قیمت گذاری خدمات تبلیغات آنلاین و قرارداد گزینه ای
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|1846||2011||11 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Electronic Commerce Research and Applications, Volume 10, Issue 1, January–February 2011, Pages 38–48
For the Internet advertisement market, we consider a contract problem between advertisers and publishers. Among several ways of pricing online advertisements, the methods based on cost-per-impression (CPM) and cost-per-click (CPC) are the two most popular. The CPC fee is proportional to the click-through rate (CTR), which is uncertain and makes decisions of advertisers and publishers difficult. In this paper, we suggest a hybrid pricing scheme: advertisers pay the minimum of CPM and CPC fees by purchasing an option from publishers. To determine the option price, we consider a Nash bargaining game for negotiation between an advertiser and a publisher and provide the solution. Further, we show that such option contracts will help the advertiser avoid high cost and the publisher generate more revenue. The option contract will also improve the contract feasibility, compared to CPM and CPC.
Over the last decade, the Internet has emerged as an important medium for advertising. According to a recent report of the Interactive Advertising Bureau (2008), over $23 billion has been spent in the US market alone. For the first 14 years of Internet advertising history after it first was introduced in 1995, the market increased by more than 400% in terms of total revenue. In 2008, the Internet advertising market was the third largest in the US after only television channels (including national, local and cable stations) and newspapers. The American Press Institute (2009) reported that the Internet advertising market took only 13 years to reach $20 billion in revenue, while newspapers took 127 years in the US. Without doubt, the Internet is the fastest growing marketing medium in history. The Interactive Advertising Bureau (2008) also reported that search-based keyword advertising achieved 45% in revenue share in 2008, while the second on the list was display banner advertising with a 20% revenue share. Retailers and financial service providers spent 22% and 13% of the total revenue, respectively, which were the top two expenditures in the market. Although the marketing media have been growing very fast, there is still much confusion over how advertisement publishers (e.g., web content providers) should charge for and how advertisers (e.g., retailers) should pay for their advertisement campaigns. When the market was first introduced, the traditional cost-per-impression method was used. It charges a fixed cost for a given number of displays of banner ads. As the biggest online advertisers are retailers, the actual benefit of advertisement campaigns usually comes from visits of customers to and purchases from the website of advertisers. In such cases, the number of displays does not reflect the benefit of advertising. Many advertisers believe that it is not reasonable to pay for advertisements that generate no value. As the number of clicks likely reflects advertising effects, later in the history of the Internet marketing, payment based on the number of clicks has become very popular. This method is called cost-per-click, or CPC, while the traditional method based on the number of display is often called cost-per-mille, or CPM (mille is a Latin word meaning thousand). The CPC is possible, because the exact number of clicks is trackable due to the nature of information technology (Hoffman and Novak 2000). The CPC is one of several performance-based payment methods. Among other such methods, the cost-per-action, or CPA method is popular. This is based on the number of observable actions, for example, purchases, subscriptions, membership registrations, etc. In 2008, CPC pricing and other performance-based pricing made up 57% of the total revenue, while CPM pricing made up 39%. The CPC pricing is particularly popular in search-based advertising, while the CPM pricing is popular in banner display advertising. The market for CPA pricing is not yet mature, because the number of actions is neither controllable nor tractable by the publisher. In CPC contracts, the click-through rate (CTR), which is the ratio of the number of clicks to the number of displays, plays an important role. The advertiser receives more marginal benefit from advertising when the CTR is higher. Although in CPC contracts the advertiser pays an advertising fee based on the number of clicks, which potentially reflects the actual advertising benefit, there is a significant drawback in CPC pricing. Click-frauds, which do not lead to purchases or subscriptions, will increase the CTR drastically without contributing to the profit of the advertiser. Many CPC contract providing publishers and agencies maintain click-fraud detection systems, but there is no perfect detection technology. Even without considering click-frauds, the CTR is never known a priori, hence both publishers and advertisers will be uncertain about it. Lahaie et al. (2007) noted that CTR can fluctuate dramatically even over small periods of time. In practice, if the same advertisement has been displayed by the same publisher, the historical estimate of CTR is used, while a forecast is used when the advertisement is completely new. As we noted earlier, different types of contract schemes have been invented to generate more revenues or extract more advertisers’ interests from the publisher’s perspective. However, among those contracts, an advertiser has to select only one of them. As mentioned, since one contract can be more beneficial than the others, the contract choice might be crucial to the publisher’s revenue. For example, suppose that an advertiser selects a CPC contract before she advertises when she is uncertain about the CTR. After its ads appear, if the CTR turns out to be high, then the advertisers might wish she had contracted to use CPM instead, since this will cost less. In this paper, we propose a risk management method to deal with the uncertainty about the CTR and analyze the potential of this method to increase the likelihood of contract agreement. As a way to hedge against risk, we consider an instrument for the minimum price guarantee for the advertiser, with which she can choose between the CPM and CPC pricing after the CTR is realized. Of course the advertiser will choose the minimum of the CPM and CPC fees. To earn this privilege, the advertiser must pay an additional fee. Obviously if the additional fee is too high, the advertiser will not consider making the contract with the minimum price guarantee. The additional fee will be determined by a negotiation between the advertiser and the publisher. We call this contract type with the minimum price guarantee an option contract and call the additional fee for the guarantee the option price, following the theory of financial options (Wilmott et al. 1993). To determine the option price, we use an approach that is risk-neutral to both contract parties based on our analysis of a Nash bargaining game (Nash 1950). Furthermore, we investigate the potential of the option contract as a new hybrid pricing scheme. We make an assumption that the advertisement effect depends only on the number of clicks, which is not restrictive when we consider that the largest group of advertisers is retailers. In summary, the research objective of this paper is to propose a novel contract, option, and then investigate the potential of the contract. The paper is organized as follows. In Section 2, we review the literature regarding online advertisements involving CPM and CPC contracts. We also discuss option contracts that have been suggested in other service industries. We define the option contract and formulate the option pricing model with the other two popular pricing contracts in Section 3. We show that the option contract has a better possibility of contract agreement compared to the standalone CPM and CPC contracts in Section 4. We provide managerial insights from the option contract in Section 5, and study the problem numerically in Section 6. We conclude in Section 7.
