AI in Retail 2.0 – From pilots to productivity

In 2024, generative AI took centre stage as one of the most talked-about innovations in retail. But in 2025, the conversation has shifted from “what’s possible?” to “what’s working?” With nearly every major retailer experimenting with AI in some form, the winners are no longer the early adopters, they’re the fast integrators.

Retailers are no longer asking “What can AI do?” They’re asking, “How do we make it work at scale, with impact, and at speed?” This shift from exploration to execution has created a new frontier for AI in retail. It’s no longer confined to experimental labs or proof-ofconcept demos. AI is becoming deeply embedded in the operational fabric of leading retailers, transforming how they forecast demand, engage customers, personalise content, and empower employees. Today, AI in retail is less about flashy outputs and more about quiet excellence. It’s in the algorithms that optimise stock replenishment. The models that predict size preferences to reduce returns. The assistants that support store staff in real time. And the co-pilots that help marketers generate hyper-relevant campaign content in minutes, not days. Multimodal AI, capable of interpreting text, image, voice, and even video has opened the door to richer, more intuitive customer experiences. Shoppers can now search using photos, chat with bots that understand context, or receive styling recommendations that adapt to both the weather and their purchase history.

“Today, AI in retail is less about flashy outputs and more about quiet excellence.”

But for all its potential, AI still fails in many retail organisations. Not because the technology isn’t ready but because the organisation isn’t. The biggest barriers are not technical, they’re cultural and operational.

AI asks businesses to work differently. It blurs traditional roles, requires cross-functional collaboration, and challenges teams to trust data over instinct. Too often, AI is launched as a project without business alignment, stakeholder ownership, or the necessary change management. Models are trained but never deployed. Dashboards are built but never used. And employees are left out of the loop.

“AI asks businesses to work differently. It blurs traditional roles, requires cross-functional collaboration, and challenges teams to trust data over instinct.”

Retailers that succeed treat AI not as a tool but as a team member. They invest in enablement, rethink workflows, and create feedback loops between users and systems. They set clear goals tied to business outcomes and measure progress in terms of adoption and value—not just outputs. 

The leading AI retailers of 2025 are finding ways to pair high-tech with high-touch. Ulta Beauty uses AI to power virtual consultations but anchors them in personalised, human service. Zalando leverages AI for styling and returns but couples it with a deep understanding of customer behaviour. Decathlon integrates AI into sustainability efforts, using it to forecast demand and reduce overproduction.

“The question is no longer if AI belongs in retail. It’s whether your organisation is ready to put it to work.”

AI is also going green. As the environmental footprint of digital services comes under scrutiny, retailers are exploring “green AI”—models designed for energy efficiency and low carbon impact. Therefore, choosing the right AI architecture is no longer just a technical decision, it’s also about taking sustainability into account. 

In 2025, AI won’t win customers on its own. But the retailers who use it to streamline operations, scale creativity, and amplify human strengths will gain a serious competitive edge. The question is no longer if AI belongs in retail. It’s whether your organisation is ready to put it to work. 

This year, the most competitive retailers are moving beyond pilot projects to embed AI directly into their operations, customer journeys, and employee experiences. The focus is no longer simply generative content or chatbots. Retailers are investing in AI that supports business-critical outcomes—from store operations and merchandising to customer service and sustainability.

What’s new in AI for 2025? 

AI’s evolution is no longer about novelty, it’s about orchestration. Here’s what’s changing:

• Multimodal AI: Retailers are now harnessing AI that can understand and generate across text, voice, and image. This unlocks richer product discovery, smarter styling assistants, and more intuitive customer service.
• Autonomous agents: AI bots now act independently to manage inventory alerts, reorder low-stock products, optimize pricing, or escalate service issues—freeing human staff for high-value work.
• AI co-pilots for employees: Store associates are getting real-time assistance from AI—whether it’s helping answer complex product questions or identifying upsell opportunities during interactions.
• Generative design & merchandising: AI tools are supporting faster trend analysis, predictive fashion forecasting, and even initial product concepting— particularly in fast fashion and DTC brands.
• Green AI: As sustainability gains ground, retailers are exploring AI models optimised for low carbon emissions, energy efficiency, and responsible data use.

Why most AI still fails: Adaptive > Technical challenges 

Despite the wave of AI tools available, many implementations still stall. Not because the tech doesn’t work, but because the adaptive change it requires hasn’t happened. In many cases, retailers underestimate the cultural, organisational, and operational shifts needed to support AI. Here are the most common blockers: 

• Lack of internal alignment on how AI connects to business outcomes
• Inadequate training and enablement for front-line employees
• Poor data governance and fragmented tech stacks
• Unrealistic expectations of AI as a magic bullet, not a co-pilot

The best-in-class retailers don’t just install AI, they integrate it culturally and operationally. They’re rethinking workflows, reskilling staff, and treating AI like a strategic partner, not a plug-in. It is a known fact that AI disrupts traditional workflows. It asks people to trust algorithms over instincts. It calls for crossfunctional collaboration between data scientists, store staff, marketers, and supply chain leaders. And it demands clear governance, continuous training, and a culture of experimentation. These are not small asks. 

Too often, AI is introduced as a technical project when it should be a strategic transformation enabler. Technology teams may build the models, but without business alignment, employee buy-in, or operational clarity, those models rarely get adopted. Retailers that succeed integrate AI into their processes, define clear goals, and make space for trial, feedback, and iteration. They involve teams from day one, educate employees on what AI can and can’t do, and embed it within their daily tools and routines.

Real-world examples

Ulta Beauty continues to lead with AIpowered beauty consultations, virtual try-ons, and replenishment reminders. Their goal? Enhancing human connection — not replacing it.

Decathlon uses AI to forecast demand and reduce overproduction, tightly linking its sustainability goals with machine learning

Zalando leverages AI not just in styling recommendations but to power their entire returns optimisation strategy — cutting both CO2 impact and costs.

Sephora deploys generative AI to help store staff deliver expert advice more consistently and confidently across all stores.

How to get AI right in 2025 

Start with a business challenge, not a tool. Whether it’s lowering return rates, increasing in-store efficiency, or scaling content creation, tie AI directly to a measurable outcome. 

• Co-design with your employees. Your staff’s insights are critical. Use AI to augment their strengths, not replace their roles.
• Focus on data readiness. Clean, unified, and accessible data is the bedrock of any meaningful AI deployment.
• Measure what matters. Instead of vanity metrics (like chatbot engagement), look at impact: time saved, revenue gained, waste reduced.
• Prepare for governance. Ethical use of AI, bias reduction, and data privacy are no longer optional—they’re expected.

Final thought: Productivity over hype

AI isn’t a future trend; it is today’s differentiator. The question for 2025 isn’t whether to adopt AI. It’s whether you can operationalise it fast enough to create value. For retailers who succeed, AI becomes not just a competitive advantage, but also a foundational layer of modern retail.

Are you interested in the full report? Find it here.