Cost Per Impression (CPM) Definition

Cite this article as:"Cost Per Impression (CPM) Definition," in The Business Professor, updated March 19, 2019, last accessed October 28, 2020,


Cost per Impression (CPM) Definition

Cost per thousand (CPM) refers to the cost of internet marketing or traditional advertising in which advertisers pay for every time that an advert has been displayed. That is, it means the expense of costs incurred for every thousand potential consumers who have viewed an advertisement. CPM is a way to measure the cost of displaying an advert on the web page. If a CPM is $50, for instance, every time that it is viewed by a thousand customers (thousand times), then the advertiser is charged $50. The term CPM stands for cost per ‘mille’ (Latin word meaning thousand).

A Little More on What is Cost Per Thousand – CPM

CPM is a common pricing approach applied in web ads. The success of this method is usually click-through rate, which refers to the percentage of people that see and click an advert. For instance, an ad which receives 3 clicks for every 100 impression has a 3% click-through rate (CTR). One cannot measure the success of an advert using CTR alone since an ad that a reader views but fails to click may still have an impact.

CPM vs. CPC and CPA

CPM is one of the pricing methods used in website advertisement. Other models include cost per click and cost per acquisition. Cost per click involves an advertiser paying each time a visitor of a web clicks an ad while cost per acquisition involves an advertiser paying only each time a visitor makes a purchase once he/she clicks an ad.

Different pricing methods have varying effectiveness depending on the ad campaign. CPM is considered the most effective for a campaign that is focused on heightening brand awareness or delivering a particular message. In such a case, the CTR does not matter since the exposure of having an advert prominently placed on a high-traffic website aids in the promotion of a company’s brand name or message even in the case that visitors don’t click on the ad.

Companies that do not focus on mass appeal and regard highly the promotion of a product to a particular audience always prefer CPC or CPA advertising methods since they only get to pay when visitors click through their site or purchase the advertised product.

Publishers prefer CPM because they get paid just for displaying advertisement. However, the CPM rates are usually low, usually $2.00, and as such requires a website to have robust traffic in order to make decent income.

In order to evaluate CPM, website publishers choose popular keywords and monitor the amount of product/service sold during the period of ad pay. Instead of calculating the total cost of the advertising campaign, CPM enables advertising companies to discover if the total costs are paying off or not. It also helps to determine to whether CPM for a particular inclusion needs of a company is working, or whether some other mentioned would be more effective.

References for Cost per Impression

Academic Research on Cost per Impression (CPM)

