Sign In

Communications of the ACM

Communications of the ACM

Drivers of Price Premium in E-Markets

As a main source of profit generation, product pricing would be an important factor for any Internet retailer. By selling high-quality products with high prices, Internet vendors could generate price premiums and profits. However, online customers enjoy low search cost with the help of Internet technology. Moreover, they often practice comparison shopping: 16% of U.S. Internet users are known to visit one or more comparison shopping Web sites (such as

In response, many Internet retailers have offered untenably low prices to attract customers based on the assumption that customers would buy products from the retailer who offers the lowest price. However, very few Internet retailers survive by adopting these low-price strategies. Instead, many online retailers have filed for bankruptcy as intense price competition prohibited them from generating enough profits.

Yet prices posted on the Internet for similar products are known to vary by as much as 47% across vendors [1]. More interestingly, even price-sensitive customers may not always buy from Internet retailers offering the lowest prices. Kim and Xu found that customers are more salient to trust than price in their purchase decision making at an online store [6]. Smith and Brynjolfsson reported that reputable online bookstores (including,, and command a $1.72 price premium over generic online bookstores [7].

Why do consumers purchase books from even though they may perceive that charges higher prices than other online bookstores for the same books? Why do higher prices have relatively low effect on consumers' purchase decisions in some cases? If an Internet retailer can lower the effect of price on consumers' purchase decisions, it could generate more price premiums and more profits by charging higher prices.

This article aims to identify the drivers of price premium by examining changes in the effect of price perception on customer purchases in electronic markets. Changes in the effect of price perception on customer purchases imply changes in customer price sensitivity—price sensitivity refers to the weight attached to price perception leading to purchase [5]. An Internet retailer can generate price premiums by lowering customers' price sensitivity in their transactions with the retailer.

Back to Top

Generating Price Premium by Lowering Price Sensitivity

Customers tend to encode prices in ways that are meaningful to them instead of remembering the actual price of a product [9]. They compare objective price (the price charged by the current retailer) with reference price (the price charged by another retailer) during shopping, and then encode the outcome as being higher or lower than the reference. Such outcomes drive the price perceptions of customers, which in turn influence their purchase decisions. As a monetary sacrifice, an increase in perceived price lowers utility of the product. This would deter the customer from wanting to purchase the product. In contrast, if an Internet retailer could lessen the weight attached to perceived price in the customer's decision, it could gain surpluses from its customers—what we term price premium. As we aim to examine changes in the effect of perceived price on purchase behavior, we measure the base relationship between perceived price and purchase intention, and test what factors moderate the base relationship.

Given the situation of high price competition and low search cost in electronic markets, finding the drivers of price premium is critical to the sustainability and profitability of online businesses.

From the standpoint of pricing theory, the full price of a product consists of purchase price, cost of search, and cost of a disappointing purchase [4]. Purchase price in the online shopping context refers to monetary payment for the product, including tax and shipping costs. Customers then encode purchase price (the perceived price). However, the effect of perceived price on purchase intention could be moderated by the other two costs: search cost and disappointment cost. This is especially true in the Internet shopping context, where uncertainty and risks are prevalent. With customer deception by Internet retailers becoming increasingly common, the present value of anticipated future losses (such as security threats) may far exceed the monetary benefit (paying a lower price compared to other retailers) of an online transaction. Thus, the full price can be low if the two costs are lowered even though the perceived price is somewhat higher. In such a case, the effect of high perceived price on purchase would be lowered.

Besides the full price of a product, three consumption costs—money, time, and effort—may influence customer purchase decisions [3]. Search cost and disappointment cost of Ehrlich and Fisher [4] can be matched mainly to non-monetary costs (time and effort) while purchase price can be matched to monetary cost (money). Downs posited that most customers regard time and effort costs as more important than monetary cost where low-cost standard products (for example, pencils) are concerned [3]. Since we consider customer price sensitivity as the weight attached to perceived price leading to a customer's purchase, price sensitivity can be moderated by the non-monetary factors.

Based on our literature review, we propose and test four factors that may lower price sensitivity: reputation, switching costs, familiarity, and pleasure. Reputation means a collective representation of the retailer's past actions and results that is held by the public. Switching costs refers to a customer's subjective perception of the one-time costs associated with the process of switching from one retailer to another. Familiarity refers to a customer's degree of understanding about the procedures and technology of transacting with a vendor. Pleasure means the degree to which a customer feels good or pleased with previous transactions with an Internet retailer.

