©Art24hr via Canva.com
January 8, 2025
Is Retail Set Up To Capitalize on AI?
A recent survey found that only 40% of retailers are fully prepared to implement artificial intelligence (AI) projects now due to challenges ranging from data quality to scalability.
The poll of 200 retail executives involved in IT — taken in June last year as part of Riverbed’s broader Global AI & Digital Experience Survey — found 87% acknowledge that accurate and timely data is critical for optimal AI. However, of those surveyed, 72% are concerned about the effectiveness of their organization’s data for AI usage, and only 45% rated their data as “excellent” for accuracy, with 42% indicating their data quality is a barrier to further AI investment.
Additionally, 77% of retailers surveyed agreed that with AI still maturing, “it’s challenging to implement AI that works and scales.” Finally, data confidentiality and security risks are another hurdle, with 91% of retailers concerned that AI would access their organization’s proprietary data in the public domain due to their company using AI.
Riverbed’s study also identified a “reality gap” with retailers. A wide majority (84%) claimed to be ahead of their peers, including 35% who say they are significantly ahead, with only 4% believing they’re slightly behind. Riverbed stated, “This gap between perception and reality indicates many leaders are overconfident about where their IT function is on their AI journey relative to their industry peers.”
The findings come as the retailers surveyed expressed strong enthusiasm around the potential benefits of AI, with 95% saying AI is a top C-suite priority. Among the retailers surveyed, 65% are ramping up their AI initiatives by increasing investments in infrastructure and talent, while an additional 25% have reached the final transformative stage, with AI fully embedded in their business operations.
An EY survey of 254 CPG and retail executives in the U.S. taken in May and June 2024 also identified an overconfidence in AI preparedness.
Of those surveyed, 74% considered their companies to be “AI mature” (rating themselves a 4 or 5 out of 5), but 52% still noted that the rapid introduction of new and emerging technology “keeps them up at night.” Plus, strategic investments are just starting to ramp up, as 47% of CPG and retail executives plan to increase investments in GenAI or machine learning (ML) in the next year, up from 31% at the end of 2023.
“We’re still seeing many brands and retailers in the use case testing phase, and they have to balance the pressure for progress with the reality of the journey to a responsible, strategic and long-term AI agenda,” said Mark Chambers, EY Americas retail sector leader.
EY’s study still identified “compelling” use cases for AI, including 41% of the retailers and CPG brands seeing AI and enhanced predictive analytics as the most effective solution for shrink, while one-third are leveraging AI to enhance customer experience personalization, support decision-making in forecasting and scenario planning, and power customer service chatbots.
PwC’s “Global Artificial Intelligence Study: Exploiting the AI Revolution” report identified the three areas with the biggest AI potential for retailers:
- Personalized design and production
- Anticipating customer demand, such as using deep learning to predict customers’ orders in advance
- Inventory and delivery management
Barriers to overcome, cited by PwC’s study, include “adapting design and production to this more agile and tailored approach. Businesses also need to strengthen trust over data usage and protection.”
Discussion Questions
Do you see data quality, still-maturing AI technology, security risks, or some other factor standing out as the primary hurdle to implementing AI projects for retailers?
Where should retailers’ AI investments be prioritized?
Poll
Loading …BrainTrust
Mark Ryski
Founder, CEO & Author, HeadCount Corporation
Shannon Flanagan
VP|GM Retail & Consumer Goods at Talkdesk
Roland Gossage
CEO, GroupBy
Recent Discussions
More Discussions
How Can Retailers Crack the Gen Z ‘Riddle’?
Gen Z is becoming an increasingly important consumer demographic, but an elusive one to map. How…Does Bath & Body Works’ Scented Holiday Campaign Pass the Sniff Test?
Bath & Body Works is kicking off a new campaign with scent being at the fore…Which Tech Will Solve In-Store Inventory Woes?
Three major tech lineups are looking to solve retail inventory problems. Which is most suited?
Anyone who says they ‘get’ A.I. is lying or stupid. “A.I.” is evolving so rapidly and continuously, it’s not a destination, it’s a journey. Absolutely, data quality is a concern, but it’s only one of many. Yes, retailers need (and many have) clean data. But identifying and implementing A.I. initiatives is time consuming and expensive. At this stage, virtually everything AI-related needs to be verified. It’s still very useful, but for mission critical tasks, it requires human oversight. Oh, and have you tried to hire an A.I. specialist lately? Good luck. The lack of affordable/retainable skilled talent is a bottleneck to broader adoption. That all said, it’s important to realize that A.I. is being embedded in many applications and services that retailers buy and use every day – so there is A.I. being used. Leveraging vendors for A.I. projects makes good sense for the retailers who can’t afford to take on A.I. projects internally.
