Here at C3, there is life before Futures of Entertainment (FOE) - and life after FOE! It is a fun but challenging event to put together - as we are committed to programming the event with panelists and themes which are at the bleeding edge of our research agenda for the 2009 - 2010 academic year. We find that if we do FOE right, we send FOE attendees back to their own professional and scholarly realms newly inspired to keep in touch with the Consortium - and to contribute in the months ahead to the creative and intellectual challenges we explore together over the course of the two days @ FOE.
By all accounts and possible metrics, Futures of Entertainment 4 was a huge success. Once again, thank you to everyone at CMS, the C3 Team and C3 Sponsor Companies for another thought-provoking and successful event. In both our Glancing at the C3 Blog and
C3 in the News sections of this newsletter, you will find links to discussions, articles and general coverage of the event (there is also a link to the recent article in the Los Angeles Times about Prof. Jenkins).
In our Opening Note, we introduce a new feature of the bimonthly C3 Newsletter, which is the publication of a Research Memo Abstract. This publication of an abstract by the C3 Researcher allows the C3 community to hear some of the formative thoughts and questions which will be expanded into a C3 Research Memo (another new format we are introducing this year). Your feedback, questions or concerns about the abstract will insure that the C3 Researcher is addressing some of the questions unique to C3 scholars and C3 sponsor company employees. So your feedback is crucial - and we encourage you to get back to the C3 Researcher directly.
C3 Researcher Ravi Inukonda is first up! Ravi provides a personal greeting to the C3 Community - as well as some professional context and his own point of view on his research topic in the Researcher Introduction section of this newsletter.
Finally, in our Closing Note, we have Part 1 of a 2 Part article (first run in the fall 2009 version of CMS' In Media Res) by current CMS Graduate Student Audubon Dougherty '10 - detailing her personal adventure with social entrepeneurship/Information and communications technology (ICT) in rural Peru.
This issue of the C3 Newsletter was prepared by Daniel Pereira. If you have any questions, please don't hesitate to email him.
In This Issue
CMS EVENTS on the MIT Campus:
12.11.09 | 5:15pm | 14E-310
Comparative Media Insights: Ken Wark: "From Gamer Theory to Critical Practice"McKenzie Wark is chair of Culture & Media and associate dean of Eugene Lang College, and an associate professor of critical studies at the New School for Social Research. He is the author of A Hacker Manifesto (Harvard UP, 2004), Gamer Theory (Harvard UP, 2007) and various other things.
12.15.09 | 5:15pm | 14E-310
Comparative Media Insights: Lisa Nakamura: "Race, Rights, and Virtual Worlds: Digital Games as Spaces of Labor Migration"
Lisa Nakamura is the Director of the Asian American Studies Program, Professor in the Institute of Communication Research and Media Studies Program and Professor of Asian American Studies at the University of Illinois, Urbana Champaign. She is the author of Digitizing Race: Visual Cultures of the Internet (University of Minnesota Press, 2007), Cybertypes: Race, Ethnicity, and Identity on the Internet (Routledge, 2002) and a co-editor of Race in Cyberspace (Routledge, 2000). She has published articles in Critical Studies in Media Communication, PMLA, Cinema Journal, The Women's Review of Books, Camera Obscura, and the Iowa Journal of Cultural Studies. She is editing a collection with Peter Chow-White entitled Digital Race: An Anthology (Routledge, forthcoming) and is working on a new monograph on Massively Multiplayer Online Role playing games, the transnational racialized labor, and avatarial capital in a "postracial" world.
Hello C3 Community
My name is Ravi Inukonda and I am a Graduate Research student with the Convergence Culture Consortium. I am also a 2nd year MBA student at the MIT Sloan School of Management. I worked with Dan and Joshua over the spring to investigate some of the strategic and tactical issues that C3 is confronting. I am glad to be back with the team this semester shifting my focus to research initiatives.
Before coming to MIT, I spent several years in strategy and product management roles at technology firms such as Microsoft and EMC. Through these roles, I have come to appreciate the beauty of technology and software. I spent this summer working at an early stage venture capital firm in the Boston area. They say venture capital is a sport a “mile wide and an inch deep.”. Contrary to popular belief being an intern, I spent a good portion of my time investigating the internet and digital media spaces.
