How Price Sensitivity Can Make or Break Your Profits |
Historically, many companies have relied on two common pricing strategies. The first “cost plus” pricing—requires firms to make constant adjustments as their costs rise. Certain cost charges such as a rent hike or collective bargaining agreement, however, can impact market participants differently, forcing some firms to raise their prices more than competitors—and above what customers are willing to pay.
A second common strategy, “competitive pricing,” involves setting prices based on the price set by rivals. This approach can be a problem if the pricing doesn’t reflect important differences in what is being offered. In addition, this approach assumes the competition is setting the most effective price for a product.
Both pricing approaches share common failings. They do not incorporate critical information on what customers are willing to pay. Second, they rely heavily on management’s subjective judgment, rather than on data-driven empirical evidence to determine the impact of different pricing levels on demand.
Demand-Based Pricing
Analyzing quantifiable data can refine a firm’s understanding of buyers’ price sensitivity and how it varies by product, channel, geography, and customer segment over time. This, in turn, forms the foundation for “demand-based” pricing.
By quantifying the trade-off between price and volume, as well as incorporating a detailed understanding of the cost-to-serve, companies are able to make adjustments to generate additional volume, additional profit or both.
Leading hospitality companies understand that business travelers are often less price sensitive than leisure travelers. And customers in high-end properties tend to be less sensitive than those in more modest ones. But in many other industries, companies continue to use “cost-plus” or “competitive” pricing. By overlooking “demand-based” approaches, these companies are foregoing a significant profit opportunity.
Estimating Elasticity
The concept of elasticity, a measure of the trade-off between price and volume, forms the foundation for “demand-based” pricing. Estimating elasticity begins with examining two or more years of transaction data and using a statistical analysis to quantify the impact of different prices on volume. But other factors also impact volume, including:
- Overall Economic Conditions: Demand is likely to be higher in good economic times, particularly for products such as raw materials;
- Business-Specific Drivers of Business Volume: Some industries, such as mortgages are highly cyclical, with volumes changing dramatically due to exogenous factors such as interest rates
- Seasonality: Some products, such as gardening tools, have seasonal demand;
- Competitive Prices / Underlying Factor Costs: A 5% increase in the price of gasoline from $3.00 to $3.15 per gallon will have a much different impact on volume if the price of competitive products is $3.00 per gallon than if it is $2.50 or $3.50.
Without explicitly incorporating such factors into the analysis, elasticity estimates can be highly inaccurate and misleading, while failing to capture the true relationship between price and volume.
Case Study: Part I¹
“ABC Auto Finance,” an indirect auto finance company, makes loans in partnership with a national dealer network, with dealers compensated for arranging the financing for consumers’ buying vehicles. Dealers have significant pricing authority, and their compensation varies depending on the “retail” interest rate they are able to negotiate with the consumer. Dealers typically maintain relationships with a number of “wholesale” lenders who provide similar products to ABC Auto Finance. As such, the dealers tend to be quite price sensitive, shifting their business from one lender to another depending on various factors including the “wholesale” interest rate being offered.
ABC Auto Finance’s management believed that an opportunity existed to improve its “wholesale” interest rates. Given that the company originated several billion dollars of loans annually, small changes that produced either higher origination volumes, higher profitability or both would have a large bottom-line impact.
Management did not understand the price sensitivity of its customers—the dealers—well enough to make meaningful decisions. The company launched an effort to analyze this issue in a highly quantitative fashion using new tools that could drill down to a granular level.
The complexity of auto finance derives from the multiple features of a single loan. In the case of this company, rate sheets were developed for each state. These had multiple rates to reflect two approval processes (standard & expedited), seven loan terms, six model years, six customer risk tiers, two loan amount ranges and two loan-to-value bands—in all, more than 100,000 different product combinations. Each offering potentially had a different elasticity which could change over time.
Despite this complexity, ABC completed an initial study in three months. Management found that auto finance elasticity coefficients could differ by more than three times depending on product characteristics, such as the loan-to-value ratio, term and risk tier. This valuable information suggested where prices could be raised with little impact on volume and where they could be lowered slightly to increase volume substantially.
Management immediately incorporated this knowledge into its weekly pricing meetings. While elasticity estimates helped identify price sensitive business segments, they didn’t provide management with explicit recommendations on what prices to offer. Thus, an optimization model was needed.
Measuring Performance
In pricing, optimization is a complex, highly mathematical activity designed to determine which prices will result in the best business performance possible.
To begin, one must determine the performance objective one intends to use. In practice more than one measure may be utilized as management typically makes a tradeoff between two key performance indicators, one volume-oriented and one profit-oriented. The volume measure can track revenues, units sold or market share. The profit measure can capture gross margin, net income, return on equity or one of a myriad of other objectives by which different businesses assess profitability.
The profit-volume tradeoff reflects the willingness of many business managers to give up short-term profits to grow their businesses. (Obviously, they expect to recoup any immediate reduction in earnings over the long-term.)
