. The drone itself does not recognize images, this operation is carried out on the "periphery" of the network either in the store itself or in the cloud. This allows you to scale the image recognition and processing system using the same learning mechanism for a large deployment network.

At the heart of such an AI system is machine learning, which is able to recognize goods on the shelf as the drone advances. Natural variations in products on the shelf allow you to constantly continue learning, making the system smarter and more reliable. In addition, an important part of the training of the AI system is that it studies the organization of the warehouse, observing changes and certain trends in replenishment of goods. Thus, the system simulates the shelf content and the required product range. Based on this, it is able to automatically detect various deviations, such as lack of goods, depletion of stocks, incorrect placement, etc. In this case, the system does not require additional reference data, which is often either unavailable, simply incorrect, or missing.

Such a form of autonomous perception, when the AI system learns through visual observation, as a person, can be used in a wide range of services - for example, to assess the condition of products on the shelf, identify leaks of liquids, problems with temperature or other equipment, etc.

Facial recognition systems hold great promise for retail use

Drones outperform ground robots in many ways. Electromechanical robots are large, bulky, expensive and cannot operate without human control, which can cause additional problems in terms of reliability. Most importantly, they still cannot provide an adequate view of all shelves and inventory. From the air, drones can see virtually everything from different perspectives and can be quickly redirected to different parts of the warehouse. Fixed cameras and smart shelves (Amazon Go-style) can be effective in certain cases - with potentially high investment and a tough build-up of physical infrastructure. The drone is much cheaper, it can be configured in less than 2 hours, its use is a more accurate and effective means of assessing stocks. However, the ultimate solution for retailers and the entire supply chain is sure to be a mix of drones, stationary cameras and ground robots.

The question of whether retail stores will be fully automated like Amazon Go raises plenty of discussions. Amazon Go uses a variety of smart sensors, ceiling cameras and other detectors to automate both checkout and evaluation of available items. In the future, this may become the most practical option for small stores and automated kiosks (for example, at the airport or for self-service employees). But it is unlikely that such a system will be applied to large retailers with a more diverse list of goods. Retail stores in general are turning into logistics centers. Amazon Go will find better use in automating cash transactions rather than inventory or the supply chain.

Giant retailers are actively mastering robots

Other industries could undoubtedly also benefit from using their own drones. Training autonomous unmanned aerial vehicles to detect errors in confined spaces finds application in all directions - throughout the supply chain, at production facilities and in the office. In fact, automated visual inspection, machine learning, image recognition and orientation in space using a standalone camera/sensors are already finding widespread use. Autonomous perception can be represented as an indexing system of the physical world over time that is able to detect expected and sudden changes.^[5]

2016: TAdviser Review: IT in Trade and Logistics 2016

Main article: Overview: IT in Trade and Logistics

Notes