  • Sponsored search: A brief history, Fain, D. C., & Pedersen, J. O. (2006). Bulletin of the american Society for Information Science and technology, 32(2), 12-13. This journal describes various web search concepts as well as critical elements associated with web searching. According to the authors, web search refers to a fundamental technology for navigating the internet, although it is freely available to consumers. The author highlights key sponsored search elements including advertiser-provided content, advertiser-provided bids, review process, matching of advertiser content to user queries, display of advertiser content, and processes that gather data, meter clicks, and charge advertisers. It is noted that when sponsored search was introduced by GoTo, each listing was associated with keyword which was open-ended and advertisers were able to add new ones. As such, GoTo ensured that the advertiser content remain relevant to the keyword used.
  • Online advertising: Pay-per-view versus pay-per-click, Mangani, A. (2004). Journal of Revenue and Pricing Management, 2(4), 295-302.  This article describes the differences existing between pay-per-view and pay-per-click in online advertising. Mangani (2004) explain how the advertising industry has changed over time due to the diffusion of the internet. The author highlights that one of the most striking innovation in advertising is the method of selling advertising space on webpages. This article examines the pricing strategy adopted by web publishers who operate in a market where they are unable to influence the price of the ad. Findings show that the distribution of editorial revenues between pay-per-view and pay-per-click methods depends on the elasticity of access and actions with respect to advertising quantity. The theoretical findings have been found to be significant since the parameters are usually available to permit quick online consumer behavior analysis.
  • Determinants of internet advertising effectiveness: an empirical study, Baltas, G. (2003). International Journal of Market Research, 45(4), 1-9. This empirical study analyses the determinants of the effectiveness of internet advertising. Baltas (2003) describes the structure of advertising effectiveness and investigates the importance of media and creative factors for banner effectiveness. Based on the author’s observation, econometric modelling of actual data on banner ads demonstrates that creative factors (banner size, message length, animation, and logos) and media factors (campaign length, off-line media use, number of host web sites, and campaign cost) influence the direct response of the target audience. This is usually measured through click-through rates.
  • A Framework for the Optimizing of WWW Advertising, Aggarwal, C. C., Wolf, J. L., & Philip, S. Y. (1998). In Trends in Distributed Systems for Electronic Commerce (pp. 1-10). Springer, Berlin, Heidelberg. This article describes the framework for optimizing WWW advertising by highlight the general methods used to optimize advert management on web servers. The author discusses the key issues that arise in web advertisement management, and describes basic mathematical techniques which are employed to address such problems. The mathematical techniques considered include statistical, optimization, and scheduling models.
  • Ad exchanges: Research issues, Muthukrishnan, S. (2009, December). In International Workshop on Internet and Network Economics (pp. 1-12). Springer, Berlin, Heidelberg. This article discusses the emerging ways that advertisers are using to sell or buy display ads. Some of the real-time two-sides markets that are used globally include RightMedia, DoubleClick Ad Exchange, and AdECN. The authors, therefore, describe the abstraction of the advertising market from which they represent several research directions as well as useful insights.
  • The value of behavioral targeting, Beales, H. (2010). Network Advertising Initiative, 1. This article provides an understanding of the effects of behaviorally targeted advertising on revenues and advertising rates. The author conducts a survey involving 12 ad networks to obtain quarterly data on the pricing (CPM rates), conversion rates, and revenues in various ad segments. The results reveal 3 key findings: behaviorally targeted ads have significantly higher advertising rates; behaviorally targeting advertising is more successful than a standard run of network advertising, thus, creating greater utility among consumers from more relevant advertisements and clear appeal for advertisers from increased ad conversion; and a majority of network advertising revenue is used in the acquisition of inventory from publishers, making behavioral targeting an important revenue source for online content, service providers, and third-party ad networks.
  • Dynamic revenue management for online display advertising, Roels, G., & Fridgeirsdottir, K. (2009). Journal of Revenue and Pricing Management, 8(5), 452-466. This article discusses the management of dynamic revenue in the case of online display advertising. The authors propose a model of dynamic optimization to maximize the online display advertising revenues for web publishers. The model selects which advertising requests to accept and dynamically delivers the promised advertising impressions to viewers in order to maximize revenue, which helps to account for uncertainty in website traffic and advertising requests. The authors also propose a Certainty Equivalent Control heuristic and show the real case study that the chosen optimization-based method outperforms common practices. The findings illustrate the importance of accounting for capacity allocation’s opportunity costs in advertisement contract negotiation for globally maximizing the revenues of online publishers.
  • A uniform allocation mechanism and costperimpression pricing for online advertising, Araman, V., & Fridgeirsdottir, K. (2011). Working paper. This article by Araman and Fridgeirsdottir explores the uniform allocation mechanism and cost-per-impression pricing model for online advertising. According to the authors, online advertising is a fast growing area in the media industry, and web publishers have been faced with operational problems in the selling advertising space. The authors develop stylized model that incorporates some essential ingredients of the advertising operation of a publisher. Based on the findings, the authors suggest a capacity allocation mechanism which shares the capacity among different advertisers and matches the requirements of a campaign with a supply of viewers. The proposed mechanism results in economies-of-scale and is proven to be optimal asymptotically.
  • Single-period balancing of pay-per-click and pay-per-view online display advertisements, Kwon, C. (2011). Journal of Revenue and Pricing Management, 10(3), 261-270. This article explores the concepts of pay-per-click and pay-per-view online display advertisements in a single period. The authors refine prior research results considering the contracts details. They examine the research problem by formalizing a simple stochastic optimization problem for a single period of advertising contracts. Further, the author investigates how pricing and other contract components affect the optimal display strategies numerically and analytically.
  • Factors influencing consumers’ willingness to accept mobile advertising: a conceptual model, Leppaniemi, M., & Karjaluoto, H. (2005). International Journal of Mobile Communications, 3(3), 197-213. This article analyses the factor that influence consumers’ willingness to accept mobile advertising using a conceptual model. The authors aim to build a model that investigates factors that influence the acceptance of mobile advertising from the point of view of both consumers and the industry. Using previously reviewed studies, the authors also propose a conceptual model of the willingness of consumers to accept mobile advertising. Based on the set research hypotheses, the model indicates that 4 factors primarily influence the willingness of consumers to accept mobile advertising. These factors include the role of mobile medium in marketing mix, technology development, one-on-one marketing medium, and regulatory factors. The findings provide several managerial and conceptual insights into the role of mobile advertising today and in the near future.
  • Optimal real-time bidding for display advertising, Zhang, W., Yuan, S., & Wang, J. (2014, August). In Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining (pp. 1077-1086). ACM. In this article, the authors study real-time bidding (RTB) optimization based on display advertisement. RTB enables advertisers to bid on a display ad impression when it is being generated in real time. The authors assert that RTB goes beyond contextual advertising through the motivation of bidding focused on user data. It is also noted that RTB is different from the sponsored search auction where the bid price is associated with keywords. On the side of demand, a fundamental technical challenge is to automate the process of bidding based on the budget, information gathered in runtime, campaign objective, and history. The findings of the study show that there is a non-linear relationship between optimal bidding and the impression level evaluation like click-through rates and conversion rates. This, according to the authors, is different from the past research which focused on the linear bidding function.
  • Impression Fraud in On-line Advertising via Pay-Per-View Networks., Springborn, K., & Barford, P. (2013, August). In USENIX Security Symposium (pp. 211-226). This paper discusses the concept of pay-per-view networks and their link to online advertising impression fraud. The authors assert that advertising is one of the key means for generating revenue among millions of mobile apps and websites. Although pay-per-click is the main online advertising revenue generator, alternative forms like video advertising and impression-based display are also becoming common. The authors, thus, investigate the problem of invalid traffic generation which aims to inflate the website advertising impressions. The study starts with the analysis of purchased traffic for a set of honeypot web sites. Information gathered from these sites offer a window into the basic mechanisms used for impression fraud thereby enabling users to identify pay-per-view networks.

Was this article helpful?