The retail industry is surrounded by conversations about AI, from hyper-personalised shopping journeys to smarter search tools and customer-facing digital experiences.

But behind the buzz, one question matters most: is AI actually changing the business?

At eTail Asia, we spoke with Delphine Dierckx, Head of Digital Partnerships, PMO and Omnichannel Centre of Excellence at DFI Retail Group, about why successful retail AI is not about telling a flashy story. It is about solving essential business problems, building the right data foundations and using technology where it can create meaningful commercial impact.

3 Takeaways from the Interview

• Move from Shiny to Foundational: True retail optimization comes from solving operational challenges, not launching surface-level gimmicks.
• Go One Level Beyond Search: AI can help retailers move from reactive search experiences to more predictive, personalised customer experiences.
• Build a Stronger Data Culture: AI will not work if teams are not actively improving data quality, governance and the processes behind the tool.

The Contrast: "Shiny Tech" vs. Foundational Impact

For DFI Retail Group, meaningful business impact comes before flashy AI adoption.

One of the biggest misconceptions in retail today is that AI is plug-and-play. Retailers cannot simply buy a tool, press play and expect it to transform the business from day one.

AI requires foundational work. That means clean data, clear governance, strong processes and teams that understand their role in making AI effective.

Changing the Business: Predictive Forecasting

While many conversations around retail AI focus on the newest customer-facing tools, DFI Retail Group is applying AI where it can directly improve business performance: forecasting.

By using data analytics supported by AI, retailers can better understand customer patterns, shopping habits, timing and demand. This helps teams make smarter decisions about which products need to be available, where they need to be, and when customers are most likely to need them.

For retailers, this has a direct impact on the business.

Better forecasting can support stronger stock availability, improve customer satisfaction, increase productivity and help minimise waste. Instead of AI being used as a flashy add-on, it becomes part of the operational foundation that helps the business run better.

Beyond Search: The Role of AI in Hyper-Personalisation

Search helps customers find what they are already looking for.

AI can help retailers go one level further.

By learning more about what customers buy, when they shop, why they shop and how those behaviours change, AI can help retailers create more personalised and relevant experiences.

This is where hyper-personalisation becomes more than a marketing concept. It becomes a way to connect customer insight with operational execution.

It is not only about recommending the right product. It is about making sure the right product is available at the right time, in the right place, for the right customer need.

The Blueprint: Three Steps to an AI-Ready Foundation

If the data foundation is weak, the AI output will be weak too.

Delphine highlighted several non-negotiable foundations retailers need to get right before AI can deliver meaningful value.

1. Improve Data Quality: AI depends on the quality of the data behind it. Retailers need to clean, structure and maintain their data so that AI tools can produce useful outputs.
2. Build a Shared Data Culture: AI is not only a technology responsibility. Teams across the business need to understand why data quality matters and how their actions affect the performance of AI tools.
3. Put Governance and SOPs in Place: Clear governance structures, processes and standard operating procedures help teams understand what needs to be done, how data should be managed and how AI tools should be used efficiently.

Without these foundations, AI can quickly become another underperforming technology investment.

AI Readiness Checklist: Is Your Retail Business Actually Ready?

Before retailers invest in the next AI tool, they need to ask whether the foundations are already in place.

Use this quick checklist to pressure-test your AI readiness:

☐ Is your customer and product data clean enough for AI to use?

☐ Do teams understand their role in maintaining data quality?

☐ Are there clear SOPs for how data is captured, cleaned and governed?

☐ Can your AI tools connect insight to real operational decisions?

☐ Are you using AI to solve business problems, not just create a flashy customer experience?

☐ Can AI help improve stock availability, forecasting, productivity or waste reduction?