Computer-Mediated Communication Magazine / Volume 1, Number 5 / September 1, 1994 / Page 13


Pyxis Cyberia: Online News Service Advertising

by Steve Brown (sbrown@ufcc.ufl.edu)

[During the spring of 1994, seven students at the University of Florida 's College of Journalism and Communications developed a prototype for a World Wide Web-based news service called the Florida Compass. While many of the lessons we learned were just new versions of guidelines already established in journalism, both these 'new versions' and the process of our re-learning might be of interest to those watching the future of the news media. This is the fourth in a series of reflections on what we learned from creating the Compass. Steve Brown handled advertising.
--Gary Ritzenthaler]

Our group working on the Florida Compass project had several areas of design well-covered including technical, visual and news. Given my past experience working for an ad agency, it was a good match for me to work on advertising. I began searching for sources that would give me insight into the subject of advertising in an electronic environment.

The book, The Media Lab, by Stewart Brand gave me some very good ideas. One section in particular, 'The Invited Persuader' in the chapter "The Politics of Broadcatch," page 201, opens with Nicholas Negroponte, director of MIT's Media Lab, dividing advertising into three categories: 'advertising as noise', 'advertising as news' and 'reverse advertising'.

Advertising-as-noise, "the Marlboro-type ads whose information content is zero," according to Negroponte, will not do well. Advertising should not aim to create need, it should aim to fulfill it. Fulfilling need involves knowing the context in which the users live: knowing what products are of most interest to them, what places they should visit or frequent, and what ideas they should learn. The technology is moving toward a level of customization that will favor information that can be associated with the user's needs. Information that is not a benefit (junk) will not do well. Information that is valuable (news) to the user will benefit from a highly customizable environment. This leads us to the next category, advertising-as-news.

Advertising-as-news, will be received and consumed by the user systematically. The user will devote a certain amount of time to the task of processing this information. It will be similar to reading a newspaper on a daily basis. Advertising-as-news will be divided into information relevant to only you and information that is relevant to everyone. An example of information relevant only to you would be a sale on jeans that matches your tastes in style, color, fit and quality. Few people are interested in information that matches your tastes. An example of information relevant to everyone would be where to buy environmentally safe products. It's within everyone's best interests to receive this information. Since we receive advertising-as-news everyday, it helps inform us about new opportunities, but what about specific problems we want to solve? This is where reverse-advertising comes in.

Reverse-advertising is advertising by the user as an individual; the cash-cow want ads so coveted by the newspaper publishers. Users are engaged in buying, selling or trading goods and services, and are using the medium as a place to do business. Users have a strong sense of what they want. They want to solve problems and get involved in the process when problems arise.


Advertising-as-news and reverse-advertising are the sort of information users will experience in a highly customizable world of the future.

Customization is a difficult, time-consuming task. This is where the best qualities of the computer come into play. Computers are designed to manage large quantities of information and the part of the computer that will do the managing will be something akin to a personal assistant or agent. Your agent, a personalized piece of software, will work on your behalf, making decisions for you, freeing you from simple tasks like sorting and counting. It will go out and collect the advertising-as-news that is most relevant to you, say the ten most important pieces of news. It checks to see what reverse-advertising is relevant to you, for example, who is selling something you want to buy. This agent will know about your likes and dislikes among other things. It will be sensitive to your schedule and exactly how much time you have to devote to monitoring incoming information. In most cases, the term 'agent' can take the place of user in the rest of the column.

There are some implications for advertising, public relations and the news services, i.e. media, in these ideas. Instead of casting your 'message in a bottle' out into the sea of news services, you build a bridge to the mainland of users in two ways. With the first way, you connect to the public directly. Obstacles like time and space are minimized when you can retrieve a document the size of the average book in seconds from a site halfway around the globe for little or no cost. This direct connection is reverse-advertising and acts to solve problems. The second way involves connecting to news services, ones that are targeted well or as stated before, highly customized. The links from news services are discrete in the form of hyperlinks and lead back to the source of information, the originating site. This is advertising-as-news and acts to create opportunities.

The Web is a network structure with the potential to reflect the system described.


Instead of senders and receivers in a passive medium, the participants are described as players all gravitating to innovations that best meet their needs.

These users of the system work to create information that fits into the categories of advertising-as-news and reverse-advertising, and try to avoid advertising-as-noise. The following is an example of how a company might utilize this network structure.

Advertising on the Web

A car dealer publicizes itself by first creating a site on the Web. It connects using all the necessary technology and develops content. It controls all content at the site and is aware of requests for information. All links terminate at the site rather than stopping at a news services's site. I will start with the simpler of the three types of advertising.

Reverse-advertising for the dealer begins with a list of cars available for sale. If a request from an agent matches one of the cars on the list, a connection is made. The agent would present a list of cars fitting the criteria the user established before making the request. Most likely, the user wanting to buy the car would take the next step and make a trip to the dealership for a closer inspection. The dealer could also make requests to buy or trade when stock is low or poor in quality. As stated before, these users are familiar with what they want and out to solve problems.