I learned a ton about the various business models in these spaces and very quickly realized that a massive change was undergoing from below us in the way advertising is produced, consumed and delivered. I saw that there was a new stakeholder in the advertising value chain that has quickly become the “efficient marketplace” for advertising. This new class of stakeholders is popularly known as “ad networks” and “ad exchanges”. These players have shifted the dynamic and are trying to put the power in the hands of the consumer and not the supplier.
After my summer internship ended, I met with Dan and proposed some work this fall that would further these insights. My focus is going to be on examining the supply/demand strategies in ad networks. With the recent consolidation in ad networks and changes in economy, advertisers and media properties face new challenges in their ad network strategy. The existing ad network models haven't offered transparency on where campaigns run creating information asymmetry between supply and demand. As liquidity improves through the exchanges and targeting audiences becomes more main stream, transparency will be a key lever in conducting business for both sides. The need for media optimization will become increasingly necessary in this new digital era. There may be bumps along the way, but technology will only make the media buying and selling process more efficient – creating significantly better yield wherever there is true value.
I am very excited to be a part of the Consortium team again this year. We had a very successful team meeting with C3 Principal Investigator Prof. William Uricchio the other day, where I was able to share my research topic with William and the C3 Team for the first time since FOE4. William pointed me in some great directions, and I look forward to getting some feedback from Prof. Jenkins in the weeks ahead. As Dan mentioned in the editor's note, we have also redesigned the C3 newsletter to include the publication of a research memo abstract (below ) prior to publication of each C3 Research Memo - in order to solicit feedback, questions and concerns from the broader C3 Community. Please get back to me directly! I can be reached at ravii@MIT.EDU.
Research Memo Abstract
Introduction to Advertising Networks
An ad network is a company that connects online advertisers (companies that want to advertise their products/services online) to online publishers (websites, blogs, social networks, and other online properties that want to host advertisements in order to generate revenue).
Ad networks have several different models, which are discussed more in depth later in this report. An ad network can buy inventory from various websites, often remnant inventory or all inventories in the case of smaller publishers, and then resells that inventory to advertisers or agencies. Essentially, in this model, ad networks make money by selling inventory for more than they buy it. So, to boil down how this type of ad network keeps its revenue flow positive just imagine a basic buy sell model; it has to buy inventory from the publishers cheaper than it sells it to the advertisers. (Business models are discussed in later sections)
Another type of ad network, and some say that is the pure ad network, is that which has a portfolio (or network) of publisher partners, which give the network exclusivity to sell their advertising inventory. In this case there is often a revenue share between publishers and the network. Basically, an ad network is a collection of online inventory which comes in various forms. Types of inventory include banner ads, rich media, e---mail, text and video ads. This inventory can be found in various places on the internet from blogs to RSS feeds to traditional internet portals. Ad networks came into existence to be a sort of media representative for publishers. Both big and small publishers need ad networks in order to keep their sites generating revenue. In the case of small publishers, better known as the long tail, ad networks are an essential part of the advertising business, as these publishers do not have the money or infrastructure to have an in---house advertising sales team. An ad network or networks is able to take the inventory from this publisher and sell it to various advertisers for a competitive price. Of course, these ad networks have inventory from various (hundreds and even thousands) of sites and are able to offer interesting packages to advertisers.
In the case of large publishers even web portals, ad networks are also a necessary part of the advertising ecosystem. There is always remnant inventory that large publishers are unable to sell for various reasons including cost/benefit for the advertiser. These publishers look to networks in order to sell this inventory and usually at a fraction of the price that the publisher would sell it for.
Yes, publishers do have mixed feelings about this--- ad networks become a necessary evil, as they sell inventory that otherwise would not be sold, but for an extremely low margin. Each ad network has a different way of operating and offering their respective inventory.
