The use of AI in retail opens up a wide range of opportunities for retailers to optimise their business processes and create a better customer experience. With the ability to process highly complex data volumes (big data) in real time and generate solutions quickly, the capabilities and potential of AI are playing a significant role in shaping the future of retail – both online and offline.
In a survey conducted in 2023 by the German Retail Federation (HDE) and Safaric Consulting, one third of the companies surveyed (145 medium-sized and large retail companies in Germany) stated that artificial intelligence is far more than the latest hype. The development of AI is proceeding at high speed and its potential is recognised by retail companies in particular. Because of the resource input required, artificial intelligence is used mainly in larger companies as a sensible area of investment. Despite growing interest in the technology, AI is neither in use nor in the planning stage in most retail companies. A lack of specific practical examples and high costs stand in the way of many AI projects.
AI applications in retail
Automated stores with self-checkout, smart shopping trolleys with barcode scanners and customer service chatbots lead to more resource-efficient processes enabling valuable time savings for both companies and consumers. AI-controlled cameras, intelligent forecasting models for inventory management and self-learning systems that control staff deployment are evidence that artificial intelligence has been integrated into the retail sector for some time and is becoming increasingly important, particularly in the areas of head office, customer experience, stores and logistics. In interview Weinheim University professor Gerald Lembke, author and expert on digital media, emphasises the urgency of coming to grips with the digital world. Retailers who complain about a lack of customers in their shop or have problems moving products off their shelves are just wasting precious time.
“The only future is one using AI tools. They’re increasingly dominating customer acquisition and sales.”
The specific question for retailers here is: Where is artificial intelligence already in use and what is the best way to get started? For an initial overview, Lembke recommends KI-TOOLPARTY.de, a portal that presents helpful AI tools, some of which are free. For a more detailed look behind the scenes of AI, we recommend you read the rest of this article for an overview of innovative applications in the retail sector.
Head office
Product range management
Effective product range management contributes decisively to the success of a company. Those organisations that regularly review and adjust their range create an attractive shopping experience for customers. By using self-learning algorithms, in-demand products can quickly be adjusted and presented in the right quantity, at the right price and in the right place.
The choice of the right tool depends on the specific requirements, goals and budgets of the company. To find out which is the most suitable software, we recommend you compare different tools. There are a number of AI tools available for retail product range management, including:
Predictive analytics tools for personalised product recommendations
Cluster analysis tools for inventory planning, demand forecasting and product range optimisation
Sentiment analysis tools for product placement and price optimisation
Dynamic price optimisation
With the help of dynamic price optimisation, it is possible to adjust prices completely automatically in real time to current customer behaviour. In addition, information such as the weather, trends and constantly changing market, environmental and competitive conditions are integrated in the price optimisation process. Here, relevant customer data is analysed in order to identify purchasing behaviour and relate it to the current business situation. The improved pricing saves resources and increases customer satisfaction.
AI-supported machine learning algorithms for analysing pricing data and sales performance
Dynamic pricing engines for adjusting prices and maximising margins.
Predictive models for improving profitability taking into account factors such as competition, demand and inventory levels
Personnel deployment planning
There’s now no more need for long, complex Excel spreadsheets when creating shift schedules. These days, easy-to-use software solutions are available for personnel deployment planning. Here, too, AI technologies use external factors such as customer purchasing behaviour and upcoming events to calculate staffing requirements and create corresponding shift schedules. This ensures sensible utilisation of the workforce and improves the efficiency of the business. The following systems can help:
Workforce management systems for planning staffing needs based on sales data, customer demand and historical data.
Automated shift planning tools for planning staffing needs, shift planning and improving staff utilisation.