These performance measures need to take into account any business rules and constraints that will be applied prior to performing the “optimization” process. These might include factors, such as how the maximum price differs across markets, whether prices must be lowered as order size increases, or if prices must end in certain digits such as 99.
In general, the fewer business rules and constraints applied the better. Each time one is added, the potential for performance improvement declines since it will tend to eliminate certain alternate pricing strategies that otherwise would be considered.
Mathematical Optimization
Once performance measures and business rules and constraints have been defined, they can then be combined with elasticity estimates to derive the “optimal” prices. Essentially, optimization involves creating thousands of pricing scenarios, estimating the volume of business that will be generated for each scenario and evaluating the results for each against a set of performance measures.
Looking across all the scenarios, an “efficient frontier” is established, defining the maximum profit that can be achieved for any level of volume. The “efficient frontier” allows management to understand what the profit loss might be if prices are changed to gain another 5% in volume.
Frequently, companies discover that they are operating at a point within the “efficient frontier” rather than along it. Firms which function within the “efficient frontier” use “sub-optimal” prices and so have an opportunity to increase profitability without reducing volume; to increase volume without reducing profitability; or to increase both profitability and volume simultaneously. Moving from within to a point along the “efficient frontier” usually leads to a substantial performance improvement. The optimization process not only indicates what type of improvement is possible, but also provides the specific price changes needed to get from operating sub-optimally to operating optimally.
Case Study: Part II
In the first part of the case study, ABC Auto Finance estimated the elasticity of demand at a granular level for its indirect auto finance business. Now, management wanted to determine how performance would improve using optimization to determine the rates offered to its dealer network.
Management first had to agree on the performance measures that would drive the optimization. Identifying the volume-oriented measure–origination volume–was straightforward. Agreeing on the profit-oriented measure proved more complex as different parts of the organization assessed “profit” in different ways.
In the end, ABC Auto Finance chose a measure that was defined as the present value of the net income generated less the costs for the capital required to support the loan. In addition, ABC Auto Finance identified business rules and constraints requiring, for example, that interest rates on used vehicle loans be no lower than those for new vehicle loans. They also noted rules affecting loans of different terms, loan-to-value ratios and borrower credit scores. The optimization revealed that, for each $1 billion of loans originated, ABC Auto Finance could improve profitability by approximately $1 million keeping the risk profile, customer demographics, and total origination volume intact.
To put this in perspective, in 2006, the average loan originations for the top 10 U.S. indirect auto finance companies was $24 billion.2 So, with similar results, one of these companies could have improved annual profitability by more than $20 million. For ABC Auto Finance the potential profit improvement equaled approximately 20% of its annual profit.
To achieve this, only slightly different rates were needed. And, to the surprise of some internal skeptics, the recommended rates were not all higher than those they had been using. Some actually would be decreased. The rates that were increased were for certain custom¬er loans where there was relatively little price sensitivity (and hence relatively little loss in volume). Those decreased were in other segments with relatively high price sensitivity.
The initial analysis was confirmed in an “in-market” test where a small por¬tion of the business used an optimi¬zation model to determine rates. The performance of that “test” group was compared to that of a carefully selected “control” group to give management confidence that any differences resulted from the new pric¬ing approach rather than other exog¬enous factors. The “in-market” test val¬idated the historical data analysis results, convincing skeptics that price optimiza¬tion could help im¬prove business per¬formance.
Successful pricing programs involve broad change management (not merely a systems enhancement) with a focus on upgrading the skills of those involved in the process and aligning the incentives of the sales force. Cutting corners in these areas isn’t thrifty. One may lower the level of investment, but the benefits will also tend to be reduced. Overall, the infrastructure and other needed invest¬ments will provide a payback within a matter of months not years.
How Demand-Base Pricing Applies to Other Businesses
While this case study involves a specific industry, many businesses share similar characteristics that make elasticity analy¬sis an effective tool for identifying differ¬ences in price sensitivity. Those business characteristics include:
- Substantial revenues (i.e., >$100 MM annually)
- Large number of transactions (i.e., >50,000 annually)
- Complex market characteristics (e.g., purchaser size, geography, industry) that drive pricing decisions
- Multiple distribution channels (e.g., retail stores, direct mail, third-party distributors)
- Large number of competitors
- Frequent price changes
Many retail, manufacturing, distribu¬tion, high-technology and financial ser¬vices businesses have these characteristics and can gain great insight into how to price their products…or, with another stage of analysis, how to truly “optimize” prices.
A company that seeks to estimate cus¬tomer elasticity and optimize its prices is embarking on a journey that will require significant effort. However, the effort can pay large financial dividends, pro¬vide enhanced insights about customers and improve organizational alignment.
Endnotes
- While this case study focuses on a B2B2C (business-to-business-to-consumer) busi¬ness, the approach has been applied in a variety of B2B (business-to-business) and B2C (business-to-consumer) markets.
- Source: Auto Finance Big Wheels 2007 (Royal Media Group); Deloitte analysis.
This article was originally published in the November 2008 edition of The Pricing Advisor, a Professional Pricing Society publication.
