KMWorld CRM Media Streaming Media Faulkner Speech Technology Unisphere/DBTA
Other ITI Websites
American Library Directory Boardwalk Empire Database Trends and Applications DestinationCRM EContentMag Faulkner Information Services Fulltext Sources Online InfoToday Europe Internet@Schools KMWorld Library Resource Literary Market Place Plexus Publishing Smart Customer Service Speech Technology Streaming Media Streaming Media Europe Streaming Media Producer Unisphere Research

Magazines > Online > March/April 2004
Back Index Forward

Online Magazine
Vol. 28 No. 2 — March/April 2004
Making Money as an Aggregator
By George Plosker

When confronted with the history of online, what surprises people the most is its longevity—the online news and magazine aggregator business has been around for approximately 30 years. Dialog celebrated its 30th anniversary in 2002; LexisNexis in 2003. Incredibly, between 1971 and 1974, in addition to Dialog and LexisNexis (then just Lexis; Nexis launched in early 1980), OCLC, SDC ORBIT, MEDLINE, The New York Times Information Bank, and Dow Jones News/Retrieval all began commercial service.

The zealots who came out of this age believed that the intelligence, energy, and alignment among these pioneer companies and their customers increased awareness of online access to useful electronic content, which eventually led to the Internet and today's World Wide Web. While this may be true, it is clear that the Web's incredible popularity and pervasiveness has rocketed past the traditional industry in terms of both number of users and number of searches. Why is this so? Even with a significant head start the traditional companies have, at best, experienced incremental growth, while the Web moves forward on a completely different, almost logarithmic, scale.

Today's aggregators find themselves in a very difficult environment. To some extent, they are stuck in the middle between publishers that want as much revenue as possible and a public who doesn't want to pay for content. The most recent challenge to growth is the "Google is good enough" phenomenon. What can the aggregator do to begin to take back market share?


From the beginning, aggregators tended to focus on different content areas and target markets. While there was some overlap, and therefore competition, the services did not parallel each other. The earliest aggregators, Dialog and SDC ORBIT, competed in terms of being the first "database supermarket service." Each began with a core of scientific and government databases, later adding business content to expand the appeal of the service.

LexisNexis traces its heritage to the Ohio State Bar Association (OSBA), which wanted a legal research system for Ohio statutes. When Mead Data Central, then the owner of Lexis, moved into the news and business arena to start Nexis, it found a marked contrast to the legal environment. A good portion of the legal information was in the public domain. In contrast, Nexis had to conduct intricate negotiations for marketing rights to information held by a great variety of publishers, including media giants like the Washington Post and the New York Times. This point gets at the core of content aggregation—the relationship between the aggregator/search service and the publishers. Dow Jones News/Retrieval, renamed as Dow Jones Interactive, is now Factiva, a joint venture of two other media giants, Dow Jones and Reuters, that own core content aggregated by Factiva. Still, Factiva must negotiate for content it doesn't own.

Another early search service was BRS, a major piece of which morphed into Ovid to focus on medical and life sciences content. A partnership among Chemical Abstracts Service, Fachinformationszentrum Karlsruhe (FIZ Karlsruhe) of Germany, and Japan Information Center of Science and Technology (JICST) created STN, which concentrates on scientific, technical, and patent literature with some accompanying business content. Today, STN has over 200 databases, "covering a broad range of scientific fields, including chemistry, engineering, life sciences, pharmaceutics, biotechnology, regulatory compliance, patents, business, and more."


All these aggregator search services had certain points in common. They grew out of technical environments that, typically, were government-funded. This enabled them to spend the large amounts of capital necessary to create large-scale data centers. All had distinct ideas of what market segments they would go after and enjoyed strong growth for many years in serving those segments, although perhaps not quite in the way they envisioned. They had a focus on obtaining—through licensing, partnerships, and, in some cases, ownership—authoritative and compelling content. As user environments and technology changed, these search services introduced interface improvements and, in some instances, new services to meet the needs of different groups of users. So, what went wrong? Why did aggregator search services never capture the breadth of demand they sought? Information industry writers and pundits have typically pointed to two key factors in terms of what inhibited the growth of the traditional aggregators—cost and complexity.


