AJR  Features
From AJR,   June 2000

A Tough Sell   

News Web sites are having a hard time attracting enough advertising to bring them into the black. That's not likely to change anytime soon.

By Paul Farhi
Senior contributing writer Paul Farhi (farhip@washpost.com) is a reporter for the Washington Post.     



SOMEDAY SOON, writes Time Inc. Editor at Large Daniel Okrent, the jig will be up for news on paper. The combined magic of digital, satellite and cellular technology will enable journalists to beam text and images to readers and viewers anytime, anywhere, via portable personal information appliances, he wrote last December in a commentary widely circulated on the Internet. Readers will like it, but advertisers will absolutely adore it, he believes: "[With] a truly interactive medium...the effectiveness of media advertising changes radically." Imagine, he says, a medium supported by ads tailored specifically to a reader's age, gender and known buying habits.
Futuristic? Actually, with a few refinements, Okrent could be describing the present. Much of what he imagines is already here, at least in theory. It's called the Internet. The Net should be a marketer's dream. Perched over their ubiquitous personal computers at work or at home, online readers can get news and information instantly, at any hour of the day--now. Most crucially, from an advertiser's perspective, the Internet's interactive nature enables participants to make instantaneous purchases. What's more, technology is available to determine exactly who's browsing around a site--a competitive advantage that newspapers, TV and radio companies will likely never enjoy. For advertisers, "the Internet is a broad-reach medium like TV and print," says Susan Bratton, vice president of market development at Excite@Home, which owns the popular portal site Excite, "but it's so much more than that."
In theory. In many ways, the Internet is turning out to be something less than the advertisers' paradise it seemed. While online advertising may someday rival or even supplant the oldline kind--virtually every news organization in America is investing in the Internet on that economic premise--how it will get to that point seems less clear. Questions abound about the effectiveness of ads online, about privacy and ethical issues, about, quite simply, whether advertising can support so many sites.
It's striking that virtually none of the many hundreds of Web sites that post original news and commentary are profitable (although a few, such as MSNBC.com, say they're getting close). That's because it's still early. In the meantime, however, Wall Street's enthusiasm for the concept has nearly died. Long before the recent decline in technology stocks, Internet content pioneers like CBS MarketWatch, TheStreet. com, Salon and the women's site iVillage.com saw their shares hammered to levels at or below their initial public offering prices. A handful of specialty news and information sites, such as the health-news and advice site drkoop.com, are flirting with bankruptcy. Indeed, after several years of giddy stories about the bright future of news online, something new is cropping up: doubt. "I've never been convinced that the ad-supported model works," says a top marketing executive of an Internet company, a veteran of the daily and weekly newspaper business, who asked not to be identified. "It's still untested."
Some analysts predict that a shakeout is imminent. "Ninety percent of the [commercial] Web sites you see won't be around in a year," says Charlene Li, who tracks Internet advertising trends for Forrester Research of Cambridge, Massachusetts. The meltdown will primarily affect "all those undifferentiated e-commerce sites," she says, but news sites will be in for some tough going, too. "It's difficult to sell advertising around news," Li says. "We call news [sites] the lost souls of the Internet."
Since few publishers charge for online content (and very little e-commerce has developed on news sites), advertising appears likely to be the primary support for news on the Web for years to come. Charging for content appears to be a nonstarter; with the exception of the Wall Street Journal, no major news organization has been able to generate much revenue from asking Web surfers to pay for news.

In fact, the trend is moving in the opposite direction. TheStreet.com last year decided to expand the amount of content it gives away, while Slate abandoned paid subscriptions altogether. The San Jose Mercury News, located in the heart of Silicon Valley, at one point tried charging subscription fees. But the response was so disappointing that the paper cut its asking price. Ultimately, the Merc recognized an established fact of the Web: except for pornography, few people are willing to pay for content online. At least for now. "It's unclear what the ultimate [economic] model is going to be," says Bruce Koon, who pioneered the Merc's Web operations before stepping up to parent Knight Ridder.com. "Yes, people are accustomed to getting it for free, but we don't know what the future will bring. Everything on the Internet still needs that frame around it."