Advertising-as-news for the dealer begins with identifying then buying advertising from news services that best target the users that fit the profile of the dealers average customer. The challenge is in developing advertising that explores user's needs. The dealer who sells the highest quality cars at the best price will have an easier time advertising than its competition. That's business, but the message is key. Remember, advertising-as-news is divided into information relevant to only you and information that is relevant to everyone. It's important to appeal to users as individuals and as a common group. Knowing users and appealing to their desires with appropriate information fulfills the need for new opportunity. Make the appeal well and agents from around the globe will carry your message home to their users.

Advertising-as-noise and advertising-as-news are only different in the amount of content they contain for users as individuals. In the user-centered world Negroponte and I have described, multimedia images are important but have to be appropriate. How many car commercials have we seen showing a slick convertible moving effortlessly down Interstate One without another car in sight followed by a logo and a slogan that would make Pavlov proud? A lot. Is this relevant to our lives? As much as they would like to artificially create the need, it is not relevant and that's why it is ignored by the majority.

One limitation of the Web involves the lack of tracking requests for information. As it is, all requests appear to come directly from the user even if it came via a hyperlink found in a news service document. The advertiser needs the ability to know and confirm the path taken by users and agents to the site. If the system doesn't change, advertisers and news services will continue to guess if an ad is worth the money by monitoring overall sales, though this is an oversimplification.

In the scenario above, the lines between advertising and public relations begin to disappear. The bureaucracy created to broadcast information in the traditional media is replaced with a new system of individual document access. The message, be it political, financial, etc., is provided at one location and can be accessed by anyone (almost). The new 'information officer' will work to develop information of value, perfectly targeted to users, and secure links with news services that attract users who fit a well defined class that would have an interest in that company. Now I'll explain how this affected our prototype.

The Compass team created advertising by put ads with stories. Any document that lead to stories was free of ads. The banner page and other main sections were all free of ads. This was a compromise between group members and we defined it as a limited intrusion-style format for display (.gif) ads. Another compromise developed: to use a regular newspaper format as a metaphor, the news section's stories did not contain advertising, as did other sections. The Money, Weather, Sports, Style and Tech sections, for example, all contained ads at the bottom of stories. If the user is interested in seeing the news (our lead section), then that's what they see and nothing else. While browsing, certain limits apply. The user is exposed to a limited conceptual space (news) and is not distracted by parallel, unrelated material. In the sections which did contain advertising, the user is interrupted only when they have concluded a story.

The decision to put the ads at the bottom of the stories was also a group compromise. The rationale behind our decisions was simple though reaching it was difficult given the different philosophies contained within the group. Opinions ranged from 'put as many ads as possible' to 'create an area for ads and that's it'. We didn't want to overwhelm users with ads and drive them away. We also didn't want to create a 'ghetto' of information by putting them all together and labeling them 'ADS' to be avoided by users.

This reflects the team's desire to provide the most relevant information while conceding advertising's worth with respects to revenue. The question of writing style, the inverted pyramid versus the standard feature, could influence how many actually see the ad. If the ad is visible when the document is loaded, then there isn't a problem. If the user must scroll down to see the ad, chances are, the ad will not be seen at times if the inverted pyramid is used. This is a complicating factor, but one the group was willing to live with.

Advertising in the Compass

The Compass provides a link to the advertisers in the form of an inline image (.gif). The advertiser chooses where and when their ad is displayed. Stories read more could bill higher rates, for example, a fixed fee plus a cost per access. The advertiser is guaranteed minor exposure because the ad is included with a story. The advertiser has the potential for major exposure if the ad is 'clicked' on. The advertiser and the news service know how many users access the story. If the ad is popular, the link is strengthened; the users want the information enough to look further, meaning they retrieve information from the advertisers site. If the ad is unpopular, the link is weakened; the advertiser and the news service agree that the users of the service are uninterested and may become annoyed enough to avoid both parties. (Remind you of Nielsen's "Passive People Meter" or what? Check out Wired December 1993, page 68, for more information.) When the limitation of path recognition is solved, this system will be effective in informing advertisers exactly where their dollar is best spent.

The reverse-advertising portion of our service was conceptually developed, but not implemented. It was planned to contain want ads from subscribers and nonsubscribers. The user would access a dedicated database for classifieds which would run along side the Web server application. Users would use forms to enter queries and receive results. Other forms would be used to enter ads online. Users could pay online (provided a secure transfer exists) or use traditional means of billing. The ability would exist to search for specific qualities within the database. For example, cost could be queried. A user could request all classifieds selling things for less than $25.

This is the area an agent could do its best work. I imagine agents wandering the Internet, checking known sources like this database above on a regular basis. If the information an agent is looking for can be converted to bits, then transportation from one user to the other is easy. On the other hand, items like car parts are difficult but not impossible. On a regular basis, items are shipped all over the world.

In the prototype, we made decisions based on our convictions. Unfortunately, we will never have the opportunity to implement our prototype. The point is to create an ideal and then, with budget and other constraints in hand, create the real thing. Questions remain. Will users share more of the cost when advertisers contribute less? Will content be valuable enough to attract enough users to the news service? Will enough people have access? Will users replace traditional media with this technology? And, are people willing to change their behavior from passive to active, from being fed information to creating it? I'll let others answer these questions (the hard ones). ¤

Steve Brown is a graduate student at the University of Florida.


This Issue / Index