Retail’s not evolving. It’s transforming. Getting left behind now, doesn’t mean playing catch up in the future, it means you might not have one.

In every element of retail, AI’s applications are not just being explored, they’re being leveraged for exponential results. Whether it’s pricing, forecasting, stock allocation, or hiring, AI is rewriting the rules of retail. 

And this isn’t in some sci-fi distant future, but here and now. 

So it’s no surprise that the pace of AI-led change was at the forefront of conversation at a recent industry roundtable I sat on, featuring leaders from Zalando, Dweet, Lancel, and La Retail Tech – everyone wants to know how to avoid getting left behind. 

(Spoiler alert: if you’re still sitting on messy data, outdated ERPs and guesswork pricing it could be you.)

Want to know how the conversation played out? Here’s my takeaways.

#1 Don’t prioritize what you don’t understand

Mike Hadjadj, founder of La Retail Tech, pointed out that retailers are getting overwhelmed by data that they don’t know how to use. Yet, they’re prioritizing AI investment. 

Here’s where retailers need to lean on partners they can trust to help them implement the right systems and maximize their output. Data overload leads to zero action.

#2 Rethink your strategy, don’t just automate what already exists

Right now, retailers are focusing their AI investments on improving operations – enhancing what already exists. But this is just scratching the surface of AI’s potential. AI will drive growth – not just efficiency, and that means rethinking strategy.

It’s about tools that don’t just crunch data, but know what to do with it. Be bold and set your expectations higher when it comes to AI.

#3 Ditch intuition, it’s vanity

When it comes to forecasting, intuition served retailers well enough over time. But it’s time to face facts, this doesn’t come close to the accuracy levels offered by AI.    

Take one of our clients, Roberto Cavalli. With our automated, AI-driven Replenishment module the system could dig deep into store-level data, granular product performance, and demand forecasts to create custom inventory allocations. No more one-size-fits-all stock distributions or guesswork, instead Cavalli’s stores got exactly what they needed, when they needed it.

Their existing tools relied heavily on manual processes, including spreadsheets – your typical legacy environment: limited integration capabilities for forecasting or predicting product performance. 

They’re now achieving a 10% increase in revenue across their network, while stockout rates have dropped from 20% to just 5%, allowing stores to meet demand more effectively. To think a hunch can compete against that is crazy.

There will always be a value to traditional industry experience – but sometimes a still tongue makes a wise head. Get out of your own way and let AI get to work

#4 No two stores are the same

Why should 100 of your stores all get the same stock? They might have been built to identical specs, but no two stores are actually the same. They’re not the same cities. Not the same customers. Use AI to break out of the one-size-fits-all logic that’s holding you back from making real progress.. 

Smarter allocation ≠ the same stock everywhere.

#5 Teams need leaders when it comes to AI adoption

We’ve all heard that AI is stealing our jobs, but to Laurent Piffaut, CEO of Dweet, it was clear that retail is still a people business.

But, there’s still an ever increasing number of AI applications to explore. The key is being ambitious enough to task it with the work that really matters, not just letting it take care of the admin. 

With this in mind, where AI can be used to improve performance, change management needs to be part of the equation.

As Laurent shared: “There’s resistance in some generations, some roles. That’s why change has to come from leadership. Someone has to give direction and explain the why.”

#6 Forget what you think you know about discounting

The more you discount the greater the sell-through, right? Wrong.

With AI-driven dynamic pricing and lower discounting levels retailers can increase margin and maintain sell-through. Some brands are already adjusting prices not just based on the competition, but also on stock volume. They raise prices as inventory levels drop. 

Forget markdown panic and blanket sales, Zalando’s pricing model proves that less discounting can actually drive better sell-through.“Selling a product at -10% in peak season works better than -50% at the wrong time. Everyone wins.”- Laura Toledano Khelif, GM France, Zalando.

#7 Retailers have a role to play in tool development

3,000 of Zalando’s colleagues work in tech – that’s 20% of their total headcount – and they don’t just adopt tools, they build them.

“It’s our responsibility to test, even if we fail sometimes, because that’s how we find the two or three solutions that will really make a difference.”

- Laura Toledano Khelif, GM France, Zalando 

Zalando develops tools for its own ecosystem, yes – but with a clear goal to offer them to partners and brands too. It’s how they push the whole industry forward. Not just themselves.

It’s exactly how our founders developed autone. They were once completely buried in the grunt work of supply chain management at Alexander McQueen, armed with nothing but outdated spreadsheets and sluggish software. It was messy, manual, maddening chaos. 

Using those experiences, they built autone, exactly what was needed for the tireless twenty-something-year old merchandiser, the allocator, the planner - for those bombarded with data but starved of insights. 

So what to do if you're leading a retail brand right now, and want it to stay that way?

• Find trusted partners to collaborate with on your AI journey.

• Don’t stockpile data, get the tools in place to do something with it. 

• Act now – you don’t have three years to figure this out.

Stay tuned for our next autone sessions Paris recap featuring Claire Casteig, Retail Director at Lancel.