In the literature multiple terms are used for the term digital product. Examples of the term are information products, information goods, information services or intangible goods. In this thesis I will stick to the term digital product, because the used definition of customization in subsection 2.2.1 refers to products. In this paragraph I will explain what digital products are, and what their basic characteristics are.
The first subsection describes the characteristics of digital products, in particular their physical nature. The section continues with a subsection about the costs structures of digital products, followed by a section on versioning. These themes are very specific for digital products when compared to physical products, and are therefore relevant for this thesis.
Characteristics of digital products
Digital products do not have a physical form or structure that can be physically consumed (Choi et al., 1997). Digital products are always experience goods, because they have to be used first before its value can be determined (Nezlek and Hidding, 2001). Suppliers of digital products need to rely more on signals that consumers send in, and therefore product customization and discriminatory pricing based on consumer types become essential for digital products (Choi et al., 1997). Examples of digital products are on-line newspapers, magazines, music, education, and searchable databases (Loebbecke, 1999). Digital products are fundamentally different from physical products in three basic characteristics (Nezlek and Hidding, 2001). These characteristics are indestructibility, transmutability and reproducibility (Choi et al., 1997). Some non-digital products share these characteristics, but only to a limited degree. A digital product is the first product that incorporates all these three characteristics.
The first physical characteristic is indestructibility. Indestructibility indicates that there is no wear and tear associated with the use of digital products. The number of uses does not reduce the quality of the product in any way (Choi et al., 1997). A product sold by a producer is equivalent to one offered in the second-hand market, sellers of digital products therefore use strategies to be able to continue selling their first-hand products. Licensing for example is a strategy they use, so the digital product has no value anymore after a period of time. Another strategy that can be used is updating the product, so the value of an earlier version is reduced.
Transmutability is the second physical characteristic of digital products. Transmutability means that digital products are easily perceptible to change, which is paradoxical to the first characteristic (Choi et al., 1997). This allows for customized and personalized products to be created and delivered to the consumer. Product differentiation is therefore a strategy that producers follow (Choi et al., 1997). This can be achieved by customizing and updating the product, and selling the product as interactive services, not just stand-alone products. The Internet makes it easy for digital products to be sold as interactive services. The transmutability raises the whole issue of customized products, individualized pricing, and the proper use of consumer-revealed information (Choi et al., 1997).
The third and most noteworthy physical characteristic of digital products is their reproducibility. It refers to the ease with which copies of digital products may be reproduced, stored, and transferred at low and constant costs (Choi et al., 1997). This characteristic is the most noteworthy because it has consequences for pricing digital products, which is very different from physical products. It means that after the initial investment costs, the marginal costs of production approaches zero.
Seen from an economic perspective, transmutability of digital products appears to work in the producers’ favour, while indestructibility and reproducibility appear to work in the consumers’ favour. The following subsection describes the costs structures that are specific for digital products, including price discrimination.
Digital products are costly to produce but cheap to reproduce, which means that the costs of production are dominated by the first-copy costs (Shapiro and Varian, 1999). The costs of producing the first copy are the fixed costs, and the costs of reproducing are the variable costs. This costs structure leads to substantial economies of scale, the more you produce, the lower your average costs of production. However, fixed costs are also determined by sunk costs. These are costs that are not recoverable if production is halted. Once the first copy of a digital product or information good has been produced, most costs are sunk and can not be recovered (Shapiro and Varian, 1999). Not only the costs are low, but also the time required for producing a copy is marginally low, making it completely unnecessary to store multiple copies of the same product (Luxem, 2000).
For a producer of digital products it is important to keep the fixed, sunk and marginal costs as low as possible. To accomplish this, and actually realize economies of scale, digital products have to be sold to many sellers. For this, there are several strategies to follow. For example, the producer of digital products has to prevent the commoditization of the product (Shapiro and Varian, 1999). When a digital product is produced with high fixed costs, the producer needs to sell the same product multiple times to be able to make a profit. If the digital product becomes a commodity, the producer can not make a profit from it anymore because the price reaches zero. To prevent commoditization of the digital product, the producer can try to protect the product with technical or juridical means.
