‘Shiny object syndrome risks killing AI’s true value’

Retail’s artificial intelligence gold rush is here. Barely a day goes by without a retailer or brand announcing that they’re integrating some form of generative AI-powered functionality into their operations.

Don’t get me wrong, that’s no bad thing. When deployed in the right way, AI can do genuinely incredible things. The problem, though – in my experience – is that the adoption of AI in retail is often tainted by what’s known as shiny object syndrome (SOS).

If you’ve not heard of it before, SOS is a psychological phenomenon. It’s what happens when, rather than focusing on what we’re meant to be doing, we get distracted by something new and exciting.

SOS can be a major issue when it comes to the commercial adoption of technology, AI included. The hype around genAI can be captivating, and understandably so – differentiation can be tough in retail, so when the chance comes along to grab a genuine first it can be difficult to turn down.

Whether it’s chatbots, AI-powered image generation, or something else entirely, SOS also risks killing AI’s true value because it draws our focus away from proven, practical applications that can deliver genuine returns for retailers.

SOS is a psychological phenomenon. It’s what happens when, rather than focusing on what we’re meant to be doing, we get distracted by something new and exciting

Take hyper-personalisation, an idea that’s been bouncing around retail’s collective brain for more than a decade now. In theory, and with enough information, just about any retailer could create a hyper-personalised experience for its customers.

The problem is scale. Retailers don’t just have to personalise for a handful of customer segments. They need to personalise to the individual preferences of millions – across multiple channels, in real-time, and in a way that feels seamless rather than intrusive.

The challenge isn’t just about having the data. It’s about making that data work. It needs to be able to deliver the right message or offer to the right person at the right time.

Crucially, it needs to do that profitably. This isn’t something that genAI can solve.

Personalisation at scale needs ‘traditional’ AI-driven automation and advanced machine learning algorithms. These models process customer data at a level of complexity no human team could match, identifying intent, preferences and even emotional triggers to deliver relevant experiences.

A similar approach can be applied to pricing, something that can be extremely challenging to get right. That’s particularly true when it comes to the concept of dynamic pricing, when retailers adjust the price of goods based on factors such as consumer demand, competitor pricing strategies and shifts in the wider economic outlook.

Done right, dynamic pricing can be an incredibly effective way for a retailer to optimise its revenues. Done wrong, it’s a very quick way to frustrate and alienate your customers.

The challenge isn’t just about having the data. It’s about making that data work

AI, again by virtue of its ability to parse and interpret large amounts of information, can help here too. Ultimately, the key to dynamic pricing is trust. For it to work, customers need to trust you – and that means not pulling the rug out from under them on the products they really value.

If prices on key products remain stable, for instance, customers are typically much more likely to tolerate some variability on others. AI’s analytical capabilities can help to develop fair, effective, and transparent, value-based pricing strategies.

Not every AI-powered use case is customer-facing of course. There are areas like category management where AI can help keep shelves stocked with the products that customers actually want.

By combining sales data with insights into market trends, customer behaviour, and more, predictive inventory management can help retailers to forecast demand with an incredibly high degree of accuracy. As well as optimising stock levels, this can help to reduce wastage as well.

What unites all three of the examples above is that these aren’t just ‘shiny objects’. They’re tangible, highly effective tactics that retailers can tap into today – and many already are. And make no mistake, the competitive edge they’re honing by doing so will only get sharper as the years go by.

Should we stop dreaming, then? Should we stop pondering the possibilities of a genAI-powered grocery business? Absolutely not. Like anything, though, balance is essential. Chasing after shiny objects is fine, just so long as it’s not the only pursuit.

AI’s future potential is enormous, but let’s not lose sight of what we can do with it today. The retailers with AI strategies that are purpose-driven with a focus on solutions that drive customer loyalty and efficiency will be those that stand out.