THE GOOD NEWS IS that online advertising has been growing at a fast pace. Although Internet and computer companies still account for about half of all online ad spending, many new advertisers took their first big plunges online last year, according to Competitive Media Reporting, which tracks ad expenditures across media.
How much advertising is actually out there? It depends whom you ask. The Internet Advertising Bureau, a trade group, estimates that advertisers spent $4.62 billion on Web ads last year. However, Forrester Research, whose numbers prop up many an article about the Internet economy, pegs online ad spending at $2.8 billion in 1999. And Media Metrix, which calls itself the "most comprehensive [and] reliable" Net measurement company, says the number is $2.8 billion--for the second half of the year alone.
The companies say they use different methodologies to arrive at their estimates. They also point out such estimates entail much guesswork, since privately held Web companies rarely disclose their revenue. What they don't say is that the guesswork is dicier because in many cases cash doesn't change hands in Internet ad deals; many ads appear as a result of trade and barter arrangements, cross-promotions among allied sites and outright giveaways to fill up space.
The question of how big the online ad pie is would be academic if it didn't suggest something more important about the Internet's immaturity as an ad medium. That is, it's hard to get reliable--or at least consistent--marketing data about many things about the Web. For advertisers who've built a consensus around Nielsen's TV ratings, Arbitron's radio rankings and the Audit Bureau of Circulations' newspaper figures, the Internet may still look like the Wild West. "The [online] numbers never agree, but they always seem to point in the same direction," says Gary Arlen, a veteran consultant to Internet companies. "With all due respect to my colleagues in the research racket, you're more likely to sell studies to people who want good news. There are a lot of people [involved in Internet businesses] who are invested in wishful thinking."
An advertiser, for instance, might reasonably want to know how many people a Web site attracts. Almost all advertising, after all, is priced on the basis of how many people will see an ad, and who these people are. But here again, it depends whom you ask. In February, Nielsen/Net/Ratings, the sister service to the Nielsen TV service, estimated that MSNBC.com was the most popular all-news site, attracting 5.3 million unique visitors at home and 5 million at work. During the same month, Media Metrix said MSNBC.com had a total of 8.6 million unique visitors--a gap of almost 20 percent from Nielsen's estimate. In the same month, Nielsen said MSNBC.com's big rival, CNN.com, had 2.8 million unique visitors at home and 3.1 million at work. But Media Metrix's number for CNN.com was 4.8 million.
Who's right? Maybe neither. Merrill Brown, editor in chief of MSNBC.com, says Nielsen and Media Metrix undercount MSNBC.com's audience because they have trouble measuring the audience at work, where MSNBC.com is stronger. Both Nielsen and Media Metrix deny this; both say they monitor the behavior of Net users at home and at the office.
But that's about the only thing the two services agree on. Bitter rivals, each regularly questions the accuracy of the other's numbers. Accurate measurement of Net users "depends on who has the cleanest sample and the most integrity," says Media Metrix spokesman Max Kalehoff. "We think we do." His counterpart until recently at Nielsen, Rich Coyle, fires back that Media Metrix's surveying methods are rife with "survey bias."
Ultimately, accurate, independent verification of traffic is critical. Many companies won't advertise at all on a Web site, or will pay reduced rates, if they think third-party data is unreliable. And doesn't the crossfire between the two rival services leave advertisers confused? Concedes Coyle: "There are so many dynamics at work that, yes, it does get confusing. The Internet is still very young."
To derive their traffic estimates, both companies track the habits of randomly selected panels of Net users. This is the same basic technique behind all survey research, be it a poll of voters or Nielsen's canvass of the TV audience. But determining the behavior of millions of Web users through this method poses special challenges.
Harold Simpson, vice president of research at the Television Bureau of Advertising, which promotes TV as an ad medium, notes that complaints about the accuracy of Nielsen's TV ratings have grown over the years as more cable networks have come on the air. As the number of viewing options has increased, says Simpson, the number of respondents watching any particular station has gotten smaller, and thus less reliable.
The Internet magnifies this problem. If it's hard to give good readings on 60 TV channels, imagine how trustworthy the data are when they're spread across millions of Web sites. "You've just multiplied the difficulty by a staggering amount," he says.
The vastness of the Web also creates a fundamental business problem: It's hard to maintain ad prices in a market as crowded as the Internet. It's not easy to start a new daily newspaper, a cable network or a radio station, but it's a snap to set up a Web site. And since Web sites aren't subject to old media's physical limitations (press capacity, time constraints), their ad holes are effectively bottomless. This endless supply gives advertisers huge negotiating leverage over Web-site operators.
The wide open spaces of the Web also fragment the market, making it difficult for advertisers to address anything like the mass audience of TV or newspapers. On any given day, even portal sites such as AOL.com, Lycos or Yahoo! won't collect more than 15 percent of the total traffic of the Net, says Forrester's Li.
That explains why the price of the average Web ad, judged on a cost-per-thousand-users basis, hasn't risen for the past 18 months, according to Excite@Home's Bratton, who also serves as vice chairman of the Internet Advertising Bureau. Ad prices "aren't rising because inventory keeps growing," she says. According to Digitrends, a Web site for interactive marketers, portals such as Alta Vista, Lycos and Excite will run banner ads for as little as $1,000 to $1,500 a month, or less than what a major metropolitan newspaper gets for a one-time, quarter-page display ad.
Although this makes Web advertising more accessible to more advertisers, it also means Web salesmen have to sell more just to stay even. Among news sites, a big audience alone won't guarantee big dollars, says Marc Ryan, director of media research for Ad Relevance, a research firm. Instead, sites that offer "focused" content (such as financial data) and attract a well-heeled audience tend to command the highest rates.

