The Latest from 360pi

Retail Pricing: Using information to improve profitability of large chain retailers – Part 2

Dr. Stephane Bratu discusses how the world of big data can help brick and mortar retailers in the second part of this 2 part series.

Retail Pricing: Using information to improve profitability of large chain retailers is a 2 part series. Read part 1

How can the world of big data help large brick and mortar retailers to improve their pricing decisions and ultimately their financial performance?

  • By better understanding their customers and better competing against pure online retailers, who often have lower costs, to improve prices, product offering and inventory productivity;
  • By better positioning their prices to compete against online retailers in almost real time, only where it is necessary (sales channel and articles) for the “commodity” products;
  • By taking advantage of captive markets where competition from online is relatively lower (example: products that typically need professional advice in remote locations like Hawaii);
  • By capturing the niche products that have enough demand on a specific area (geographic product differentiation);
  • By improving service to recurring customers by offering applications that will help the customer better navigate the store to find the products he is looking for and tell him how many units are left of the product he is looking for. If the product is not in stock then the customer will be guided to another option, such as going to the closest store that has the inventory or going online to purchase the product.  Another application is real time promotion based on the customer profile.

The key to success is to provide analytics to drive actionable insights into customer behaviors and improve the long term profitability of the firm.  But as retailers are extensively collecting competitors’ prices, the paradigm for retail pricing is actually to get away from price matching and come back to value pricing when there is perceived value of purchasing.  How to add the value to the matching prices?

  • By grouping existing customers into segment profiles by what drives them in the store in the first place to better understand what article prices customers actually care about;
  • By understanding competition, not only on prices but also on product portfolio offering;
  • By understanding service values provided to their customers (shopper-friendly experience) and finally by aligning promotions and pricing strategy to achieve financial objectives and increase foot traffic;
  • By managing new products and minimizing risks of profit loss:  If the retailer wants to introduce a new line of seasonal products, for which he has no solid historical sales data (3)


Life cycle price management solutions have been offered to retailers by application software companies, such as ProfitLogic (Oracle), DemandTec (IBM) or Khimetrics (SAP), to help clients improve their pricing strategies, promotions, markdowns. But more needs to be done:

  • Modeling of customer behavior at the point of sales level:  Clustering the points of sales improve forecast accuracy but overlook each individual store inventory. The granularity of the point-of-sales/SKU pairs make it hard to forecast.  Setting the same price for a SKU on the store cluster can be sub-optimal, particularly if there are significant inventory differences across stores in the same cluster, which is often based on geography.
  • Pricing and product portfolio:  The objective is to put pressure on commodity products’ prices in order to sell more high-margin technology products with added value.  If a retailer manages to do that it will meet his target both in terms of revenue and profit.
  • Relative pricing: pricing of substitute products (good/better/best) by better  understanding  cross elasticity of demand and the utility gap of different product with respect to the price difference.
  • Price testing experiments in real life to experiment and learn from it.  This is called dynamic pricing where the price response function depends on what has been learned through price testing.  There are some challenges in understanding what should be the price of substitute products, as the demand has to be de-seasonalized before being aggregated.
  • Understanding pricing with respect to product life cycle is critical particularly for the short life cycle markets (e.g., electronics).  Indeed, when the new technology arrives it is critical to move demand to this new product by changing the price of the current substitute products.  There is usually more profit to be made in the new technology so it is important to make sure that the demand is present.
  • Pricing and logistics:  Assignment of products to store in coordination with pricing strategy, particularly important for expensive seasonal products (like fashion);
  • Pricing and marketing: Promotion campaign and impact of sales lift (market basket analysis, product affinity, etc.);
  • Pricing and shopper profiles: identify articles that a given customer segment typically purchases and the drivers that brought them to purchase in the first place.
  • Pricing and competitors’ prices: How competitors are pricing the products and where are they located with respect to potential and existing shoppers? Profitability analysis tool that capture price demand elasticity in each market.  But matching competitors’ prices can be dangerous and misleading and cannot be done for some commodity products:  BOGO or product bundle promotions such as “Buy item A to get a free item B” (4).  Such an across the board strategy will erode margins by spiraling-down prices.

The new technologies to mine the internet will offer more transparency for major retailers to better position themselves and better understand their customers.  The danger is to over react and match competitors’ prices every time.  The challenge is therefore to identify the articles that need to be competitive and others that provide a perceived value to customers, because of their availability, and time and cost of being serviced by the competition.



(1) “Wal-Mart steps up its online game with help from stores,” Reuters, Jessica Wohl and Alistair Barr, March 26, 2013.


(3) Felipe Caro and Jérémie Gallien, “Clearance Pricing Optimization for a Fast-Fashion Retailer,” Operations Research, November/December 2012, 60:1404-1422


About the Author