Historically, aggregator search service licensed all or most of their content from outside publishers. The publisher or content provider obtains royalty payments from the search service depending on how often customers use their material. In the transactional model, a search service charges users for each document viewed and pays the publisher each time this occurs. In the subscription model, the aggregator sets up a defined royalty pool derived from the total amount of subscriptions sold. The aggregator still tracks usage and the publisher receives a payment based on a percentage of the total hits within its content in the royalty pool. However, the details make the changing mechanisms of the large aggregators frustrating for the client. First, transactional pricing can be very complicated. Second, subscriptions are not true flat fees.

Connect-time pricing originated with Dialog at the very beginning of online. Early customers paid three variables: (1) connect time, (2) record charges, and (3) charges for the telecommunications network. Today, Dialog has well over 500 databases. Each database has distinct charges for connect time and records. Connect time varies from $0.42 to $5.93 per minute. Record charges range from $0.13 for the King James version of the Bible, to over $200 per record for corporate structure reports and a handbook of organic chemistry. In one especially valuable and complex patent database, there are 36 separate formats. Dialog charges additional separate charges for Alerts, redistribution or archiving fees, and powerful ranking and reporting features.

Subscription pricing from a large aggregator is a good way for the customer to gain control and budget predictability—or so it would seem. This is true for the period of the contract. As has been observed by many practitioners and industry writers, companies track actual (transactional) charges during the budget period. When it comes time for renewal, the aggregator bases the renewal on the relationship between the previous subscription amount and the actual usage. Assume you have a subscription contract with the aggregator for $50,000 but you actually "use" $100,000 worth of product. To show "goodwill," the aggregator will often "split the difference" so that the renewal amount, in this example, would be $75,000. That's still a hefty increase for the customer.


Knowledgeable administrators are aware of this potential discrepancy, and some attempt to limit the use of the product during the first contract period to minimize increases at renewal. In some cases, intermediaries refuse to allow end users access to the product, realizing the resulting impact on their budget. In other cases, the administering individual or department attempts to obtain a portion of the fees from end-user departments. This has varying reactions depending on how valuable the end-user department views the aggregated data. In too many cases, a subscription contract, designed to increase visibility, usage, and utility, has the opposite effect when administrators refuse to promote availability of the product to the departments that might use it.

First-time visitors to major aggregator Web sites are often confused when they cannot find how much the service costs. In fact, it is actually quite difficult for an aggregator to answer what seems like a simple question: "How much will a subscription cost?" There may be contract language that the first 3 months will cost X amount and then there will be an evaluation to determine charges for the next period.

One step forward: The "big three" now offer immediate access to their services via online sign-up with billing to a credit card. For example, LexisNexis offers two credit card options for nonsubscription clients. The first is a pay-as-you-go option. In this model, there is no charge for searching; users only pay for the complete text of documents viewed. Fees include $3 for newspapers, trade journals, newsletters, and wire services. Company and financial information ranges from $4 to $12. The second model is pay by day or week. Costs range, based on what specific content you wish to use. Available content categories include Newspapers, Business and Financial databases ($75/day; $250/week), Business Sources ($50/day; $150/week), and Major Papers ($30/day; $75/week).

Factiva offers individual subscriptions for $69/year plus $2.95 per document. The individual subscription also includes alert capabilities for $9.95 per Track Folder per month. However, this subscription option excludes select premium content, including Historical Market Data and Quotes, Company Quick Search, Business Newsstand, Investext, and Dun & Bradstreet.

Many potential users and observers see online connect time as a major inhibitor to use. These same observers comment that major aggregator fees are set up in a way that discourages adoption and success. On the other hand, charging mechanisms that provide searching and browsing for free and only charge for what is actually read are seen as much more user-friendly.


