The Price is Right – OR IS IT?
Setting prices is one of the trickiest aspects of CPG retailing. How do you develop a strategy that will deliver a good return but won't turn customers away? Recognizing the challenges inherent in the pricing equation is the first step on the road to optimization, industry experts say.
"One of the biggest challenges is the scale of the number of prices that have to be set. If a grocer has 40,000 items per store and 250 stores, then they have to price 10 million item/store combinations," says Graeme McVie, vice president and general manager of Precima, a company that provides consulting and analytics, execution and measurement, and data management services for retailers and manufacturers.
In addition, customers make purchase decisions in different ways–some buy primarily based on price or value, while others take factors unrelated to price and value into consideration, McVie notes. "And what customers want in one trade area can be vastly different than what shoppers want in a different trade area," he adds.
Another challenge is that pricing is "so dramatically intertwined with product positioning and availability that it can't be done in a vacuum," says Colin Hare, senior vice president and practice lead of growth analytics and solutions at 4i Consulting, a company that provides end-to-end analytical business solutions.
As challenging as the process has become, establishing a comprehensive pricing strategy based on thorough analysis is more important than ever in food retailing.
"Food costs are rising faster than disposable income, so pricing is becoming more critical," Hare says. "What we're seeing with pricing today is that there is limited growth at retail, and to [increase] pricing to get growth isn't sustainable or giving value."
Analytics is a tool retailers can use to examine factors such as Everyday Low Price (EDLP) versus hi-lo strategy, the impact of trade promotions, the competition in a given market area, and price elasticity to help them set prices in an optimal way.
EDLP OR HI-LO
"Should I use EDLP or employ a hi-lo strategy?"
That is a question most retailers struggle with–one that has a different answer depending on the area in which they operate, customer demographics, the time of year and even the specific product category they are setting prices for.
"A few years ago it was pretty clear who was doing EDLP–Walmart and Family Dollar–and hi-lo was the realm of traditional grocers," Hare explains. "Then they swam toward each other. Walmart now selectively discounts more, so it has more of a hi-lo feel; meanwhile, the Krogers of the world are choosing select items to be EDLP. It is muddying that whole dichotomy."
The short answer is to favor an EDLP approach where consumers are more sensitive to changes in everyday prices, and to favor a hi-lo approach where they are more responsive to promotions, McVie says.
However, EDLP isn't necessarily the right approach for everyone. Kurt Jetta, CEO and founder of TABS Group, is not a proponent. "Movement to EDLP is almost never a good strategy," he says. "Consumers rarely respond to the lower prices enough to offset lower margins. The biggest challenge is to resist the allure of going head-to-head against Walmart on everyday pricing unless you are positive that your operating costs are competitive with theirs."
THE IMPACT OF TRADE FUNDS
How much a retailer has to spend to promote a particular item or category plays a big role in the pricing equation.
"If a lot of category sales are generated by items that respond strongly to changes in everyday prices, then the retailer should favor putting more of the trade funds into EDLP, " McVie says. "If a lot of category sales are generated by items that respond strongly to promotions, then the retailer should favor putting more of the trade funds into promotions."
The next step is to determine the appropriate promotional intensity. That involves assessing how frequent and how deep the discounts should be "to generate positive incremental sales and profits as can be determined from the promotional price elasticity generated by the econometric models," McVie explains.
"Deeper is usually better for ad execution," Jetta says. "Sales lifts increase exponentially with discount, not linearly. Thus, higher units more than offset revenue loss from lower prices." Once the optimal promotional discount has been determined, and the promotional volume has been forecast, the retailer can calculate how much of their trade funds to use for promotion.
Hare, however, foresees problems on the horizon where trade promotions are concerned.
"Manufacturers' trade budgets are the biggest budgets companies have, and that number always goes up because retailers demand more–'Give me more money to promote this in my store,'" he explains. "But there are substantial diminishing returns to manufacturers and retailers. They spend less on advertising and product development because all funds are going into trade. That is starting to be a big problem, and in four to five years it will be huge. The lack of innovation will start to become more noticeable."
ELASTICITY AND COMPETITION
Price elasticity is another factor retailers must consider in the quest to optimize prices. It measures the impact every 1 percent raise in a product's price has on volume sold, Hare explains.
A high correlation exists between price elasticity and competitive intensity, according to McVie.
"Where more competitors are present and the shopper has more viable shopping options, there is a tendency for price sensitivity to be higher, and vice versa," he says. "It is important to gather competitors' prices at a very local level and to set competitive rules in the optimization for the specific competitive sets that exist in specific trade areas–there is no point in matching a competitor if the nearest store is 50 miles away."
It is also important to understand that price elasticity is fluid and hinges on several factors.
"For a single item, the price elasticity can vary dramatically from one store to another by up to a factor of four or more," McVie says. "Within a single store, the price elasticity for all the items in a category can also vary dramatically."
The price elasticity of a single item can vary dramatically by time of year, as well. Ice cream, for example, is substantially more price sensitive in summer than in winter because customers buy it more frequently in warm weather, McVie says.
Hare uses the example of sun-care items. "They have low elasticity during summer because customers are going to the pool and are going to buy it. However, the products within that category are highly elastic," he explains.
Hare believes a deeper understanding of elasticity is needed–particularly as it relates to channel switching and the competitive impact outside of the immediate category.
WHAT RETAILERS NEED
To conduct analytics that will lead to optimal pricing, retailers can tap a variety of data:
Sales data. In order to build elasticity models at the store/SKU level, McVie says retailers need two years' worth of past sales data at the store/SKU/week level, as well as the product hierarchy file, the store hierarchy file and the promotional file.
"In order to be able to optimize, you will also need the cost file if you want to include gross margin and the competitive price file," he explains. "And if you want to include the customer or segment perspective, you will need data at the transaction/item/store/customer level for all transactions, items, stores and customers over the past two years. It would also be beneficial to include the customer file."
Syndicated data. According to Hare, standard Nielsen or IRI data can provide a good baseline assessment of promotional success. "Retailers can leverage syndicated data from Nielsen or IRI to understand changes in market share and to provide average and most common market prices," he says.
"Household panel data from Nielsen or IRI is preferable to loyalty card data because the data gives visibility to what consumer do when they are not in your store," Jetta adds. "Loyalty data has more value for operational decisions such as store hours and locations."
Supplier data. According to McVie, retailers do not need data from suppliers to build core econometric models and optimize prices. "The models can be built using the retailer's sales data," he says. "[But] if a supplier is attempting to build models without using the retailer's data, they can sometimes use shipment data as a proxy for the sales data–but this usually only works in a DSD [direct store delivery] environment and not for warehouse products. The more product attributes that are available, the more robust the models will be."