Articles & Papers

The Big Picture

A successful portfolio pricing strategy begins with a broader outlook. Rob Arnett

Marketing Research

Winter 2002

Executive Summary
Economists have long recognized that pricing is one of the most powerful levers available for managing demand. Yet many companies seem to treat pricing decisions as either an afterthought or a tactical detail. By developing a portfolio-level pricing strategy, companies can better understand how their markets operate, exploit strategic alternatives, and position themselves to make better tactical pricing decisions when circumstances dictate.

Companies spend millions of dollars developing, testing, and implementing advertising strategies, sometimes with mixed or even mysterious results. They employ scores of focus groups, copy tests, heavy-up spend tests, test markets, and tracking studies to explore and evaluate better ways to use advertising, even though its effects can be inconsistent.

Pricing, on the other hand, has effects certain enough to be canonized as an economic law-the law of supply and demand. In spite of a level of certainty regarding its general effects, pricing does not generally receive the same attention as other strategic areas like advertising. Instead, corporate planners are stuck in tactical, reactive pricing mode.

Tactical or reactive pricing is usually caused by one of three stimuli. The first is performance. Here, pricing actions happen as a reaction to adverse changes in business performance. Soft demand leads to price decreases designed to prop up sales. The second is profit pressure. Responding to pressure to fatten the bottom line, companies increase prices to improve profit margins. The third stimulus is competition. In this situation, a competitor drops its price, forcing the company to match the change, or raises its price, providing cover for a matching price increase.

In each case, companies adjust their pricing strategy in response to some external stimulus, often without a clear idea of how a given pricing action will affect their own portfolio, much less those of their competitors. A string of unconnected, reactive pricing moves can sometimes lead to adverse long-term consequences. To make matters worse, tactical pricing decisions are frequently made on a single product or segment of the product line, without considering their impact on the portfolio as a whole.

For most business firms, locating and effectively targeting unique market segments is both a reality and a necessity in today's competitive market place. Creative market segmentation strategies usually afford the business organization a strategic advantage over their competition and provide marketing efficiencies that greatly improve customer retention and profitability. If a firm can address its markets by way of a creative new vision of how that market is structured and operates, and can uncover the needs and wants of the segments therein, then it has the opportunity to act on that vision to enhance its own profitability, often at the expense of the competition.

Portfolio-Level Pricing

Taking a broader, more planned view of the pricing process can help companies get out of reactive mode. Most important, examining pricing strategy at the portfolio level encourages firms to look at how all of the products in their portfolio work together to generate company profits. This approach considers pricing not only in the context of competitive products, but acknowledges the potential effect that products within the portfolio have on each other. Cannibalization is only the most obvious of those potential effects.

It's helpful to think of developing a portfolio-level pricing strategy in the context of the marketing strategy planning cycle. (See Figure 1.) The process begins, and is grounded in, the company's goals and objectives and moves through a series of stages to the use of performance metrics to evaluate results and provide input into the company's goals and objectives. Each step has a critical role to play in the development of pricing strategy.

Figure 1


Goals and objectives Goals relate to desired performance outcomes three to five years in the future, while objectives are more specific, highly measurable outcomes for the next 12 months. It's crucial to understand from the goals and objectives what key outcomes pricing should bring about at the portfolio level. It's also critical to understand how the products in the portfolio are managed within the organization. It's important to address the following questions:

  • Is the company looking to maximize revenues, volume/share, or profit? In all likelihood a pricing strategy that maximizes one of these outcomes will have an adverse effect on at least one of the other two.
  • At what level should this maximization take place? Corporate? Product level? Brand level?
  • Do different business units manage individual products?
  • Are individual products treated as separate profit centers?
  • How does this management structure support the achievement of corporate performance goals?
  • What is the relationship between "corporate goals" and "business unit goals?"

Resource allocation. Developing a portfolio pricing strategy will require resources for research, testing, analysis, and strategy formation. A new strategy will also require resources for implementation. Pricing isn't the only tool available to your company to achieve its goals and objectives, even just within the marketing realm. Before developing a portfolio pricing strategy, it's important to commit the resources necessary to see the task through to company-wide implementation.

This is where many pricing strategy efforts are stillborn. After considering the costs involved (and perhaps not adequately weighing the benefits), many companies suddenly decide their pricing is fine the way it is and move on to hotter fires.