If we are evaluating whether or not retailers are ready, or able, to take advantage of AI, it would be a mistake to overgeneralize. The majority of retailers should get on board sooner rather than later, otherwise, they will lose out to their competition, and the world!
However, integrating AI into existing systems can be a complex and expensive process requiring significant investment in technology and talent. Additionally, retailers may face challenges related to data privacy and security, as well as the need to upskill their workforce. Get help right from the start using true hands-on experts.
Smaller retailers might also find it challenging to implement advanced AI solutions due to their scalability and resource demands. Bringing in able educators is a good idea if your organization is not ready. Learning should involve all departments including marketers, accountants, advertisers, and managers.
Several of your comments resonate with what I see in my everyday conversations with retailers (e.g. there are different AI adoption patterns across the broad umbrella of “retail”) but I would suggest we broaden the need to bring in educators. There is a larger need to “do AI”. Of course retailers (like all industries) need to better understand AI principles, emerging trends within the technology and core dependencies to drive wider organizational AI adoption; however, there are well proven patterns (inclusive of foundational frameworks) that can be quickly implemented to start developing an AI muscle. With limited resources, I’d argue that retailers need to bring in system integrators to bring science to their art, accelerating the benefits statement and driving efficiencies that AI promises.
I must be living on another planet. Reality gap is an understatement!
74% considered their companies to be “AI mature” (rating themselves a 4 or 5 out of 5). 25% have reached the final transformative stage, with AI fully embedded in their business operations. Are you kidding me??? For real deal, LOL.
I did an overview of AI to a women in tech group of 50 at a $10b retailer today. I conducted a few polls to kick things off. Are you using AI enable solutions to do your work? Only 33% said yes. I was floored.
Here’s another reality gap. Everyone’s talking about data and it being the #1 barrier to adoption, yes, of course; however, what’s not being talked enough is about the change management that comes along with transforming to a truly AI enabled organization. Simple things, like a shared vision, connected strategy and execution, with some sort of governance model to realize results, let alone an updated operating model that reflects the modern day, are lacking from the majority of retailers I’ve interacted with.
The biggest hurdle to implementing AI projects for retailers is poor data quality. Without clean, accurate, and timely data, even the best AI algorithms are useless.
Retailers often overestimate their readiness and focus on AI initiatives without fixing their data foundations. Security risks and immature technology are real concerns, but they pale in comparison to the chaos caused by bad data.
I think retailers should first invest in improving data collection and management systems. Once that’s solid, they should prioritize AI projects that directly impact customer experience, like personalized recommendations and predictive inventory management.
Agree 100%, @Anil Patel. Without gold standard data integrity, applying any technology – including AI – simply perpetuates inaccuracies and useless results.
Now that it’s 2025, we need to stop generalizing “all things AI” together. It simply doesn’t make sense to consider AI use across various functional areas – consumer experience, forecasting, product design, pricing, etc. – as all being similar or at the same stage of maturity. In some cases, data quality or even data availability is a challenge. In other areas, the AI solutions themselves aren’t quite mature yet. And in other cases, most businesses simply aren’t prepared for the magnitude of change required. For new and significant IT investments and projects, leadership should be asking about key criteria including business value, user readiness/sustainment, consumer implications, risk mitigation, etc. – and AI-related projects are no different.
The question is definitely, “Is AI ready to help retailers?” The answer is, at least in my mind, No (exception is machine learning for the supply chain). It is truly a solution looking for a problem in retail, and one reason we can’t quite see whether retailers are “ready for it” is we’re looking at a relatively immature solution that we hope will save the industry money.
I’ve been around this industry for a minute, and I don’t think I’ve ever seen a technology this over-hyped. Maybe RFID in the aughts (remember the Walmart 2005 mandate that solved….nothing and set that industry back a decade or more).
I’m so old-fashioned, I’d really like to see us focus on how we genuinely improve the customer experience. That’d be really, really novel. An industry-wide push.
Sorry to be so blunt. There just aren’t any magic bullets.
Wait, 74% of retail executives considered their companies to be ‘AI mature’ and yet 52% of the same executives said the rapid progress of AI technology keeps them up at night? What? These folks are either living in LaLa Land or have no clue what they’re doing. I was speaking with the head of data & analytics for a very large, very successful billion-dollar company last week about product portfolio pricing and how can he leverage AI to find opportunities to understand his true cost so as to maximize his net landed profitability. Before I could even ask a question, his very next thing out of his mouth was the concern over data integrity and quality, i.e. he knows the impacts of messy data. He gets the reality that, clearly, most executives in these surveys don’t. Perhaps these executives should spend some time in the technical trenches to learn more.
We are still well into the experimental stage of AI. And, sadly, for many, that is not going to stop AI implementations, because no one wants to be left behind in this particular implementation gold rush.