Inventory management
The use of artificial intelligence enables efficient inventory management. And with the aid of virtual reality, it is also possible for the user themselves to test out different store designs, product placements and customer flow within the store. In this way, order quantities can be optimised and bottlenecks and overstocking avoided. Relevant useful AI tools include:
Inventory management system solutions that provide forecasts for demand, inventory planning and order quantity optimisation
Inventory analysis tools for managing seasonal trends and optimising ordering strategies
Energy management
With the help of energy-efficient and sustainable operating concepts, it is possible to gain a decisive competitive edge. Significantly reducing energy costs increases the attractiveness of the company since environmental and climate-conscious actions can influence today’s purchasing behaviour. Relevant applications for energy management are:
Building optimisation for improved energy efficiency
Energy monitoring and analysis to identify savings potentials and inefficient consumption patterns
Customer experience
Intelligent customer service chatbot for personalised contact (AI-created image).
Personalised marketing
Personalised offers can positively influence purchasing decisions and significantly increase sales success. In the field of personalised marketing, AI is used to analyse customer behaviour, identify preferences and interests and create personalised content and offers. You can put the theory into practice with:
Customer data platforms (CDP) for complete customer profiling
Predictive analytics tools for predicting future customer preferences and actions
Chatbots and virtual assistants for personalised interaction with customers
Visual product searching
Image recognition and computer vision algorithms help when searching for a specific product. Instead of tediously trying to search for a product name or keywords, you can use a photo to find a similar product or information about the desired product. To be able to offer this service requires, among other things:
Image recognition algorithms to identify and analyse the visual characteristics of uploaded images and so identify products.
Product databases managed and optimised by AI tools for effective visual product searching.
Personalised purchasing advice
AI can take on the role of purchase advisor by analysing customers’ preferences based on past purchases, demographic data and other available information. Using this data, AI can identify those products that match the customer’s individual needs and desires. The following applications can assist in this process:
Collaborative and content-based filtering for personalised product recommendations. This uses the purchasing behaviour of other customers with similar preferences.
Recommendation algorithms for predicting future preferences.
Fraud detection
By applying pattern recognition, artificial intelligence monitors processes in the PoS system and can point out irregularities. This enables early intervention in the case of fraudulent intent and ensures security:
Machine learning (ML) algorithms for early detection of anomalies and identification of fraudulent transactions
Network and behavioural analytics for monitoring user behaviour to identify security breaches as quickly as possible
Biometric authentication for identity verification
Chatbots
Chatbots can support staff by interacting with customers in natural language and answering their queries. This can improve customer service and reduce employee workload. Relevant examples include:
Chatbots for efficient customer service and rapid support
Chatbots for order and payment processing
Chatbots for shipping and delivery information
Chatbots can also be extremely helpful when creating job advertisements, answering customer complaints or writing job references. ChatGPT is a proven all-rounder and effectively supports retailers in time-consuming tasks. Stephan Knecht, owner of retail consultancy Fleet40, emphasises the time savings that can be achieved with the use of chatbots – and time is a valuable resource for the future.
“Mankind has always been afraid of the new. Retailers should not wait until people are no longer fearful of AI, because by then it will already be too late. My advice is to use AI early. Don’t fight it. Google is yesterday’s news. ChatGPT is the future.”
Stores
Self-check-in/checkout
Self-check-in and checkout for high-street retail enables customers to complete the shopping process smoothly and independently. Customers can check out their purchases directly via an AI-controlled system instead of having to wait in a long queue at check-out. To achieve this, the system requires:
Image and object recognition systems for price calculation
RFID technology for product tagging
Advice/ sales robot
Robots provide further support and reduce the workload for sales staff. Robotic systems can also make targeted recommendations and help with the pre-selection of products.
Customer segmentation for classification of customers
AI systems can control and automate processes along the value chain. This allows deliveries to be planned more efficiently and bottlenecks to be identified at an early stage. The following tools can be beneficial:
AI-based predictive analytics tools for identifying bottlenecks and responding quickly to changes in the supply chain.
Demand forecasting tools for identifying future demand and better inventory planning.
AI-supported solutions for monitoring and transparency across the supply chain.
Route planning
Sustainability and efficiency are key: Empty runs and standing times can be reduced with AI-supported route planning. Here, too, external factors such as seasonal events, weather and customer purchasing behaviour are taken into account by the AI system.
Routing algorithms for analysing multiple factors such as distance, traffic connections, delivery windows and load capacities.