Market assessment. This is where the homework begins. A sound pricing strategy depends on a thorough understanding of the competitive environment, the existing product and brand space, and the current consumer mindset.

A key consideration in studying the competitive environment, product, and brand space is to understand how (or if) the brands in the portfolio work together to create value. The consumer mindset not only includes perceptions, but also how they derive value from the products in that market. While this last point may seem obvious for some categories like cars or food, the functional benefits derived (i.e., transportation and nourishment) don't provide the complete picture. Secondary benefits such as image and convenience are bigger drivers of perceived value, and therefore pricing.

Opportunity analysis. Doing homework in the market assessment phase helps identify pricing opportunities. Each opportunity needs to be evaluated against the potential to support the company goals and objectives across the entire portfolio. One product opportunity may shift volume, revenues, or profits from one end of the portfolio to the other, without growing the portfolio itself.

Target evaluation and selection. Marketplace success requires focused effort toward specific groups of customers with similar needs. Similarly, a successful pricing strategy needs to be focused at the segment or target level. Because our objective is to develop a portfolio pricing strategy, the entire strategy may address the majority of the market or the whole thing. Still it must operate at the segment level.

Marketing-mix optimization. Portfolio pricing strategy by definition focuses on one key aspect of marketing-pricing. Often, though, pricing can't be changed without some impact on other aspects of the marketing mix such as the product itself, advertising and promotion support, and distribution. Because we're working at the portfolio level, pricing changes on one brand may require adjustments in pricing or other marketing-mix elements on other brands in the portfolio.

A simple example from consumer packaged goods (CPG) relates to changing prices vs. changing the amount of product sold. Many CPG marketers have found that it's safer (in the short run) to "weight-out" the product and leave the price point alone than to change the price itself. In numerous services categories, there may be trade-offs between price and service time, or price and service features.

Framing these types of trade-offs is at the heart of the market assessment phase of the process, in essence understanding and defining "what is." Marketing-mix optimization is about using that knowledge to test and explore "what could be."

The technology and processes for exploring "what could be" using research have come a long way in the last five to 10 years. Choice modeling (and sometimes conjoint) has matured through application into a robust tool for simulation and forecasting.

Marketing implementation. No matter how well the previous steps are followed, the success of any portfolio pricing strategy will depend on the quality (and speed) of its execution within the company and in the market. The number of areas within a company potentially affected by a pricing change is staggering (i.e., marketing, sales, finance, accounting, customer service, and information services).

External implementation is equally complex, considering it will affect every customer and consumer, as well as prospective customers and consumers. Some companies and industries, such as airlines and technology firms, make price changes seem easy, but they're not. And mistakes can cost lots of money.

Performance metrics. Developing a portfolio pricing strategy and managing its implementation and performance is an ongoing process. It requires continuous performance feedback to (1) compare its performance back to the original corporate goals and objectives across the entire portfolio, (2) provide information for making mid-course corrections, and (3) build the foundation for further strategy development and refinement. Wise companies begin designing their performance metrics early in the process, so they'll be well-positioned to support all three of these objectives.

Key Elements of Success

Market assessment. Pricing is all about understanding what drives value in a category and industry, and research is the only way to truly understand the levers of value and the factors that drive brand choice. Without this, it's hard to know where to begin developing a pricing strategy. One way to develop this understanding is through the use of the brand value model, developed by Park and Srinivasan in 1994 and applied by Bill Neal and others at SDR Consulting. The core of the brand value concept is that consumers' perceptions of value drive their choice of brands.

Value perceptions can be decomposed into the following components:

  • Price, including payment terms, fees, and promotions that impact price perceptions
  • Product features, covering all the tangible aspects of the product itself
  • Channel features, which relate to how the consumer acquires the product or service
  • Brand equity, which is really the value placed on the brand by consumers over beyond the values due to price, product features, and channel features

With this view of brand value, it's easy to see how brands or offerings with very different components could have equal or nearly equal brand value. This opens up a greatly expanded view of pricing and the levers available in developing pricing strategy. This view also drives a deeper understanding of the contribution of each brand in the portfolio to the overall company value proposition.