Advertisers look to ad networks not just to get inventory cheap, but also to be able to target their ads more efficiently. There are various types of targeting as well as more focused vertical networks that will be discussed in more detail in later sections. Ad networks allow advertisers to reach a wider audience simply based on the fact that the ad network provides the link between an ample array of websites (not just big internet portals) instead of the advertiser or agency having to have a relationship with each website. Not only does an ad network help amplify the audience possibilities for advertisers, but it has the ability to make for a more targeted and focused ad campaign and at the same time makes for more lucrative media vehicles for publishers by monetizing the internet [long tail] more effectively.
Over the last decade there have been hundreds of ad networks that have come into the online internet advertising space in order to fulfill a need for publishers to maximize their ad inventory and for advertisers to buy ad space more efficiently. There are several key differentiating factors which characterize ad networks:
• Inventory acquisition
• Type of inventory
• Particular media type (video, display)
• Pricing models
• Technology (algorithms)
There are several different models of targeting that an ad network can adopt, such as geographical, behavioral or contextual. There are also ad networks that focus on a specific vertical such as women, sports or outdoor living. These are known as vertical ad networks and also use targeting models inside of these verticals to reach a desired audience. Returning to the basics of targeting, ad networks offer advertisers ways to reach their target audience without cringing thinking that their ads could be on just about any site anywhere [in that country] and have little efficiency. Certain ad networks offer combinations of various types of targeting to focus even more and have the best click---through rate possible.
Geographical targeting is just that, targeting advertising messages based on geography. Let’s say the advertiser is a chain of restaurants in California. The advertiser will want to target his ads to people who live in the state of California and will not have interest in having people in Idaho see his ads as they will most probably not result in clicks, and definitely not in actual visits to his restaurant. With geographical targeting, the advertiser can decide down to the city to which he would like his ad to appear.
In the case of behavioral targeting, there are ad networks that call themselves behavioral ad networks as they specialize in this type of targeting. Behavioral advertising is based upon the notion that an advertiser should care more about a narrowly defined group of consumers viewing a message than about the context of a message or broadly defined demographic. Behavioral targeting is done by first using cookies to track user behaviors which then are used to create user profiles. For example, user A visits espn.com, foxsports.com and also facebook.com every day. With this information it can be deduced that user A is male between the ages of 16---25 who likes sports and is also an avid social network user. There would be tens, hundreds or thousands of users who would have similar profiles to user A and would be great candidates to receive advertising regarding Nike sportswear, for example. Nike can go to this ad network with behavioral targeting and specify his demographic or profile of his ideal target and the ad network can pull up user A + co’s profile. This is a simplified example of behavioral targeting and it can get much more complex than this, but it is a good reference point. Ad networks that are considered the leaders in behavioral targeting are Tacoda (now part of AOL), Revenue Science as well as Blue Lithium (now part of Yahoo!).
Contextual targeting is based on the belief that the context of a site is most important in determining what advertising message should be displayed. Therefore, when a user is navigating on a site, ads will appear that have to do with the content of that particular site. There are a few different types of contextual targeting, but the most well---known and used is Google’s AdSense which shows text ads on publisher sites based on the content of that page or site. AdSense is heavily used on long tail sites. Another example of contextual targeting would be Contra’s in---text ad network which provides ads embedded in specific key words within the body of a webpage. When talking about targeting, it is also important to discuss a little about vertical ad networks themselves. Vertical ad networks are those that narrowly focus on certain industries. A great example of this is Martha Stewart Living ad network Martha’s Circle, which has a portfolio of publishers that focus on home and lifestyle. An advertiser, for example Unilever, who is interested in advertising their newest detergent, would look to this ad network to reach a certain kind of audience and certain types of publishers that deal with home and lifestyle content. A plethora of vertical networks have come into the market in the last years such as ESPN’s network focused on sports websites and The Washington Post’s ad network which in its turn focuses on news journal sites. And, as shown in these examples, the line between ad network and publisher can become quite blurry as lots of these networks have their own publishers.
The most important element of an ad network’s business model from a financial perspective is the strategy they use to acquire inventory, and of course, then sell it. The following is a description of the various types of inventory acquisition:
In this model the ad network does not have possession of inventory; instead, it represents publishers and gains revenues based on commission. It is therefore a service business with less fixed costs and of course does not have technology integration. In turn, this model is less scalable than others. Example: Gorilla Nation
Direct Revenue Share
This model also does not have possession of inventory, but it does have targeting and optimization technologies. Different than commission based model, there is a more integrated relationship between the network and the publishers.
