How Is Predictive Analytics Being Used in Grocery?
Predictive analytics – that is, mining data for information to be used in predicting trends and behavior
patterns – is being used by some grocery retailers in different areas. They include category management,
pricing, inventory planning, e-commerce, mobile marketing, among others.
What is the number one area that grocers should focus on with predictive analytics?
The biggest opportunity for grocery retailers is in customizing their shopping experience for each shopper. Whether this is in brick-and-mortar stores or online, finding ways to help each shopper more easily shop for their groceries in the way they want to will provide them with more loyal customers as the grocery shopping world changes. Finding data that can help them identify ways to improve the experience for each shopper will make that shopper much more willing to return to that store for their next trip, and increase the amount they are spending by only shopping at that store. What works for one shopper won’t work for all, so the data is key in unlocking how each individual can be catered to, creating a loyal shopper.
Rich Scamehorn, Chief Research Officer and Co-Founder of InContext Solutions
The first step is to understand the business systems that will provide a singular view of your customer data. This digital core, a full set of centralized data points, is a critical success factor.
Data consolidation and the creation of a digital core will allow grocers to implement custom promotions, enhanced product searches, variable pricing and more. This is just the starting point though. Retailers and their suppliers need a holistic view of their supply chain to respond quickly to consumer demand. While a recent Penton study said that more than 70 percent of grocery retailers surveyed had real-time visibility into their own inventory, it is the holistic supply chain view that is needed. In the long term, whether a purchase is digital or in-store, a tighter supply chain will enable retailers to use predictive insights to alert a supplier of a coming need. And that translates into a better customer experience.
Randy Evins, Sr. Principal IVE Food & Drug at SAP Retail
Pricing is one of the single largest value levers that a grocer can pull to impact their business. Predictive analytics lend themselves very well to the pricing process. Pricing involves understanding a large variety of data elements: price elasticity, seasonality, trend, demand transfer, halo effects, competitor prices and costs/margins. Pricing also involves a large volume of data points - a grocer with 40,000 items per store and 250 stores has one million store/SKU combinations. Pricing has to achieve two major objectives that can often be viewed as being in direct contrast to one another: achieving sales and profit targets while delivering a positive price perception – predictive analytics is ideal for tasks such as this.
In addition, pricing decisions can be taken at a centralized level and can involve a small team of individuals, which reduces the amount of change management that is required in order to have the new process adopted. Category reset pricing takes place on a quarterly or annual basis, which enables the analytics to be employed at key strategic points in time. As a result of these factors any predictive analytics can be used to rapidly set new prices for all items in all stores and can be rapidly deployed through the price hosting system to deliver immediate, tangible and significant value.
Graeme McVie, General Manager and Vice President of Business Development for Precima
Predictive analytics around customer personalization should be a key focus area. Personalization can provide grocery retailers with substantial value across a multitude of business areas. The use of sophisticated models can enhance the customer’s shopping experience, boost sales, increase conversion rates and drive deeper contributions from vendors due to improved returns on their marketing investments.
In addition, predictive analytics can be integrated to improve demand forecasting, inventory availability, merchandising and labor scheduling. The overall customer experience is enhanced by providing customers with relevant offers, and by ensuring that that the products are available when and where customers choose to purchase and take possession of those products.
More evolved companies may be more inclined to add more depth and sophistication to already existing predictive analytics in areas such as product recommendation and also in predicting and preventing attrition of both customers and employees. These companies may also want to focus on emerging Internet of Things opportunities in areas such as predictive equipment failure.
Tony Kleiner, Senior Business Consultant at Teradata
There are many areas for predictive analytics of course but the number one area should be Targeted Promotions and Campaigns. This area would involve the use of predictive analytics to assess -- and fine-tune over time -- which customer segments are most lucrative, sensitive to promotions and best served in ways that they are driven to shop more and more frequently. Not only does predictive analytics have the power to increase customer engagement, it can also garner a larger share of shoppers' grocery budgets. It makes for better-served customers, thus more loyal shoppers. It also allows the grocer to determine and focus on those shopper segments that are most likely to respond to such promotions. This could include digital offers, promotions, online continuity campaigns, loyalty programs, personal pricing and recurring programs that pay.
Reena Kapoor, VP of Product at YOU Technology
In general, the focus should be on the greatest return for the least effort. You can argue that clearance optimization (not typical to grocers) is a means of fixing in-season pricing mistakes, in-season pricing optimization is a means of fixing inventory mistakes, and inventory optimization is a way of addressing poor assortment decisions. The further upstream you get the process right, the less you need to worry about optimizing downstream. What are the upstream processes? Assortment rationalization, macro and micro-space planning - things that part of Category Management. Inventory and pricing decisions all come later.
Ziad Nejmeldeen, Chief Scientist at Infor
It’s loyalty because today’s grocers are in a much different environment than 10-15 years ago, and the reason for that is consumers will shop in a highly competitive market and shop 3-4 grocery stores for different items. Some people go to Trader Joes to buy wine and apps because it’s a specialty item, but then buy water at Target because it’s cheaper, and normal groceries at Whole foods. There isn’t enough loyalty to stores from shoppers today, but a retailer’s understanding of consumer’s mobile data and where they shop can entice them to understand that consumer and why they shop at different places. They can also use that data to retarget them. Understanding someone shops at 5 different stores, you can talk to a consumer before they go to a specific store for an item and get them to your store instead.
Adam Meshekow, EVP of Strategy & National Sales at SITO Mobile