Fortunately for researchers, this type of structure is relatively straightforward to uncover using trade-off methods (e.g., choice or conjoint). Using conjoint as an example, the design would include a number of relevant attributes representing each of the four components listed in Figure 2. The utilities derived from the conjoint model serve as the building blocks for estimating the contribution of each of the four components to brand value. The contribution of price would be the sum of the utilities for the price-related attributes.

Figure 2


One virtue of this type of approach is that it's very amenable to segmentation. Segmentation will produce groups of consumers, each with similar brand value structures. There's usually a price-oriented segment and may be a brand-loyal segment, a channel-loyal segment, and so on. The approach can be adapted to an existing segmentation (e.g., channel-based segmentation).

Matching each brand in the portfolio to the most appropriate value segment is key to developing a compelling strategy. When married to the appropriate segment, brands can fine-tune their value proposition for that segment, which tends to enhance the overall profitability of the portfolio. This type of approach provides the type and depth of information to understand the dynamics of the market, and to identify opportunities for strategic pricing actions.

Marketing-mix optimization. Armed with the knowledge of the market's brand value structure, we're in an excellent position to develop and test alternative strategies for delivering value to consumers and profit to shareholders. Perhaps the best currently available tool for simultaneously exploring and testing alternative pricing strategies is choice-based forecasting.

The foundation of choice-based forecasting is a carefully designed choice experiment, where key elements of the brand value equation (including price) are systematically varied and presented to respondents as a series of scenarios. Respondents choose the most appealing alternative from a set of competitive alternatives, or allocate a set of purchases over the competitive set. Results of these choices over a sample of consumers are then aggregated to produce a model of how the marketplace will respond to alternative portfolio pricing strategies.

Key advantages of choice modeling are the ability to directly incorporate competitive context and response into the model, and to simulate hundreds or even thousands of market scenarios from the results of one study. Most important for this application, choice modeling allows us to look at the impact of any pricing action or change in the value equation across the entire portfolio.

For all the strengths of choice modeling, there are limitations when used alone. Choice models tend to overstate price sensitivity because respondents can see prices changing relative to each other, and they learn to play along. They also overstate demand because they're based on "perfect world" scenarios where all buyers have perfect information about choices, and all choices are available. Choice models also tend to overstate the impact of feature changes for the same reasons. The key to accurate choice-based forecasting is calibration, using data external to the model.

Impact of marketing activity on awareness and availability. The modeling activities described to this point essentially estimate demand for a service (i.e., its sales potential given universal awareness and availability among potential buyers). Real-world marketers never achieve universal marketing reach, so they must estimate marketing reach to convert demand into a sales forecast. The best approach is to actually develop one or more marketing plans and estimate awareness and distribution based on these plans. Short of this level of rigor, reasonable assumptions about awareness and distribution can frequently be made based on history.

Information on current business performance and share. To make sure the model reflects real world sales, volume, and share levels, it's important to calibrate the choice model to current business performance data. Results of the model will then closely approximate expected business performance under a given portfolio pricing scenario. Using these two external sources, choice modeling becomes a dynamic tool for forecasting, while dampening or eliminating its sources of overstatement.

One final piece of external data that must be incorporated into the model is cost or margin information. With this information embedded in the simulator, companies must look beyond the revenue and volume implications of a particular portfolio pricing strategy to include its profit implications as well. This provides a critical link back to the company goals and objectives.

With a properly calibrated choice-based forecasting tool, companies can test and simulate business results for scores of strategy alternatives in a confidential and risk-free environment. These forecasts provide an important mechanism not only for exploring and choosing among strategic options, but also for setting organizational expectations for business performance based on the chosen alternative.

A recent study conducted by SDR for one of our clients illustrated both the value of choice modeling and the importance of studying pricing effects at the portfolio level. The client sells a product that is currently available at grocery stores, drugstores, and mass merchandisers-except Wal-Mart. Wal-Mart sells a private label version of our clients' product that lines up with every product in their portfolio, plus a low-end, entry-level version. After suffering share losses to Wal-Mart over an extended period, our client decided to test the introduction of an entry-level product at a variety of price points using choice modeling.

Not surprisingly, the entry-level product did a good job of attracting price-sensitive buyers back to food, drug, and competitive mass merchandise outlets. And the lower the price, the better it worked. The unexpected outcome is that the company lost volume on nearly every other line item in the portfolio, at higher margins than the entry-level product. Launching the product would have produced a huge financial black eye.

