Ad Pricing by Multi-Channel Platforms: How to Make Viewers and Advertisers Prefer the Same Channel?

Abstract

Ad-financed TV channels are two-sided platforms where media houses provide communication from advertisers to viewers. Most media houses air several channels, some of which are particularly valuable to advertisers. At first glance, one might expect the ad volumes to be highest for the channels that are the advertisers' favorites. However, a crucial management challenge for media houses is to ensure that viewers go where the potential for raising advertising revenue is greatest. Because viewers dislike ads, we show that this implies that advertising volumes will be relatively low (and advertising prices relatively high) in such channels. Indeed, other things equal, the ad volume in a channel is inversely related to its attractiveness to the advertising market. Only if the costs of using alternative tools to attract viewers to the advertisers' favorite channels are sufficiently small will the advertising volume in channels with high demand for ads be larger than in channels with low demand for ads.

Notes

¹This feature of a two-sided market is also found in other ad-financed media markets, where the platform provider links the audience and the advertisers, such as newspapers, social networks (like Facebook®), and search engines (Google™ links advertisers and searchers). See, for example, Rysman (2009) and Parker and van Alstyne (2005) for more examples.

²The most common price structure in the TV market is one where the viewers pay a fixed fee for accessing a channel (or a bundle of channels) independent of actual viewing time. Pay-per-view TV still has a relatively small share of the market.

³In other media markets, consumers may not dislike ads. Kaiser and Wright (2006), for instance, showed that readers of women's magazines appreciate ads. Rysman (2004) analyzed the market for Yellow Pages directories, and he found that consumers' utility increases with the number of pages with ads.

⁴ Eisenmann, Parker, and van Alstyne (2006) provided a guide to business strategies in two-sided markets.

⁵When advertisers' preferences have an impact on the quality, this obviously raises welfare issues. Anderson and Gabzewicz (2006) discussed how advertisers' preferences may distort the newspaper content away from the readers' preferences. We do not address welfare issues in this article.

⁶Utility function (1) is due to Shubik and Levitan (1980), and is a modification of the standard quadratic utility function. Under the standard quadratic utility function, a change in the parameter s would affect both the substitutability between the goods and the size of the market (see, e.g., McGuire & Staelin, 1983). This is not the case with the Shubik and Levitan utility function, where s is a unique measure of channel substitutability. Our qualitative results would also go through with the standard quadratic utility function, but then an increase in s would both reduce the size of the market and increase the substitutability. This makes it more difficult to perform comparative statics. See Motta (2004) for a general discussion of the advantages of the Shubik and Levitan utility function over the standard quadratic utility function. Specific applications of the Shubik and Levitan utility function are, for instance, Shaffer (1991) and Foros, Hagen, and Kind (2009).

⁷Note that the absolute value of these effects is increasing in s. This captures the fact that the better substitutes the public perceives the channels to be, the more willing they are to shift from a channel with a high advertising level to a channel with a low advertising level.