How Retailers Use AI, Mapping to Boost ROI on Store Remodels
Walmart, Walgreens, and Sephora are all using artificial intelligence technology to improve the retail experience. While the majority of use cases for AI in retail have focused on enhancing the shopping experience for customers, forward-thinking analytics firms are innovating and developing new uses for their existing AI technology.
The analytics firm Fractal Analytics is pushing forward in the retail space with its own solution that relies on AI to forecast the cost of retail store remodels, as well as determine the ROI from large-scale renovation projects. Although Fractal works solely with Fortune 500 companies, the solutions it is developing could be adopted more broadly throughout the retail space.
Retailers nationwide will spend billions of dollars on remodeling this year. Target alone has spent hundreds of millions on remodeling projects since 2017. The company is on track to remodel 1,100 of its fleet of 1,800 stores by the end of 2020, with upgrades that include boutique-style layouts for clothing and accessories, special counters for beauty products, and wider spaces for home furnishings.
Target isn’t the only retailer spending big on store remodeling. Walmart is also making major investments in its retail spaces, as it works to widen aisles, add more self-checkout kiosks, and improve the lighting in its grocery departments. Walmart’s plans to remodel hundreds of stores in the coming years is part of an $11 billion investment strategy. Many of the store remodels are taking place in Florida and Texas.
The team at Fractal believes that artificial intelligence could help retailers, like Target, Walmart, and others, forecast the cost of their store remodels, which would ultimately lead to greater ROI.
“Retailers look towards renovation-remodel to drive incremental revenue for their stores,” says Bhaskar Roy, client partner for retail at Fractal Analytics.
Key considerations that Roy says can influence store remodels include the reason behind the renovation, or remodel, and the elements of the remodel that will generate the greatest ROI. The most common reason Roy sees retailers remodeling stores is to refresh their brand presence in the market, but expanding store layouts and changes in company-wide strategy can sometimes play a role in these decisions, as well.
“Ultimately [it] boils down to a decision driven by competition,” Roy says.
Different AI algorithms can be used to understand future demand and sales in a robust manner, while also helping analytics firms like Fractal arrive at various thresholds to apply for driving better decision making for remodels. It’s not uncommon for retailers to test store renovations before making major changes, but Roy says the most effective way of testing a remodeling change is to perform an in-market test.
Many of Fractal’s retail customers invest millions of dollars in remodeling, and those types of capital expenditure heavy changes call for accurate testing to measure the impact of change.
“Without such measurements at a low scale before [a] larger roll-out, such investments can be a risky affair,” Roy says. “Our customers have successfully used Trial Run (Fractal’s test management product) to test a variety of store remodeling ideas.”
Using Fractal’s Trial Run product, retailers are able to test things like whether making changes to fixtures and shelves will drive increased sales. They can also analyze how sales volume will be impacted by widening aisles or by adding large counters for product specialists to interact with customers.
In addition to store-wide remodeling projects, Roy says the increasing availability of location and mapping technologies is helping retailers create temporary store remodels that are tied for specific events, like Black Friday.
“Mapping [and] location technologies are being used to understand people movements patterns in and around the store,” Roy says, “[and] to serve as critical data points in the store reno-remodel decisioning process.”
Stephanie Miles is a senior editor at Street Fight.