Back to Top

Data Collection

This study employed the survey research methodology. We developed the data collection instrument by adopting existing validated questions. All questions were anchored on a seven-point Likert scale. As the context of this study, we chose an online bookstore that sold about 18,000 books daily. The data for this study was collected over 10 days, during which a banner on the home page of the bookstore publicized the survey and directed respondents to it. A total of 810 valid responses were collected via the Internet survey: 63% of the respondents were female; the mean age was 30; they all had at least one transaction with the vendor.

Back to Top

Findings and Discussion

The table here shows the moderated regression testing results. In addition to their main effects, the results show the four proposed factors exerting significant moderating effects, that is, they could lower price sensitivity. In discussing the first factor, reputation, we note that information asymmetry exists in a customer-vendor relationship, with the customer having limited information about the vendor. A solution to this information problem is for the customer to find a signal about the vendor's characteristics. Reputation is such a signal, allowing the customer to distinguish between good vendors and bad ones. Vendors of good reputation who engage in untrustworthy behavior would ruin their reputation and forfeit the investment they have made in building it. Consequently, the reputation of a vendor enables customers to perceive fewer risks and lower cost of a disappointing purchase when buying from the vendor. Thus, high vendor reputation reduces the weight of perceived price in a transaction with an online vendor.

In discussing the second factor, familiarity, we focus on transaction-related understanding about the use of the vendor's Web site, and of the whole procedure of transaction with the vendor based on previous experience. In an uncertain and risky transaction context, such as over the Internet, familiarity enables customers to perceive higher certainty and lower disappointment cost in their transactions with the vendor. As customers become familiar with the relevant procedures and technology of transactions with a vendor, they would expend less time and effort to search for a product and complete a transaction with the vendor. As Downs posited, most customers regard time and effort costs as more important than monetary cost when purchasing low-cost standard products [3]. Thus, high familiarity reduces the weight of perceived price in a transaction with an online vendor.

The third factor, switching costs, arises from a variety of factors and may be classified along three dimensions: procedural switching costs (setup costs and learning costs), financial switching costs (monetary loss costs), and relational switching costs (psychological or emotional discomfort) [2]. According to this factor, switching from one vendor to another increases the expenditure of time and effort. In the presence of high switching costs, customers may not easily switch to another vendor unless the alternative vendor's price is significantly lower than the current vendor's and the price savings are greater than the switching costs. Thus, high switching costs reduce the weight of perceived price in a transaction with an online vendor.

The fourth factor, pleasure, may be considered a customer's emotional response or feelings to previous transactions with an online vendor. Pleasure from previous satisfactory transactions with the vendor would resolve emotional unsettlement arising from information asymmetry between buyer and seller and from the uncertainties of Internet transactions. Consequently, pleasure enables customers to perceive lower risks and lower cost of a disappointing purchase when buying from the vendor. In addition, pleasure enables customers to perceive emotional value—the utility derived from feelings resulting from the transaction [8]. Thus, a high level of pleasure reduces the weight of perceived price in a transaction with an online vendor.

Using the subgroup analysis procedure, we further analyzed the significant interaction effect noted in the table here to determine the exact nature of its impact. The moderating factors were dichotomized at the median. Perceived price was regressed on purchase intention under conditions of low and high subgroups of each moderator; the figure here shows the graphical representation of the moderating effects. All the graphs depict negative gradients, which show that purchase intention decreases as perceived price increases. Lower perceived price (near 1 on the scale) represents greater monetary savings for customers, which should lead to higher purchase intention. In contrast, higher perceived price (near 7 on the scale) represents greater monetary loss for customers, which should lead to lower purchase intention.

When perceived price is high, customers who perceive a high level of vendor reputation (or switching costs, familiarity, pleasure) have higher purchase intentions compared to customers who perceive a low level of vendor reputation (or switching costs, familiarity, pleasure). Thus, the graphs show that customers who view a vendor as having a high level of each moderator are less sensitive to price perception in their online purchase decisions than those who view the vendor as having a low level of each moderator. By lessening the weight attached to perceived price in a customer decision, each of the four factors could enable an online vendor to generate price premiums.

