Artificial intelligence is one of the topics of the year, but it is no longer a novelty in the retail sector.

For years, intelligent algorithms have been optimizing forecasting and replenishment processes, supporting customer service staff and managing personalized dialogues with consumers. However, with the arrival of generative AI, the topic has gained new momentum.

What is truly new is not so much the technology itself, but its strategic dimension: how can retail companies integrate artificial intelligence effectively, sustainably, and holistically into their structures?

From project to practice

The new EHI study AI Integration in Retail reveals that almost all major retailers are now addressing AI at a strategic level. What began a few years ago as pilot projects is now part of concrete transformation initiatives. Nevertheless, implementation remains selective in many cases: AI is used in isolated applications, but is rarely integrated across the organization. In the context of the European AI Regulation (EU AI Act), which sets new requirements for transparency and traceability, it becomes clear that successful AI integration is less an IT project and more a transformation process that affects all levels of the organization.

A survey of AI managers shows that 67% consider their data integration to be poor or relatively poor, while only 13% rate it as good. Although large volumes of data are available, they are distributed across different systems and often suffer from insufficient quality. The lack of interfaces and the historical development of IT environments further complicate seamless integration. These structural deficits are compounded by unclear responsibilities in data management, as in many cases it is not clearly defined who is responsible for maintenance, quality assurance, and governance. Forecasts and applications lose reliability if the underlying data is inconsistent or incomplete. At the same time, early initiatives—such as the creation of data lakes or central platforms—demonstrate that targeted investments can be effective, although so far they remain the exception rather than the rule.

Price management systems

Without clear responsibilities, uniform standards, and a well-defined data strategy, the potential of AI can only be leveraged to a limited extent. The three main obstacles are clearly identified: the general availability of data, its technical integration into existing systems, and ensuring its quality and usability. AI can only fully realize its potential in retail when these three dimensions are addressed systematically.

In this context, price management systems are an illustrative example of how technology and data strategy must work hand in hand. At EuroShop, providers such as GK Software demonstrate how AI-based price optimization works when supported by a reliable data architecture.

The key to data architecture lies in finding the right balance. It is not necessarily a case of ‘the more data, the better’; it is far more important to create a clean, robust, and structured database that allows pricing algorithms to operate optimally, explains Sandy Preuß of GK Software.

According to the expert, flexible interfaces with PIM, ERP, or order management systems, as well as with external sources such as competitive price data, are particularly important in this area. These integrations prevent manual data silos, increase the consistency of inputs, and make automated decision-making processes reliable.

At the same time, hybrid integration models and the option to keep sensitive data in European or secure environments directly address the concerns identified in the study regarding loss of control and dependence on hyperscalers. For retail decision-makers, this means that investments in data pipelines, governance, and interfaces are not a secondary IT task, but a strategic prerequisite for AI applications—such as pricing—to function reliably and in compliance with legal requirements.

AI integrates across all dimensions of EuroShop

Artificial intelligence now influences almost all thematic dimensions of EuroShop. In particular, with the advent of generative AI, its use has expanded far beyond traditional areas such as replenishment, pricing, forecasting, and inventory management. Across all areas, a convergence of technology, data intelligence, and operational practice can be observed—from store planning and energy and building management to shelf management—contributing step by step to a smarter and more efficient retail environment.

For example, AI is used in the design and simulation of retail spaces. Modern 3D planning tools make it possible to virtually replicate real stores and modify them interactively: shelving systems, displays, and lighting elements can be integrated into the floor plan via drag and drop and immediately visualized as photorealistic 3D representations. This allows customer pathways, sightlines, and lighting atmospheres to be optimized in advance. Umdasch Shopfitting, for example, offers shop.up, a 3D planning program that enables retailers to visualize their store concept in real time within the real environment using augmented reality. They can also test different lighting scenarios on pilot shelving. Thanks to AI-assisted design tools like these, design cycles are significantly shortened and store layouts can be adapted more flexibly.

Key findings of the study 

• Responsibilities defined: 100% of the companies surveyed have established clear responsibilities for AI within the organization, and 71% have specialized teams.
• Early adopters remain the exception: only 16% of respondents consider themselves active pioneers.
• Data integration remains a weak point: 67% rate their current data integration as poor or relatively poor.
• Sovereignty gains importance: 21% prioritize processing sensitive data exclusively on European servers.
• Beyond technology: the results show that technological progress alone is not sufficient; strong leadership, clear governance, targeted training, and the active involvement of employees are key factors.