5 Use Cases for Real-Time AI Analytics
With retail shifting online, it becomes tougher for physical stores to attract customers and increase their sales. Globally, the share of online shopping marks 17% of all retail sales, and by the year 2025, e-commerce is estimated to reach 25%. In Finland, the situation remains slightly different with only 8% of sales coming from e-commerce. Surprisingly, in-store shopping is not falling behind as much as expected. This is majorly due to the slow adaptation of online stores, and the recovery from pandemic with more people wanting to spend time outside their homes. In the US, in-store sales have gone up 13.7% compared to pre-pandemic levels, according to Mastercard SpendingPulse.
However, when it comes to in-store shopping, retailers must remain relevant and find ways to attract customers and make them come back. The way to sweep the board is to focus on omnichannel model, the combination of both online and offline stores.
AI can be used to drive two different functions for retail analytics: improving the customer experience and optimizing the sales.
What offline retail struggles with, is the ability to generate real-time analytics similar to online shopping analytics. AI can be used to drive two different functions for retail analytics: improving the customer experience and optimizing the sales.
Using Real-Time AI Analytics for Improving Customer Experience
According to Sarah Ouakim, the Nordics Market Engagement Lead on Ventures & Open Innovation at Accenture, the change in retail is emerging from new needs and preferences of consumers. They way we shop is changing, with subscription services and home deliveries, for instance. Convenience is key, and getting up from your couch needs a little extra motivation. What consumers are craving, is outstanding customer experience in stores – and a reason to return.
With the help of real-time AI analytics, we can drill into the shopping behaviors of consumers inside the store and gain valuable information about what goes down in the stores before the purchase is made – the data that cannot be gathered with traditional analytics tools.
So, how exactly can you use real-time AI analytics to improve the customer experience?
1. Queue Forecasting and Checkout Line Abandonment
American consumers will abandon a checkout line and leave a store without making a purchase after eight minutes of waiting in a checkout line, states the study conducted by Omnico Group. Their British counterparts are even less patient with a waiting time of seven minutes before walking away. This is where every minute counts.
Consumers will abandon a checkout line and leave a store without making a purchase after eight minutes of waiting in a checkout line.
With real-time AI analytics, we can improve customer service and forecast the formation of possible lines before they actually happen. This can be done, for instance, with an estimation of 15-minute delay from walk-in, and information about soon-to-form queues can be transmitted directly to the staff, who have time to prepare for the situation by opening new cash registers.
Additionally, real-time AI can be used to identify number of customers who walked into the line but abandoned the line without making the purchase.
2. Duration of Purchase Process at the Register
How long does it take for a consumer the go through the purchase process from payment to packing their goods? 77% of Americans are less likely to return to a store where they experienced long checkout lines, according to the Omnico Group study.
By measuring the time spent on the register and bagging area, we can analyze the customer service experience in the store and the efficiency of the checkouts. By comparing these analytics to the data on queues, potential bottlenecks can also be identified, resulting with smoother operations on the checkout areas.
Using Real-Time AI Analytics for Optimizing Sales
3. Monitoring the Performance of Categories
With real-time AI analytics, we can define areas where data is collected for different categories. These can be entire aisles, separate shelf units or, for example, individual shelves horizontally, top shelf vs. bottom shelf. Stores have access to similar data through the sales, however, what they are lacking is the data from the decision-making processes in-store. By analyzing the number of customers and the time they spend at certain locations within the store can be compared to the generated sales. This can be used, for instance, monitoring the performance of a newly introduced product or category.
4. Trends About Non-Purchasing Customers
Non-purchasing customers are a serious point of issue for retailers. Understanding why majority of customers leave empty-handed can help to increase the sales. With real-time AI analytics, this can be done with inflow analytics compared to sales.
In a smaller scale brick-and-mortar stores, real-time AI can additionally be used for analyzing residence times within the stores. By combining data with inflows we can explore times when most visitors who enter also make a purchase decision, and times when visitors just wander around the store based on the inflow analytics.
5. Campaign and Promotion Effectiveness
The perk of online advertising is the amount of data generated; the duration spend on site, click-rates, and so on. In-store campaign performance is not as easily analyzed. When campaign signs or stands are used in a store, we are able to use AI to track the number of customers and the time customers spend around the stands. This data can be then used for the conversion into sales. Campaign performance can be compared to the data from regular shelves, to determine the effect of the ad on the purchase decision.
For stores with multiple locations, it is possible to analyze category performance and campaign effectiveness through A/B testing, too. Different store locations often have slight changes in demographics, which might affect the sales. With real-time AI analytics, data can be collected from multiple locations without time-consuming installations and turned into actual performance insights.
After all, does any of this matter? We are seeing the change, and so have others; according to the Gartner CEO and business executive survey 2021, 95% of retail CEOs are looking to boost their technology to enhance their brand’s in-store and online shopping experiences and retail capability.
About the author
Julia Peltonen is an analyst at Elisa with an in-depth focus on technology solutions, AI, and innovation.