Future Mind
The retail industry has plenty of AI enthusiasm. What it lacks is consensus on where the real value lies and in what order to go after it. We asked experts from Future Mind and Solita to share their honest takes on AI's greatest transformation potential in retail, and where companies should direct their efforts first.
If you want to dig deeper into the data behind these themes, this year’s edition of Solita's European Retail Barometer is a good place to start.
41.7% of retailers see AI-driven automation as the largest opportunity, while 29% highlight personalization. The real transformation lies in connecting these end to end. Forecasting, replenishment, pricing, and promotion optimization remain high-impact areas, but value accelerates when customer insight directly informs operational decisions in real time. Preventing tomorrow’s stockout matters more than explaining yesterday’s.
Start with high-impact operational use cases, prove measurable value, and scale toward an integrated model where customer data, supply chain intelligence, and automation reinforce each other.
I clearly see the greatest transformation potential in store operations. Not because customer-facing AI lacks impact, but because store ops sit at the center of cost, complexity and omni-channel execution. Although omni-channel is now a structural reality, physical stores remain the dominant revenue engine in most retail categories, and especially in grocery. This means store execution continues to determine both customer experience and cost efficiency.
Even in an omni-channel world, the store is not just one channel among many. It is the operational backbone. Store operations carry a high-cost base and directly affect working capital, availability and customer trust. Yet it remains heavily dependent on manual processes and imperfect data — especially inventory accuracy.
Key transformation levers include:
But none of this can scale without accurate store-level inventory data. Overstated stock creates phantom availability and lost sales. Understated stock drives excess safety stock and tied-up capital. The real transformation will not come from adding another algorithm.
It will come from treating operational data quality as a strategic capability.
It is difficult to point to a single, isolated operational area, because the biggest change is currently taking place in customer interaction itself and in the way products are searched for. In my view, in the coming years a priority for the retail sector will be optimizing presence and visibility within LLM (Large Language Model) environments.
This can be described as a new form of positioning (SEO for AI). I base this thesis on three observations:
Shift in traffic sources
We are observing a clear increase in traffic redirected to retail websites directly from platforms such as ChatGPT or Gemini. Users are increasingly treating language models as the first line of shopping advice, instead of traditional search engines.
New methods of technical integration
Tools are emerging that allow retailers to integrate more deeply into AI ecosystems. One example is the OpenAI Apps SDK, which enables applications to communicate directly with the model. For retail, this means the ability to display products, inventory levels, or prices directly within the user's chat window, in response to a specific query (purchase intent).
Risk and direction of development
A key strategic question remains how this trend will ultimately evolve. There is a risk that LLM environments will begin to function like advanced price comparison engines, forcing companies to compete primarily on margin. The challenge for Data and Marketing teams will therefore be to steer integrations in a way that builds added value, rather than competing solely on price.
AI assistants and chatbots are becoming a new, preferred "shopping interface," and it is precisely in the area of customer-store interaction that I see the greatest potential for transformation. The ability to search for products, summarize reviews, or compare offers using natural language will make the purchasing process faster and more intuitive.
Solutions such as Rufus (Amazon's shopping assistant) demonstrate how important the intelligent combination of product data with a language model is. AI, properly supplied with data and embedded within the e-commerce ecosystem, can meaningfully shorten the path from customer intent to purchase and take over a large share of today's interactions with the store interface.
We see strong opportunities in both using AI to improve internal processes and to create innovation solutions for customers. Early AI adoption focused on internal efficiency such as automating invoicing, accelerating workflows, and improving decision-making
These are still important; however, today there are many transformative opportunities taking place when it comes to customer-facing innovation. Retailers are using AI to personalize experiences, predict demand, and create entirely new digital services.
Organizations that don't just optimize processes that already exist, but who turn AI capabilities into competitive advantage by embedding AI into core business models and reimagining customer engagement are the ones who are starting to see how this technology can really help their business.
In the short term, improving internal processes typically delivers the fastest measurable returns. Automation, workforce optimization, and better forecasting protect margins and improve operational control. That financial discipline is essential in a cost-pressured market. But long-term differentiation will be customer-driven. AI enables more relevant assortments, smarter pricing, and seamless omnichannel experiences.
The most successful retailers will not choose between efficiency and innovation — they will use operational improvements to directly enhance customer value. Real advantage is created when internal intelligence translates into a better customer experience.
Right now, I clearly see greater opportunities in improving internal processes. Not because customer innovation lacks potential, but because its success depends on operational excellence. AI can power impressive customer-facing solutions. But sustainable impact starts with fixing core fundamentals. Inventory accuracy is again a clear example. If store stock data is unreliable, customer promises cannot be kept.
Typical internal opportunities include:
When inventory is overstated, customers face empty shelves and broken promises. When it is understated, capital is tied up and margins erode. Customer innovation depends on operational truth. And especially in grocery, where margins are thin, volumes are high, most perishables have a short lifespan, and store execution is everything — operational AI is not just an efficiency lever. It is a competitive advantage.
The greatest potential of AI can be realized wherever there is direct interaction between the user and the system. Whether we are talking about internal processes or customer-facing solutions, the key factor is the ability to translate natural language into concrete technical actions. This allows AI to automate subsequent stages of work, shorten task execution time, and relieve operational teams.
At the same time, these same mechanisms make it possible to create entirely new customer experiences that are more intuitive, faster, and better tailored to user needs. As a result, the advantage does not lie on one side or the other, but in the effective use of AI wherever it can simplify complex interactions and accelerate the flow of information.
No single answer emerges from these conversations, and that is probably the most honest thing about them. Where AI creates the most value depends on where you are starting from. But a common thread runs through almost every response: customer-facing ambition and operational discipline reinforce each other. The question is rarely which one to pursue. It is usually which one to fix first.
K. Szmaj - VP of Strategy & Design
K. Borowiec - Business Development Manager
E. Waszkowski - VP of Commercial Development
A. Rakowski - Partner Relationship Manager