"DISAGGREGATED," OR SPECIALIZED, information may be the best bet for Internet survival, some observers suggest. In other words, the old newspaper model--providing a supermarket of general news and features--may not work in a world in which Net users are clicking around. "If you fly across the U.S., local newspapers tend to look and read the same," says Li. "They all have the same AP and Reuters stories. The only unique thing they offer is about four pages of local news. So, if I want national news, I don't need my paper. The trouble is, I can get my national news from a million places on the Web, like Yahoo! The thing that will bring me to a specific site often is the information that isn't generic. Someone may go to Yahoo! once a month, but they'll visit the fly-fishing site frequently."
Li agrees with Larry Kramer, chairman and CEO of CBS MarketWatch, who says the Internet is unparalleled in its ability to gather together communities of people with a common interest--and the advertisers who want to reach them. From an advertiser's perspective, this is highly efficient. "We're not looking to get the big, full-page ad for Macy's," Kramer says. "But there is no question in my mind that we will own the classified ads someday."
But is anyone paying attention to ads on the Internet? To be sure, that question has dogged the old media for years, too. Are people running to the kitchen or the bathroom during commercial breaks? Are magazine readers paging quickly past those vivid four-color ad spreads? Are traffic-bound drivers punching their pre-set buttons like jackhammers every time a commercial comes on the radio? Some research indicates that the Internet has those same defects. The speed with which Web surfers can browse seems to work against advertisers' efforts to roadblock would-be customers.
In fact, visitors may come to a Web site, but it's not clear that they "drill down"--the term of art for clicking on many pages within a site. According to Nielsen, homepages account for an immense percentage of the traffic on most sites, which indicates that many visitors never see the ads located well inside the site.
For example, Nielsen said that 95 percent of the visitors to washingtonpost.com in February came to the site's homepage. But no other part of the site attracted more than 15 percent of the traffic; about 2 percent went to the financial page and a mere .11 percent clicked through to Newsweek's site within a site, according to Nielsen. The pattern is repeated throughout the news category. The homepages of the Wall Street Journal, USA Today, the Chicago Tribune and the New York Times all attracted the lion's share of traffic, with the rest of these sites getting very small fragments of this audience, Nielsen says.
For this reason, Internet executives often talk about making their sites "stickier"--creating content that keeps visitors hanging around longer. "My challenge is not only to get a large audience, but to get them to come back again and again," says MSNBC.com's Brown.
Even if visitors linger, though, the amount of interaction between advertisers and viewers has often been limited. Nigel Hollis, executive vice president of Millward Brown Interactive, a San Francisco online ad firm, says click-through rates on banner ads have been dropping for months. Hollis estimates that a meager .6 percent of Web visitors actually click through, down from 2 percent in 1997.
"People are saturated by them," he says. "They've gotten used to online advertising. A couple of years ago, it was ŒOoh, this looks interesting. Let's see where this takes me.' " Now, plenty of other options beckon.