The cost issue goes beyond perception and presentation. It is expensive to be in the large-scale aggregator/search service business. Aside from the issue of sharing revenues with publishers and other content providers, as discussed above, aggregators must have large-scale data centers, which are expensive to maintain, especially when literally thousands of simultaneous users are demanding instantaneous response time.

To their credit, aggregators are continuing to experiment with pricing in attempts to maintain existing customer loyalty as well as to entice new users with a reasonably priced value-proposition. According to Roy Martin Jr., president and CEO of Dialog in an interview with Matthew McBride published in Searcher, May 2003, "...over time, pricing models in the industry will continue to evolve as customer and user requirements change. As information
delivery continues to migrate across the enterprise, different pricing models more attuned to this growing user group—especially those that involve fixed-price options—are likely to continue to emerge. For power searchers, who have a level of sophistication and knowledge of content and sources that goes beyond that of most enterprise users, traditional pricing models that emphasize and reward precision and sophistication will likely remain."


The aggregator/search service business is not a small industry. Although all the major aggregators are subsidiaries or units of publicly traded companies, and as such are not required to publish their revenue and profit figures, occasionally the industry press finds useful numbers.

For a brief period between 1997 and 2000, Dialog was a publicly traded company, both on the London Stock Exchange and the NASDAQ, allowing a rare glimpse into its results. The May 1999 issue of ONLINE, quoting a Dialog press release, showed 1998 revenues of $284.1 million and operating profit of $42.6 million. The Triangle Business Journal, in its April 14, 2000, issue, said that Dialog revenue for 1999 was $281.2 million. This article also reported that the company lost $8.7 million in operating income for the year. Since many industry observers felt that Dialog revenues were declining at this point, it seems safe to say that prior to the late 1990s, the company exceeded $300 million in annual sales.

Electronic Information Report, March 25, 2002, stated, "While Factiva, which is a subsidiary of Reuters and Dow Jones, does not report earnings, Dow Jones reports that Factiva generates about $250 million in annual sales with an annual growth rate of around 5.5 percent." This is not to say that all is rosy. The same publication on December 12, 1997, reported that then LexisNexis president and chief executive officer, Ira Siegel, "was forced to take an early retirement because of the unit's poor sales, which according to the Dayton Daily News, may decline 50 percent by the end of the year."


The aggregators developed their systems at a time when there was no standard for an online searching interface. Remember, these systems developed pre-Web, indeed, pre-personal computer and pre-Windows. Their respective providers saw the early interfaces as a competitive advantage. The early adopters, especially the search intermediaries, went through the sometimes substantial learning curve. Training, online practice, and careful pre-search planning were all part of the drill.

The aggregators were slow to change their systems to adapt to changes in technologies, or to meet different types of user needs. The complexity of the interface became the search services' hallmark. As search became ubiquitous, the aggregators took a somewhat predictable, but unproductive stance. At first, they viewed the Web as amateurish and probably (hopefully!) a flash in the pan. They viewed open Web content as not comparable with the content in their databases. While this is for the most part true, the typical Web searcher did not grasp or appreciate the significance of the difference. Now, in their efforts to update their interfaces and make them have more appeal, the aggregators are trying to look like the open Web—at least in terms of interface.

Aggregators are also looking to business development and partnerships to increase the use of their content. The recent announcement by Microsoft that it will integrate external content into the next generation of its Office products brings home the importance, not only of quality content, but also of search itself. Recently, LexisNexis and Siebel Systems announced a joint initiative to bring Nexis business, industry, and global news information, at the appropriate point-of-need, to users of Siebel's eBusiness applications.


The choice of online distribution channels allows publishers to focus on specific market segments, including the corporate or consumer user, specific industries, departments, even job titles. The 'big three" aggregators all targeted the corporate user—and all attempted to broaden the use of their products beyond their early core users.