Examples: Burst, ValueClick
Here the ad network buys inventory from the publisher for an extremely low price (mainly due to the fact that otherwise they would not sell this inventory at all) and then sells the inventory to advertisers in larger aggregated blocks at a slightly higher price. Their margin is not big here, but the scale usually large. The only problem in using this model in terms of the greater impact is the fact that publisher will have a distorted larger scale as inventory is being bought, but not necessarily it will be resold to advertisers.
Price Model Arbitrage
This model is higher risk than the prior, but with this risk comes a higher margin. The network buys inventory from the publisher on a CPM basis and then sells it to advertisers on a CPC or CPA basis, reducing the risk for advertisers. The network makes money from the difference of the cheaper CPM price with the higher CPC or CPA price.
Types of Inventory
Depending on the ad network’s focus, the type of inventory that it has varies. For example, there are certain ad networks that focus on remnant inventory from large sites and others that represent or buy all inventories from small publishers (long tail) as a sort of ad sales representative or resell house of all inventory. This normally occurs with small or even medium sized publishers (mid tail) that do not have the infrastructure to employ an in---house sales team. A good example of an ad network that primarily attracts long tail publishers but also is interesting for larger publishers who prefer the aesthetics of text ads is Google AdSense. AdSense has the advantage that it works on a self---service basis making it cost---effective for Google and easy for the publishers. Other ad networks are more focused on Mid Tail publishers. Microsoft’s DrivePM network represents unsold inventory on sites with some degree of name recognition and with that offers advertisers a certain level of quality control. Networks that work with a fewer number of sites give the advantage to the advertiser of a good level of transparency in terms of what kinds of sites their ads will appear.
When we talk about inventory, we are talking about a wide range of media types that make up that inventory and each publisher has certain media or inventory type(s) that it offers that it considers adequate for their specific needs. In this report we will be focusing on certain types of inventory, but it is important to briefly discuss a wider range.
Use to be categorized as “banner ads”, but now has a broader definition which includes different ad formats such as video and pop---ups and placements as well as different media types such as Flash.
The king of text ads continues to be Google. Google’s AdSense matches ads to a publisher site’s content in the form of text (but also offers image ads as well). Text ads are often an interesting way to integrate ads into a site without being and looking obtrusive.
The new wave on the net is clearly video with YouTube in the Top 5 sites of tens of countries around the world. The question, of course, has come up as to how to monetize these videos. Video ads come in various formats such as pre---roll, post---roll and layover in order to seamlessly integrate with the video content. Video ad networks such as VideoEgg have popped up in order to fill this demand Mobile Ads Mobile telephone penetration is booming around the globe and in the last couple of years mobile operators have been putting the effort into how to monetize this space. Mobile ads mainly come in simple formats such as mobile banner ads which could appear during games or when receiving a text message. In many countries, although mobile penetration is growing rapidly, advertisers are not sold on this media just yet. We should be seeing more activity here in the next 12---18 months. The biggest mobile ad network is AdMob (purchased by Google for $750mm in November 2009). In the U.S., as well as Europe, there has been an evolution away from banner ad into rich media and search formats. Actually, in the U.S. search ads count for more than 50% of all online advertising spending. It is cost effective and more targeted than other formats.
Ravi Inukonda is a Graduate Research student with the Convergence Culture Consortium. Ravi’s research is focused on ad networks, ad exchanges and creating efficient marketplaces for ads.Ravi is also a 2nd year MBA student at the MIT Sloan School of Management and is a member of the Entrepreneurship and Innovation Program. Prior to Sloan, Ravi held several management positions in enterprise software companies, most recently leading a strategy group in Microsoft's Enterprise Security Division where he led the incubation of several new products. Furthering his entrepreneurial passion, Ravi has also started two companies in the software space. Ravi has a M.S in Computer Science from Clemson University, SC.
Glancing at the C3 Blog