To counter these limitations, news operators on the Web are trying to diversify revenue sources. APBnews.com, which covers news about crime and the criminal justice system, may be typical. Advertising is still its main source of revenue, but the site also has about 30 "syndication" clients that pay to reprint its stories. It announced recently that it is joining with book publisher Prentice Hall to create subscription-based, college-level courses via the Internet. And APBnews.com just added an electronic store, selling hats, T-shirts and books. As prosaic as that sounds, e-commerce creates ethical dilemmas, notes Joe Krakoviak, a spokesman for the company. "We are sensitive to the idea that we might be doing something in our content that could help our e-commerce side," he says. "There has to be a crystal clear distinction about what is news and what is commerce." Another grand promise of the Internet--the ability to identify who is coming to a site--is also a big fog at the moment.
Some sites, such as the New York Times', require visitors to sign in, which provides minimal identification. And many sites automatically exchange "cookies," special computer codes, with visitors. This enables a site to determine whether that computer has visited before, although not who is actually operating the computer. Some users will surrender bits of personal data if given incentives to do so, says Millward Brown's Hollis, but most prefer to remain anonymous.
Indeed, efforts to identify Web users have been met with hostility. Earlier this year, DoubleClick Inc., the Internet's biggest ad seller, unveiled plans to merge millions of cookie files with a database of names and addresses, in a massive attempt to identify anonymous Web surfers. The plan was so roundly criticized, by individual users and Web sites alike, that New York-based DoubleClick was forced to back down.
The company, which is now under investigation by the Federal Trade Commission for possible violations of privacy laws, declined to comment for this article. But in an op-ed piece in the Wall Street Journal, company Chairman and CEO Kevin O'Connor wrote that handcuffing the advertising firms' ability to identify consumers' preferences threatens the free use of the Internet, since "the vast majority of Web sites today lose money." Simply broadcasting ads to masses of anonymous people won't work, wrote O'Connor, suggesting the Internet should be used to collect, as well as to send, data for and about consumers. Despite his ardor, he has promised to wait for government approval and new industry ethical guidelines before moving ahead with DoubleClick's original plan.
Some technology proponents--like Time Inc.'s Okrent--believe we haven't really seen anything yet. Someday, they argue, Web users will enjoy greater bandwidth that will make surfing the Internet faster, easier and more rewarding. This enables the traditional news "product"--text and photos--to grow to include sound and video over the Web. "It's going to mean the consumer will have more options on how to absorb and take in news and information," says Knight Ridder's Koon. It will also likely change the role of journalists: Staffers at the Washington Post were amused and intrigued not long ago to receive a memo from Managing Editor Steve Coll suggesting that the Post's extensive forays onto the Web might someday require reporters to cover stories with video cameras on their hats.
In the future, tech fans say, the Internet will become truly ubiquitous, as sophisticated handheld devices free users from their desktops. While it's always hard to see the future, it's difficult to see how Internet advertising will overcome its present limitations and become the force that conquers the traditional media.
"Maybe at the end of the day, we're going to find that Internet advertising has more in common with direct-mail advertising than with TV or print advertising," says former Nielsen spokesman Coyle. "It may be a question of the Internet finding its place in a broad mix of media. This is a new medium. People are still trying to figure out what works and what doesn't."

###