However, experience and results have shown that, to varying degrees, the major online aggregators had difficulty moving beyond the core markets that they worked with going back to the 1970s. The early users of Dialog and Nexis were librarians and corporate information specialists, rather than the broader business users whom the aggregators desired. Of course, the early users of Lexis were lawyers. Factiva and its earlier Dow Jones and Reuters iterations, had strong name recognition based on exclusive Dow Jones and Reuters content, with both information and financial professionals. In terms of moving beyond the information professional, Factiva, with its Wall Street Journal coverage, certainly has the edge in the business world


Initially, only information intermediaries used aggregator systems, hence, marketing was directed at them. The individual who actually needed the information did not conduct the search. The information intermediary, often a librarian or other information specialist, conducted a detailed interview with the information requestor. These interviews would cover the purpose of the search, what work had already been done on the research subject, amount of information required, due date, nature of the content types that the "end user" was looking for, and other details, such as the format that would be most useful, and so on. The subject aspect of the pre-search interview would often take the form of a negotiation, with the discussion going back and forth to make sure that both parties were clear on exactly what the search topic was. Often, the intermediary would add value in terms of clarifying the question for the researcher, while the researcher could provide technical terminology, synonyms, known authors, and papers. This back and forth would then allow the intermediary to utilize their expertise with the online system, database selection, and choice of search strategy to deliver a precise, comprehensive and, most importantly, cost-
effective search.

Intermediaries viewed the complexity and cost of these systems as inappropriate for the occasional searcher. This viewpoint, whether perception or reality, made it very difficult for online aggregators to grow their business beyond the intermediary market. As time went on, the online aggregators sought to modify both their interfaces and their pricing policies to enable them to penetrate an organization beyond the intermediary.


Sometime between the emergence of the personal computer and the coming of the Web, the aggregators began to experiment with pricing and simplified interfaces to grow their markets. Commands were simplified, menus were introduced, core searches were canned, and alert or push services were emphasized as mechanisms to grow the businesses. As the Internet and intranets came into play, the search services developed tools for integrating external content into the intranet, enabling corporate users to locate relevant external and internal information in one place.

The notion of integrating content into workflow became common. In many cases, this approach has been more successful than asking busy knowledge workers to learn and access one system internally and another for external content. Vendors now offer a myriad of choices to provide content to the knowledge worker. For corporate users, a key factor is working closely with the customer contact team from the aggregator to make sure that the implementation really meets the needs of the organization. Issues like content preferences, search mechanism, depth and breadth of search, currency, and of course, pricing, require in-depth discussion with the vendor team. From the seller perspective, the team must contain the content, technical, and sales expertise to succeed in this complex process.


As the open Web has gained popularity, a new generation of content aggregators has emerged, offering consumer-oriented subject matter. These aggregators offer much lower price points than the traditional aggregators and typically target the home, not corporate, user. Interfaces reflect users' broad familiarity with Web search; some actually use the same search engines that deal with intranet or Web content.

One such offering is eLibrary, a product of HighBeam Research, a new-age vendor targeting the consumer segment. For flat fees ranging from $19.95 per month or $99.95 per year, eLibrary clients can access "hundreds and hundreds of full-text periodicals, newspapers, newswires, and a collection of encyclopedias, dictionaries, almanacs, and a thesaurus."

Another new player recently entered the consumer content space—Keep-Media. Louis Borders, founder of Borders bookstores and the ill-fated WebVan, founded KeepMedia. Working with Douglas Herrington, CEO of KeepMedia, Borders plans to leverage his experience with content and retail into the successful sale of premium content to the broad consumer market. KeepMedia launched with 140 well-known magazines and charges $4.95 per month.

In a conversation with Herrington, he stressed that KeepMedia sees the challenges in getting the public attuned to purchasing content as a "retailing problem." He wants customers to be able to easily shop and buy, have the right merchandise and selection, and, of course, with pricing acceptable to a broad audience. Herrington feels that when consumers care about a content issue, they want to deal with a vendor that provides a high "quality threshold" somewhat similar to the Nordstrom's model.

