Analytics in grocery
Grocery retailing is becoming increasingly tough.
It is facing sustained pressures in virtually all developed economies as the cost-of-living crisis in the wake of rising inflation squeezes shoppers’ disposable income, while retailers contend with a sharp rise in the cost of labor for store and logistics, energy, and the cost of goods sold (COGS).
While global macroeconomic headwinds caused by the Ukraine-Russia war, the enormous increase in energy prices, and the COVID-19 pandemic, which disrupted supply chains, have impacted many economies, the question for grocery retailers is how to respond to this new business environment.
Grocery store size, according to the US Bureau of Labor Statistics, has been growing steadily since the 1950s, with product assortment also growing from an average of just over 14,000 products in 1980 to about 51,000 products in 2008. This, however, was reduced to about 33,000 by 2018 in the wake of the recession caused by the global financial crisis.
It is tempting for grocers to add more products to the assortment in each category, but each new product will have to compete in an increasingly crowded category, frequently resulting in higher costs across the supply chain that are not fully offset by sales revenue growth.
Grocery retailing, like many other businesses, relies on adequate funding for uninterrupted trading. Profits made in the ’golden quarter’ typically provide the financial breathing space to weather the challenging winter and spring trading seasons.
While cost control can improve the financial health of a retailer, finding areas of material savings that don’t negatively impact the shopper experience is increasingly difficult as many of the ‘low-hanging fruits’ were taken out during the recent COVID-19 pandemic.
Increasing prices for shoppers is also an important part of retailers’ financial survival strategy. However, in grocery in particular, the marketplace is very competitive, with various ’price match’ offers being marketed aggressively to increase shopper loyalty. This reduces the headroom for improving profitability through price inflation.
Several retailers have recently bolstered their funding ability by selling and leasing back logistics assets and freehold stores. However, this does not address the core challenge of changing shoppers’ buying behavior as recessionary pressures increase. The only way grocery retailers can survive this is by adapting to the change in shopper behavior.
Range rationalization informed by data analytics will help them improve profitability and retain shopper loyalty.
With unprecedented volumes of data on their product ranges and target customers at their disposal, grocery retailers will have to harness the power of data to make the right business decisions to drive profitable growth in the current difficult environment.
A two-dimensional model that utilizes product and shopper data will enable grocery retailing executives to determine products driving profitability and those that are both unattractive to customers and make minimal or negative contributions to profitability.
Retailers can make data-driven range-rationalization decisions with a product profitability and customer commitment matrix.
The matrix will determine the true profitability of a product based on absorption costing (cost involved in getting the product to the customer) against the level of commitment that target customer segments have for that product.
Both customer commitment and product profitability can be calculated sufficiently accurately to inform decision-making using advanced data analytics and accessing large-scale business transactional data.
The benefit of this two-dimensional analysis is that it allows grocery retailers to rapidly identify products in the assortment that are neither important to their target customer segments nor contributing sufficiently to cash profitability. Rationalizing these products out of the assortment brings multiple benefits for the retailer and the shopper. These products for rationalizing out of the range will appear in the bottom left quadrant of the matrix. Products that appear in the top left quadrant are candidates for either packaging change, product reformulation, or price optimization, as customers have sustained demand for them. However, their profitability is weak.
Store-keeping unit profitability
Measuring product profitability is considerably easier now that data science allows large-scale data sets to be interrogated.
For the purpose of the matrix mentioned above, product profitability is the cash contribution a product makes to the overall profitability of the retailer. The advent of big data and advanced analytics has transformed the ability of grocers to extract value from their transactional data to accurately attribute costs based on absorption costing methods. The benefit of advanced analytics over traditional cost absorption modeling is that each direct labor hour can be allocated to the relevant product that an employee was moving, receiving, storing, picking, or replenishing. Similarly, costs associated with storage and space can be accurately allocated based on the actual movement of each pallet. The transactional and cost data can be taken from across the inbound supply chain to accurately calculate the cost of placing the product into each store. The revenue and cost elements would include the following:
The breadth and depth of these data sources allow accurate product profitability calculation, providing the horizontal axis data for the product profitability-customer commitment matrix.
Customer commitment can be measured by analyzing customer segments and personas informed with real purchase data.
For the purpose of the matrix, customer commitment is the measure of the importance of the product in the assortment from the perspective of the key customer segments that the retailer is targeting. Grocery retailers who have loyalty cards or loyalty apps will already have purchase history data that can be analyzed for each customer segment. The model is flexible across retail channels (main store, convenience, and online). The factors used to measure customer commitment include:
For retailers with loyalty data, measuring customer commitment is a straightforward process. But it is feasible to create customer segments and interpret EPOS data even without loyalty data through the use of basket data and known personas.
Profitable range consolidation
Data can aid grocery retailers in effectively consolidating product ranges that are profitable and are valued by core customer segments.
Shoppers are curtailing their spending in response to the increasingly challenging macroeconomic environment. In response, grocery retailers must strengthen their funding arrangements to weather a possible downturn while also rapidly adapting to customer requirements. During such a tricky business environment, range rationalization will act as one of the important levers of profitability for grocery retailers.
The profitability matrix approach described will help retailers adopt a data-driven approach to range rationalization. Products that appear in the bottom left quadrant are candidates for removal from the range as their profitability is low, and the retailer’s target customers have a low commitment to these products. The products in the top left quadrant require a different strategy as the customer commitment is high. For these products, the retailer should investigate ways to improve product profitability, including packing changes, product reformulation, and price optimization.
To make range rationalization work, grocery retailers should leverage the scale of their transactional data to calculate product profitability and customer commitment. They can use these inputs to refresh their assortments with the objective of stripping out costs across their supply chain without compromising on their core customer proposition.