Modeling each customer to optimize business performance
According to a report by industry analyst Forrester, many retailers today still set prices with their gut and a particular margin in mind. The problem is that profits get squeezed because retailers either raise prices too high, suppressing sales, or shortchange themselves with paper thin margins.*
What if a store manager could avoid over- and under-discounting by better understanding the purchasing behavior of each customer? Retailers can examine past behavior for patterns and uncover clues about their customers' future buying decisions. Most retailers already track their customers' shopping habits through loyalty programs. But, many retailers lack the analytical tools to convert this knowledge into an insightful strategy to boost store profits and simultaneously increase customer satisfaction.
Achieving a pricing scenario that attracts customers while optimizing profits for retailers is the objective of the Accenture Personalized Pricing Tool, a prototype developed by Accenture Technology Labs. Embracing the concept of one-to-one marketing, the tool uses advanced analytic techniques to understand each individual customer, rather than appealing broadly to demographic clusters of customers. As a result, stores can take a bold step to transform how they increase customer satisfaction, boost sales, and optimize profits.
How It Works
As a customer enters a store, he or she approaches a kiosk and inserts a loyalty card to receive coupons that are not only tailored to their product preferences and price sensitivities, but also designed to maximize profits per store. A kiosk issues paper coupons or transmits special offers to a personal digital assistant. The offer the customer receives is based on their buying patterns and the retail manager's objectives.
Behind the scenes, the Accenture Personalized Pricing Tool mines data collected in past transactions at a particular store or chain of stores to model each customer's buying patterns. It then simulates how the customer might react to discounted pricing on certain products in the future. This enables a store manager to test various simulations each week—for example, a five compared to a 10 percent discount—and observe the probable results before executing a new strategy. In this way, the store manager is able to balance the relative importance of three goals—profits, sales and customer satisfaction—much as a stereo aficionado adjusts the controls on an equalizer. The relative setting of the three levers determines the optimal results for both the customer and the store.
The Technology
What is revolutionary about the Accenture Personalized Pricing Tool is innovative software that combines, for the first time, three pioneering information insight techniques—advanced data-mining algorithms, agent-based simulations and evolutionary computing.
Advanced data-mining algorithms enable a database to respond to each new shopping visit. Agent-based simulations model each customer's shopping habits and enable the store manager to see on screen how a given promotional strategy will affect customer satisfaction, sales and store profitability. Evolutionary computation is the third element. This technology selects only the most effective solutions to a retailing challenge. Just as in biology, the weaker solutions die off with each new generation. The information evolution, however, takes place in mere minutes, rather than over centuries. Only the highest performing pricing scenarios—those that maximize profits and customer satisfaction—win in this "survival of the fittest" contest.
The Implications
Consumers receive a mind-numbing range of product incentives: coupons in newspapers and magazines, through the mail, online, at stores in the aisles and on the back of receipts. The Personalized Pricing Tool simplifies the experience by giving customers incentives at the point of need for the products they are likely to want to buy.