At one level, this outcome was the result of the awareness artifact of choice modeling. In the course of doing the choice exercise, consumers became aware of the pricing not only for the entry-level product, but all of the products at both the client's outlets and at Wal-Mart. Other products in our client's portfolio were priced anywhere from 20% to 60% higher than the corresponding product at Wal-Mart.

Adding a competitively priced entry-level product didn't help, but rather called attention to the rather large price gaps elsewhere in the portfolio. Consumers then did what rational consumers do: They moved in droves to the store with stronger brand value.

The choice model both replicated and explained, in fast motion, a process that had bedeviled our client. By calling attention to the price gaps across the line, it forced them to see where the real problem lay, and to begin work on rationalizing the pricing of the entire portfolio.

Performance metrics. Some of the key metrics necessary for measuring the success of a portfolio pricing strategy are pretty elemental: sales, volume, share, and profit. No pricing strategy can be successful without moving these four measures in the desired direction.

Keep in mind the desired, or at least expected, direction may not be upward for all four metrics. Goals and objectives again dictate the trade-offs that may need to be made. A goal to achieve the leading share in the category may entail short-term profit sacrifices. Conversely, a portfolio pricing strategy designed to improve profitability could dampen sales. The nature and level of the trade-offs that need to be made can be identified in the marketing-mix optimization stage, but it's important to carry through expectations created at that stage to the measurement of in-market performance. Nothing can derail a portfolio pricing process like a preventable mismatch between expectations and performance.

But while these four measures are necessary, they're not entirely sufficient. One important drawback of all three is that they're backward-looking. By the time you find out that your sales and profits aren't on target, your pricing may have been off kilter for some time.

Pricing isn't the kind of thing that typically delights customers and consumers, but it does have the potential to be a serious disaster. If switching costs are high, the product has a dominant market position, or the current competition is fragmented and unable to respond, then the effects may temporarily go unnoticed in business results (sales and profits).

However, in the current marketplace, deficiencies in the delivery of value to consumers rarely go unpunished in the long run. Markets are extremely efficient over time in identifying and exploiting opportunities to deliver greater value and steal customers from unwary competitors and award them to those with an ear to the ground. This is why a customer satisfaction measurement system can be helpful in spotting pricing issues before they become problems-assuming you're asking the right questions. (See sidebar at bottom of page.)

Overcoming Barriers

As with most new initiatives, the two greatest barriers to implementing the portfolio planning process are commitment and resources. Commitment is the larger hurdle because a company's thinking about pricing tends to become ingrained or even dogmatic. Overcoming this inertia is a big challenge.

Unfortunately the stimulus for rethinking a company's approach to pricing may not provide the most fertile environment for a planned, reasoned approach to strategy development. An intense competitive threat, for example, can certainly shake the status quo, but it doesn't facilitate long-term thinking. The best time to get started is now, before a threat materializes. That way your company will know how to handle the threat when it arrives.

  SIDEBAR: Measuring Customer Satisfaction with Pricing

A solid portfolio pricing strategy pays close attention to the following customer satisfaction measures:

Overall value for the money. Attitudinally, this is where the rubber meets the road on pricing strategy. A perceived imbalance in price paid vs. value derived is likely to show up first in a decline in value ratings following a pricing action.

Overall rating of the product or service. Price increases tend to raise the bar consumers use to evaluate the products they buy. If we go back to the brand value model, it's clear that if price goes up, one of the other components-brand equity, product features, or channel features-will have to pick up the slack. Otherwise consumers' perception of brand value will suffer.

Ratings of non-price items related to value. For diagnostic purposes it can be helpful to collect more specific and actionable information on the types of issues consumers may be having with the products in your portfolio. Sometimes the symptoms of value problems may present themselves in unexpected ways, like dissatisfaction with the value or frequency of discounts and promotions.

As the planning cycle moves from performance metrics into the next round of setting goals and objectives, the information collected in this final stage becomes a critical element of the next round of planning and strategy development.

Additional Reading

Park, Chan Su and Srinivasan (1994), "A Survey-Based Method for Measuring and Understanding Brand Equity and Its Extendibility," Journal of Marketing Research, 31 (May), 271-288.