نتیجه گیری انگلیسی
In this paper, we proposed a new type of option contract for online advertising, and evaluated its properties. In Internet advertising, the cost-per-impression (CPM) and the cost-per-click (CPC) are widely used as the basis for contracting. However, because of uncertainty in click-through rates (CTR), one contract may cost more than the other to an advertiser or bring more revenue to a publisher and vice versa. For this reason, the selection of contract can be a crucial decision making issue to both publishers and advertisers. In this context, publishers and advertisers are willing to hedge the uncertainty of future advertising cost, as a means to achieve higher revenue. Thus, this research suggests an option to pay the minimum of CPM and CPC fees as a way to avoid the ambiguity of the contract type choice. The advertiser can avoid future high costs by purchasing the option in advance, while the publisher can generate additional revenue by selling the option. We derived an appropriate pricing rule for such options by solving for a game-theoretic equilibrium involving two players. From the proposed model, we have shown that the option contract is beneficial in the aspects of (i) contract feasibility, (ii) publisher’s utility and (iii) advertiser’s regret. First, our result shows that the option contract will increase the chance of a contract between the advertiser and publisher being made, because an option contract can be agreed to even for a relatively lower CTR. Also, this result shows that a publisher can have more potential demand (advertisers) using the option contract, when we consider the whole market. When CTR is predicted to be very low for an advertiser’s campaign, the advertiser prefers CPC, while the publisher prefers CPM. Therefore, sometimes no agreement is possible only with CPM and CPC. However, an option contract can be used as a good tool to mitigate the conflicting interests of both parties, because it has a lower threshold of CTR. Consequently, the publisher can contract with more advertisers even for a low CTR. Second, another result showed that a publisher can make the most profits by introducing an option contract, when the difference between the expected and realized CTRs is small. This can be a great advantage to the publisher, but also suggests that the publisher should try to accurately estimate the CTR. Finally, we showed that the advertiser can expect a stable level of regret regardless of CTR. As we know, CPM and CPC might lead to losses in some cases such as too low CTR and the existence of click-fraud. In this environment, contract selection could be a source of regret. However, an advertiser who chooses an option contract does not have to worry about this environmental issue or extreme scenarios. In practice, option contracts can be used in a number of different ways. An advertiser can avoid confusion in online advertising contract selection (risk-hedging). Although display advertisements are usually contracted based on CPM in current practice as they have been in traditional media, there are many debates about how online advertisement contracts should be made. The main source of this confusion is CTR. Should an advertisement campaign receive many clicks, the advertiser will prefer to pay based on CPM. On the other hand, when an advertisement campaign receives a small number of clicks, the advertiser will prefer to pay based on CPC. However, the publisher will have exactly the opposite preferences. Using the option contract we proposed in this paper, we can eliminate a source of confusion to a certain degree. Moreover, using an option contract offers a good opportunity to a publisher who hopes to make more profit. More importantly, online advertisement is basically beneficial to both parties, because both can earn profits from advertisement. But, due to their conflicting interests, they could lose this opportunity. However, our proposed option contract can play a role in mitigating the conflict and support the simultaneous interests of an advertiser and a publisher who are worried about inappropriate contract selection and profitability. This research has a number of limitations, since we did not consider some contract factors to keep our analysis relatively simple. To make the model more realistic, we could consider, for example, duration of contract, advertising format (still image, digital video and rich media), and so on. More importantly, we could consider how a banner is displayed. It can be displayed as fixed in a page for a day or a week, or can be displayed with other banners mixed for a longer period. In the former case, for a day or a week, it will receive more attention, but predicting CTR will be harder. This research also can be extended to consider each player’s risk attitude. Some players are more afraid of uncertainty than others. This attitude will increase or decrease the option value. Another extension is to address the possibility of asymmetric information. Usually, a publisher may have more information than an advertiser for the number of incoming customers. This can result in advantage for the publisher and in this situation the option price will change also. Thus, addressing the potential for a moral hazard problem will be an interesting topic to analyze further using a principal agent model and a game-theoretic approach. Although our focus has been on the direct negotiation between an advertiser and a publisher, we may extend our approach to consider a negotiation between advertisers and the network agency that connects advertisers with multiple publishers. In practice, many publishers maintain two contracting channels: one for direct negotiation with advertisers as studied in this paper, and the other with agencies. To apply our approach to the second case, we need to resolve the issue of measuring click-through rates, because an advertisement through agencies is often displayed by multiple numbers of web sites. A simple way to do this is to use the distribution of the average click-through rates of the many web sites.