Because of the costs structure, digital products do not have to be priced according to the production costs. Most physical products have costs almost equal to the marginal costs because these marginal costs are high. Digital products can be priced according to their value for the consumer. The value of a digital product for the consumer increases when the product can be personalized or customized (Shapiro and Varian, 1999). Detailed data on consumer preferences are more abundant for digital products which are used on the Internet, and as a result, consumers obtain a higher degree of satisfaction from customized products (Choi et al., 1997). With the costs structure of digital products, it is even possible to personalize prices as well. If the information products are highly tuned to the consumers’ interests, there is more flexibility in pricing (Shapiro and Varian, 1999).
Information technology allows for fine-grained observation and analysis of consumer behaviour (Varian et al., 2004). This allows for various kinds of marketing strategies that were previously extremely difficult to carry out. For example, a seller can offer prices and goods that are differentiated by individual behaviour and characteristics. Price discrimination is a strategy that can be applied to digital products. It may be an especially attractive strategy for digital products because of high fixed or first-copy costs, and low marginal costs. There are three forms of price discrimination. First-degree price discrimination can be applied when digital products are highly personalized. This is the case for mass customization and personalization (Varian et al., 2004). In the extreme case, with first-degree price discrimination, companies will charge the highest price they can to each consumer for the same product. This is only possible when the seller has collected information of the consumer that their competitors do not have (Varian et al., 2004). Also, because digital products can be customized without much added costs and consumers can be charged independently, first-degree price discrimination can be applied easily for digital products sold over the Internet (Choi et al., 1997). The Internet makes it possible to directly interact with a seller, and as a result consumers can buy products for a personalized price. Second-degree price discrimination is pricing products according to different market segments. Different versions of digital products can be sold at different prices. Third-degree price discrimination refers to selling at different prices to different groups.
Second-degree price discrimination can be achieved by selling digital products in different versions. Versioning refers to offering digital products in different versions for different market segments (Shapiro and Varian, 1999). Versioning is a strategy to achieve economies of scale, which is very suitable for products with low variable costs, such as digital products. Because digital products are transmutable, versioning is an even more attractive strategy. Digital products can be versioned in different ways. One of them is delay. Some digital products are worth more when they are new, so by selling these digital products to different market segments at different times, the product can be sold to more consumers. Another product dimension where versioning can be applied is the user interface, such as versions for advanced users and beginning users. An example is an advanced search interface (Shapiro and Varian, 1999). Other product dimensions where versioning can be applied are on-line and off-line versions, or quality of the product. Different versions can be applied by producing a high-quality version first, and then subtract value from it by removing features or degrading the product (Shapiro and Varian, 1999).
It is not always needed or desired to version along all possible product dimensions. One version does not suffice, because too few market segments can be served. From the consumer perspective, too much versions or too much variety can lead to complexity (Schwartz, 2000; Shapiro and Varian, 1999; Piller et al., 2005). Digital products such as DVDs can be sold as a standard version, and an enhanced collectors’ edition (Varian et al., 2004). If the digital product is subject to network externalities, you may want to restrict the number of versions you offer. Because many digital products are subject to network externalities, they only become valuable once a large number of people are using them. Free versions can therefore be a good way to bring a product’s use up to a critical mass (Shapiro and Varian, 1998; Gallaugher et al., 2001).
Versioning is one way to achieve economies of scale, bundling is another, and can be seen as a special form of versioning (Shapiro and Varian, 1999). Bundling refers to selling two or more distinct goods together at a single price (Adams and Yellen, 1976, in Varian et al., 2004; Choi et al., 1997; Bakos, 1998). This is particularly attractive for digital products since the marginal costs of adding an extra good is negligible. Bundling can improve consumer value, because a bundle of two distinct digital products can be sold for less than the two products sold separately. Consumers that are only interested in one of the products can choose the single product. Bundling digital products is comparable with mass customization. There are companies on the Internet that allow consumers to customize a CD, by letting them choose which songs they would like on it.