MARKETING INFORMATION - A COMPETITIVE ANALYSIS

Selling information that is later used in decision making constitutes an increasingly important business in modern economies (Jensen 1991). Information is sold under a large variety of forms: industry reports, consulting services, database access, and/or professional opinions giv en by medical, engineering, accounting/financial, and legal profession als, among others. This paper is the first attempt in marketing to stu dy competition in the rapidly emerging information sector. Specificall y, we are interested in answering the following questions: (1) Is comp etition fundamentally different when competing firms sell information rather than traditional goods and services, and-if yes-why? (2) What a re the implications of such differences for decision makers (marketers and regulators)? (3) Can we explain some of the observed marketing st rategies in the information industry? As such, the audience of the pap er includes academics as well as professionals who are interested in u nderstanding what is specific about competition in information markets . Familiarity with the practical implications of such differences and understanding of the mechanisms that drive them is essential for those who are faced with the problem of marketing information. To answer th e above research questions we build a simple game-theoretic model that consists of two firms selling information to a population of consumer s who are heterogeneous in their willingness to pay for the quality of information. The most important features of the model are the followi ng. Information products sold by the two firms are modeled as random d raws from two normal distributions having equal mean. The variances of these distributions and their correlatedness constitute the product-a ttribute space, which is assumed to be common knowledge. Consumers are interested in assessing the mean of the distributions and to do so th ey can buy the sample from any of the firms or they can buy both sampl es and combine them to obtain a more accurate estimate. Quality of inf ormation is linked to the accuracy of consumers' estimate of the mean which in turn is influenced by the accuracy of each sample as well as by their correlatedness. Consumers' utility depends on the quality of information they purchased, on their inherent utility for quality (tas te), and on the total price they paid to acquire information. Knowing consumer preferences, firms simultaneously price their information pro ducts. The main finding of the paper is that information markets face unique competitive structures. In particular, the qualitative nature o f competition changes depending on basic product characteristics. Whil e traditional products and services compete either as substitutes or a s complements in the relevant product-attribute space, information may be one or the other, depending on its position within the same produc t-attribute space. Said differently, the nature of competition changes qualitatively with a continuous change in basic product-attribute lev els. The intuition behind this finding is the following. When purchasi ng information, consumers facing important decisions may find it benef icial to purchase from several information sellers. This is more likel y to happen when the reliability of information is low and the sources of information are independent. Under such conditions information pro ducts tend to be complements and, as a result, competition between sel lers is relatively mild. In the opposite case, when information is rel iable and/or sellers' sources are highly correlated, consumers are sat isfied after consulting a single source. In this case, information pro ducts are substitutes and sellers tend to undercut one another's price s to induce consumers to choose their brand. Understanding this discon tinuity in competitive structures has important implications for decis ion makers as very different strategies are optimal under different pr oduct characteristics. Under substitution, traditional strategies to a void competition (e.g., differentiation) are recommended. When the com peting products' reliability is generally low (they are complements) f irms are better off accommodating competition. In fact, we find that a firm may benefit from ''inviting'' a competitor. Finally, our finding s are also important for regulators of information markets. As the lit erature on complementarity suggests, price fixing agreements between f irms offering complementary products may benefit firms as well as cons umers.