Back to Top


Given the situation of high price competition and low search cost in electronic markets, finding the drivers of price premium is critical to the sustainability and profitability of online businesses. In this study, we have found four factors that function as drivers of price premium by lowering the price sensitivity of online customers. These four factors not only moderate price sensitivity but also influence purchase intention. However, it is useful to replicate this study across a variety of Internet vendors to cross-validate the robustness of the results in the future. While this study focuses on an Internet bookstore where product quality varies very little, future studies can examine Internet vendors that sell products with larger variation in quality.

With all these precautions, we may draw several implications from the study. First, an Internet retailer may gain surpluses from customers by paying due attention to the drivers of price premium. Many Internet retailers have competed with each other by adopting low price strategies. Yet low price strategies have failed many Internet vendors, causing them to file for bankruptcy over their failure to generate profits. Our results offer Internet retailers a way out of their dilemma by showing customers' purchase decision making is not simply dominated by price perception. Rather, the effect of perceived price on purchase intention is moderated by the perception of each of the four found factors.

Second, to leverage vendor reputation in pricing, an Internet vendor may want to put effort into reputation building, such as leveraging the word-of-mouth effect and the level of advertising.

Third, an Internet vendor should consider Web site design and transaction procedures from the perspective of the customer. The Internet vendor could provide customers with more trial experiences at its Web site. This would allow customers to increase familiarity with the relevant procedures and technology of transactions at the Web site.

Fourth, an Internet vendor should provide customers with pleasurable or satisfactory transaction experiences to increase the mitigating effect of pleasure against price perception for the customer. Indeed, emotion marketing advocates posit that emotion wins customer loyalty ( mag/marketing_emotion_wins_loyalty/). Customer pleasure may arise from a variety factors such as Web site quality, service quality, and on-time delivery.

Finally, an Internet vendor could make it less worthwhile for customers to switch to another vendor. The vendor could increase switching costs for customers by providing customized services and frequency/loyalty programs, and put in place conditions that prevent or discourage customers from switching vendors. Switching costs could enable the vendor to hold its price advantage.

According to a study by McKinsey and Company, a one-percent increase in price produces an average increase in profitability of 7.4%. If sales volume remains constant, retailers can capitalize on high reputation, high switching costs, high familiarity, or high pleasure that their customers perceive, and employ their own price levels against their competitors to generate price premiums. Note, once again, that successful online bookstores command a $1.72 price premium over generic online bookstores. Internet retailers can leverage each of the four factors we have discussed to adjust their prices suitably to increase profitability.

Back to Top


1. Brynjolfsson, E. and Smith, M.D. Frictionless commerce? A comparison of Internet and conventional retailers. Marketing Science 46, 4 (Apr. 2000), 563–585.

2. Burnham, T.A., Frels, J.K., and Mahajan, V. Consumer switching costs: A typology, antecedents, and consequences. Journal of the Academy of Marketing Science 31, 2 (Feb. 2003), 109–126.

3. Downs, A. A theory of consumer efficiency. Journal of Retailing (Spring 1961), 6–12.

4. Ehrlich, I. and Fisher, L. The derived demand for advertising: A theoretical and empirical investigation. The American Economic Review 72, 3 (Mar. 1982), 366–388.

5. Erdem, T., Swait, J., and Louviere, J. The impact of brand credibility on consumer price sensitivity. International Journal of Research in Marketing 19, 1 (Jan. 2002), 1–19.

6. Kim, H.W. and Xu, Y. Internet shopping: Is it a matter of perceived price or trust? In Proceedings of the Twenty-Fifth International Conference on Information Systems (2004), 831–842.

7. Smith, M.D. and Brynjolfsson, E. Consumer decision-making at an Internet shopbot: Brand still matters. The Journal of Industrial Economics 49, 4 (Apr. 2001), 541–558.

8. Sweeney, J.C. and Soutar, G.N. Consumer perceived value: The development of a multiple item scale. Journal of Retailing 77, 2 (Feb. 2001), 203–220.

9. Zeithaml, V.A. Consumer response to in-store price information environments. Journal of Consumer Research: An Interdisciplinary Quarterly 8, 4 (1982), 357–369.

Back to Top


Hee-Woong Kim ( is an assistant professor in the Department of Information Systems at the National University of Singapore.

Yunjie Xu ( is an assistant professor in the Department of Information Systems at the National University of Singapore.

Back to Top


UF1Figure. Graphical representation of moderating effects.

Back to Top


UT1Table. Regression models testing main and interaction effects.

Back to top

©2007 ACM  0001-0782/07/1100  $5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2007 ACM, Inc.


No entries found