So we will just have to suffer through it until AI is more hardened and stress tested as a technology.
Wow! The confidence levels cited in the surveys and self-proclaimed maturity of AI implementations are surprising, particularly coming from retailers. There are a lot of experiments, fewer seriously investing in high-priority business projects, and even less with mature implementations driving real results. In 2024, Gartner assessed the vast majority of AI solutions in the pre-“trough of disillusionment” stages, and I haven’t seen anything to suggest that that is different for retail in early 2025.
What’s interesting about AI in retail is that, because it has been used for so long, but only become a super hot topic of conversation more recently, retailers can feel mature in their use while also worrying they’re not moving fast enough with it. The future of retail inherently relies on new use cases and data management improvements for this purpose on an ongoing basis.
AI still is in its infancy, so most retailers are in an experimental phase. As such, I find it astonishing that almost three-quarters say that they are mature in this area. We are still in the foothills of what AI can deliver, so it sounds like a lot of retailers are deluded.
Though data quality, still-maturing AI technology and security risks are factors that retailers need to consider as they implement AI, they shouldn’t consider them a complete hindrance. For instance, there are Catalog-as-a-Service (CaaS) solutions available to help retailers with data enrichment, enabling them to seamlessly clean up insufficient data like missing product info, inconsistent SKUs and errors. There are also readily matured AI solutions already on the market. Retailers should prioritize solutions built with AI from the ground up to maximize their investments rather than an add-on or plugin.
First, every retailer should have already hopped onto the AI train – even if the destination is open. Take a seat, start enjoying the view, and imagine the possibilities.
Second, data is at the center of any technology – especially AI. Remember the old adage, “Garbage In, Garbage Out?” With AI it is immediate .. and far more destructive.
Third, waiting for “perfection” or AI “maturity” is a pathway to significantly greater challenge. It’s time to accept that and get on board.
When Duke Ellington’s Orchestra and Ella Fitzgerald performed this song decades ago, they had no idea it would become the “AI Train” with lyrics that resonate in 2025:”Hurry, get on, now it’s coming. Listen to those rails a-thrumming. All aboard, get on the “A” train.”
Well put Dave. Get in there. Learn. Build skills. Get your data cleaned or your outputs will be garbage. You will get ahead but acting now. Scale later.
AI is changing so quickly that today’s “AI mature” retailer must keep up to continue to hold this self-appointed title. AI is still in its infancy as far as its capabilities go. It’s hard to predict what’s next, and it’s changing so quickly. That said, retailers can’t afford to sit on the sidelines to decide the best use for AI. Dive in now. It is affordable. If a retailer chooses to “wait and see,” they will play catch-up, which isn’t about catching up with technology. It’s about recapturing lost business and market share to competitors who are using AI.
Unified commerce’s advantages multiply with AI — since AI can pull insights from across an organization, having a unified system makes it easier to deliver the necessary data and make those insights actionable. Great times ahead for sure!
The upside in embracing AI at retail and food brand level is promising in managing inventory, shrink, customer personalization, supply chain partnerships, marketing, among many others. But AI is also in its infancy. So rather than taking a big bite of the apple and assuming you’re “there” (who on earth is really AI mature at this point?), take a long-view with testing, trial and learning. Start somewhere. Start in a narrowly where you are free to quickly measure and grow efficiencies, improvements.
Retailers first need to focus on building a strong data foundation. Investing in tools and technologies that capture accurate, real-time data should be a top priority. Without reliable data, AI will struggle to deliver meaningful insights or drive effective decision-making.
Well…here’s a topic I’ve been waiting to talk about here for a long time!
My view, informed by conversations with a number of leaders from across our industry, is that data quality, and by extension, the elimination of outdated data siloes that segment online and in-store data, is a necessary foundational element for the future of retail not only as it relates to AI, but personalization, retail media, supply chain, planning & forecasting, and product assortment strategy. Fixing the foundations of data quality now is critical in my view to unlocking the future value of this valuable asset for our industry.
With that said AI has the potential to permeate so many areas of retail that I fear many firms are chasing the buzzwords and lack a deep understanding of what AI is and how it can improve the business. Investing in education on AI basics for your leadership team seems like a critical investment in the early days of 2025. From that effort, a prioritized listing of what area of the business can be improved, and the investments necessary to bring that business benefit to life, are critical next steps.
Finally, establishing a clear set of ethical guidelines for how AI will and will not be utilized is a necessary step in the early days of the integration of this powerful technology that will prevent utilization that plants the firm, and its reputation, in hot water at some future date.
Regulation will likely not be able to keep up with the progress of this technology, so brands across retailing will have to set their own initial guardrails for the use of this technology in a way that benefits the business, improves customer experiences but does not legal or reputational hard to the business that cannot be overcome.