AI tools for real-time updates and adjustments to respond to unforeseen events such as accidents or weather conditions.
Order picking
AI tools also play a major role in goods picking systems. They can result in more efficient resource utilisation, faster order processing and better customer service. The following tools can be useful:
Digital twin for efficient inventory management and optimized storage times. The digital twin, as a virtual copy of a physical store, creates data through nightly shelf scanning to enhance internal logistics. According to Lebensmittelzeitung.net, the open-source platform for digital twins is now available to every retailer.
Image recognition and AI-controlled vision systems for identifying previously scanned goods and labels.
Collaborative robots (cobots) for repetitive tasks, increasing productivity and optimising picking.
Smart wearables
Wearables are computer systems that are either worn on the body or integrated into clothing. Functions such as motion detection and pattern recognition by self-learning AI sensors can help users in numerous areas, including monitoring and improving health. Wearables have various useful features including:
Activity recognition algorithm for automatic recognition of movement patterns.
Predictive maintenance for monitoring machines or equipment and for proactive maintenance.
Natural language processing for understanding and generating natural speech.
Smartwatch that delivers movement data and messages or can be used as a navigation aid in logistics processes (AI-created image).
How does retail benefit from AI?
Established technologies in our everyday lives clearly show how artificial intelligence has revolutionised the business of online and offline commerce. According to a forecast by industry analysts, labour productivity could increase further through AI developments. Whereas an increase by 2035 of just 11% is forecast for Spain, the estimate for Germany is around 29%, which lies somewhere in the middle. Stronger productivity growth is estimated for Austria (30%), Japan (34%) and the USA (35%). With a forecast increase of 37% Sweden looks likely to make the best use of AI.
Other useful information
The use of AI offers unimagined opportunities for the retail sector. As a guarantor of success in modern retail, artificial intelligence promises an improved customer experience, cost reductions and data-driven decision-making. Optimised inventory management and logistics increase the efficiency of supply chains, thus avoiding bottlenecks and overstocking. Chatbots and virtual assistants enable 24/7 customer service, which both increases customer satisfaction and reduces employee workload. In her interview, Marilyn Repp, an expert in the fields of innovation, online trends and digitalisation, emphasises that the use of artificial intelligence is invaluable for retailers in the face of increasing personnel shortages. However, she also states:
“AI can support staff, particularly in repetitive and analytical tasks, [...] but not replace them.”
What are the challenges and risks of using AI in retail?
Integration of AI in the retail sector poses technical, economic, political and legal challenges that, especially in Western countries, are often viewed more critically than in Asia. While the challenges of integrating AI in retail are recognised globally, there often seems to be a greater willingness in Asian countries to adapt to new technologies and exploit the opportunities. In Western countries, on the other hand, the impact is assessed thoroughly at various levels to ensure that the path of integration of AI in retail is taken responsibly.
In a study by EHI Retail and RELEX (2021), 25 retail companies (23 with headquarters in Germany, 2 in Switzerland and 1 in Austria) were asked about the reasons for obstacles to the introduction of AI-driven systems. From a technical viewpoint, many retailers face the challenging task of adapting their existing systems and processes to the requirements of introducing artificial intelligence. Careful planning is therefore essential when implementing the relevant systems. In addition, lack of personnel qualified in the field of AI and level of investment are cited as obstacles to the introduction of artificial intelligence. From a business viewpoint, AI can represent both an opportunity and a threat. For example, large retailers have already reaped benefits in terms of increasing margins and improving efficiency. At the same time, though, traditional high-street retailers feel particularly threatened by advancing digitalisation and automation. Also, appropriate regulation is crucial for the responsible use of artificial intelligence. Data security, privacy and liability issues need to be clearly defined in order to increase consumer confidence and minimise potential risks.
A quick overview
Conclusion:
The potential of AI in retail is still far from being fully exploited. As digitalisation continues, many more innovative projects and solutions will emerge in future that will make retail even more efficient and customer-oriented. Retailers who are forward-looking and skilful in their use of artificial intelligence can gain a clear advantage and respond successfully and competitively to the constantly changing demands of modern retail.
