As the Holidays Approach, Are Retailers AI-Ready to Improve CX?

Consumers may not be ready to hear it yet, but the holiday shopping season is here. Of course, it may still be invisible to them, but as retail’s biggest moment approaches like clockwork, retailers are scrambling to ensure that shelves are properly stocked and customers remain satisfied.

According to Mastercard, a 3.2% increase in holiday retail sales should be expected in 2024, despite the economic pressures that continue to weigh on shoppers. Therefore, it is incumbent on retailers to capitalize on demand, although the ways they do so may look a little different this year. The advent of AI has disrupted countless industries, and while the retail space is filled with useful applications, retailers must implement the technology in a way that resonates with their target audience.

Research from e-commerce enabler Digital River highlighted that chatbots were considered a pain point for shoppers, whose questions were often misinterpreted or too complex to receive useful guidance. In fact, 69% of shoppers said they valued interacting with a human employee.

“Some early resistance to chatbots should not hinder AI adoption by retailers – quite the opposite! Instead, they must learn from that experience to stay competitive in operations and customer experience,” said JM Erlendson, Head of Transformation Engineering at Software AG. “Logistics, supply chain and inventory management are just a few areas that can benefit from next-generation automation, and many retailers rightly see this as a path to winning the competitive market.Effective use of the right artificial intelligence can sharpen a competitive edge this holiday season, with some key benefits that will increase 2024 revenue, minimize cost, accelerate business and inform future success.

Highlighting inefficiencies with AI-powered analysis

Retail is an inherently unpredictable industry. Supply chain disruptions, large and small, can disrupt timelines, create gaps in inventory, and lead to unhappy customers. According to a recent survey by Prosper Insights & Analytics, only 20% of shoppers plan to spend more this holiday season.

65% of those buyers are doing so only because of the high prices. AI could be retailers’ answer to winning in this year’s competitive market.

“People are used to many ‘common’ uses of AI such as image generation, automatic captioning and virtual assistants, but industrial AI can be a game-changer in high-stress operational scenarios,” said Erlendson. “Supply chain and logistics people know better than most how the word goes about ‘best laid plans,’ but AI can give them the power to predict process bottlenecks ahead things go wrong and customers get angry.”

Software AG research from last year found that 86% of businesses had significantly outgrown their technology stacks, with 76% saying it had led to more chaos. In a retail environment, mixed messages in forecasting can make all the difference between a successful season and a failed one. The inclusion of artificial intelligence holds the promise of bridging this divide by extracting detailed process insights from volumes of data, making that analysis accessible and pointing people in the right direction to intervene.

This type of analysis is not strictly about best and worst case scenarios. While these are useful to see, smaller deviations are likely to be closer to the lived realities of retailers. AI-powered scenario planning allows comparison of different results and highlights areas that may sacrifice overall efficiency. This allows for quick fixes and a more efficient workflow in the long run.

Accelerate manual workflows for measurable results

Measurement should be central to retailers’ AI strategies, as otherwise it’s incredibly difficult to determine whether and how AI adoption is helping behind the scenes. For example, research from Prosper Insights & Analytics found that over 57% of consumers reported that they saw sales and promotions as more important this year, a sentiment that may have only been clear to salespeople retroactively in the past. With current iterations of AI, businesses can gather insights and experiment in real time.

Another use case is inventory management. When dealing with holiday peaks, changes in workflows are to be expected throughout the year. If a vendor follows structured purchasing and shipping cycles tied to demand forecasts, the change should show a measurable effect, either positive or negative. However, unseen nuances can complicate matters in unforeseen ways.

Backend AI has the capacity to address the root causes of process inefficiencies and illuminate the hidden reasons behind them. In the inventory example, let’s say the same company has a recurring order every 14 days for most of the year, but needs to shorten that window to 10 during the holidays due to increased demand. This may seem like a straightforward change, but its effects can be felt throughout the process chain.

For example, it may put some shipments on a weekend schedule and further complicate logistics, a consequence that may not be noticed until it is too late. AI models can predict this negative impact, or work quickly enough to spot that trend, explain the correlation, and help course-correct before the damage is done.

Contextualizing AI results for continuous improvement

It’s important to remember that tracking the right metrics can only do so much for a business – after all, the new data points that AI and analytics provide must be a spur to action if organizations are to improve continuously (or even continue.

“Data may be king, but context is its crown,” Erlendson said. “While AI can give retailers unprecedented visibility into their behind-the-scenes operations, human decision-making still bears responsibility for organizational improvement. This may change as companies set their sights on proprietary models built on proprietary databases, but for now, guidance from more generalized models should be viewed with a critical eye.”

Although AI permeates more industries every day, there is a level of personal responsibility that accompanies its use. Blind faith in its results unnecessarily increases the risks, and the real-world context of a situation must override theoretical guidance from models that are advanced but still learning.

This is not to discourage the use of AI in a retail environment. From this point forward, companies’ deployment of emerging technology will be a determinant of success not only during busy seasons, but for years to come. What is vital is HOW is being used and the weight given to its results.

Applications that deepen understanding in AI and advance business goals will prove more fruitful in both the short and long term. These have the added benefit of instilling users with confidence in their abilities and promoting process intelligence—an intimate knowledge of workflows, the ripple effects of adjusting them, and adjustments to potential deviations—at an organizational level.

It should also be noted that AI results are not limited to text blocks. Some AI tools can generate visual representations of workflows, tailored to suit different learning styles of employees or stakeholder needs.

In any industry, a strong AI strategy is a flexible one. Viewing AI as a blanket fix or failing to hone in on a differentiated approach will be a recipe for failure for retailers, and one that widens the gap between them and competitors who were smart in their adoption. Holiday shopping may never look the same behind the scenes, and the 2024 season presents an opportunity to lay a solid foundation for the future.