KeepMedia started its content pool with popular, highly regarded magazines that offer "powerful branding and well-known, strong content." Herrington told me that timing is everything: Publishers are looking at this new online venture as a way to reinforce the customer relationship and increase revenues from their archives. Data will flow to KeepMedia after the next issue hits the newsstand. In addition to online access, KeepMedia will also offer a mechanism for users to subscribe to print. Herrington feels that publishers have moved past the point of wanting to offer content via their own Web sites, recognizing that the cost/benefit of doing their own sites is weak.

The KeepMedia interface does not emulate the large search services. Rather, it supports customer browsing, keeping-up (alerting), and research on the same platform. KeepMedia offers a high degree of personalization that will lead to new displays of content based on past reading and searching behavior. Moving beyond collaborative filtering and customer profiling (which don't work, according to Herrington) to content profiling, KeepMedia will utilize the latest content categorization tools to provide a much higher level of precision and granularity than other "more like this" tools.

Writing in the July 24, 2003, issue of The Washington Post, Leslie Walker said, "Still, KeepMedia faces a huge marketing challenge to line up subscribers. Steven Brill's Contentville attempted something similar, selling piecemeal access to magazines, books and academic papers, before it closed for lack of readers in 2001."


Here's another key price variation: One aggregator will charge different rates for the content in different databases, while another tends to provide one price for virtually all articles regardless of source. Clearly, an aggregator's approach to the business model and its relationships with the publishers determine what makes sense for them. However, from the customer perspective, the difficulty of having hundreds of different prices for output is daunting and adds to the complexity.

A potentially larger point looms in terms of the effort to gain new users and increase revenues and market share. The most important difference between premium content aggregators and the open Web is the nature of the content itself. Why is premium aggregator content premium? What exactly makes it worth the amount of money the aggregator is charging?

Intermediate users of these systems were sufficiently familiar with the content and sources to understand that a refereed article is more authoritative than a press release; that business and popular literature are still more carefully researched, fact-checked, and edited than open Web sources; that patent literature is critical when doing certain kinds of research; that investment reports are more expensive than typical published articles; and that sophisticated access points and elaborate search strategies may save time even if the system is more expensive.

The aggregators generally provide considerable documentation and support materials, including highly detailed database descriptions that note the nature of the content, its creation process, typical applications, and unique value. However, the documentation is rarely available at the point-of-use when a novice online user is making a buy decision. Moreover, the meter is running, which precludes the idea of reading support materials to determine if you want to spend your carefully budgeted money on some particular output. The aggregators seem to expect that the user already knows the content and its value. While this may be true for the power user, experience shows that this is not the case with the much larger group of potential users. To today's Web user, all content is equal. In order to demonstrate higher value to a new generation of users, aggregators need to take a new and more aggressive approach to user education, particularly at the point of use.


What do aggregators need to do to make money? Here are the key points:

• Take care of your existing customers and leverage this base to grow use to affiliated professionals.

• Look to find ways to maintain a balance between cost and value delivered.

• Work with content suppliers to increase and define value and move to new price points that are in everyone's best interest.

• Take the steps to make users understand how content is differentiated and what different content types mean to them.

• Listen to your customers and give them the tools and interface features that increase functionality and loyalty.

• Look at current Web trends and features and both take advantage and go beyond them.

Therefore, I welcome the attempts of new aggregators to bring news, periodical, and other premium content to the vast numbers of people who are using the open Web to access information. Many of those involved in the information industry have called their products "access to humankind's recorded knowledge." I cannot help but agree that broadening this access is a good thing.


George Plosker [] is principal of George Plosker & Associates, a customer-centric consulting firm based in San Carlos, California.

Comments? E-mail letters to the editor to

       Back to top