|By the Project for Excellence in Journalism
The economics of the Internet remain journalism’s billion dollar mystery. And in 2006, that mystery only seemed to deepen.
Advertising revenues online continued to grow rapidly, but by year’s end there were already subtle signs that the growth rate was beginning to slow. And the medium still has a long way to go before it can begin to compare with the economics of the traditional media it was beginning to challenge.
At the same time, some of the euphoria about the medium’s potential is beginning to give way to harder-nosed questions about exactly how much things should cost. Some ad companies and industry analysts are beginning to demand a more accurate way to measure Web traffic.
Then there is another question, possibly even more baffling, to be sorted out — the idea that advertising works so differently online, the economics need to be rethought more completely.
The old journalism model of advertising was broadly based on audiences coming to look at content and staying to read or view the ads.
Online, that connection is looser. When people visit the Web for news, they often do so in short bursts, with little attention to ads. Advertising online is more like the Yellow Page listings (except for big display and video ads). People are looking for a good or service or researching a topic. (That is the commercial essence of search, and it also applies to classifieds.)
The poses little problem for online sites such as Yahoo or Google, in which search is part of the advertising structure. But it is a major issue for news sites, where ads are incidental to the reason people visit.
If that continues to be true — and some think targeted search will only grow — advertising may never grow to the point where it can pay for journalism on a scale to which we are accustomed. Those in journalism will need to develop a new economic model, not wait for this one to grow.
Online Advertising Revenue Over All
Growth in 2006
In 2006, the online advertising market appeared headed for yet another record-setting year. Through the first nine months, ad revenue reached roughly $12.1 billion, an increase of roughly 36% over the first nine months of 2005. By year-end, eMarketer projected it would reach $16.4 billion, a 31% increase.1
In all but 2 of the last 10 years, the year-to-year increases have exceeded 20%. Indeed, the increase in 2006 is in line with the kind to which the industry has become accustomed.
This extraordinary rates of growth are particularly impressive when compared with those in other media industries. According to Veronis Suhler Stevenson, a private equity and investment firm, advertising in “pure play” companies, or those who solely online such as Amazon.com, is projected to grow 20% from 2005 to 2010. That contrasts to the much more modest growth Veronis anticipates for other media industries: broadcast television (4%); cable and satellite television (10%); newspapers (2.2%); and radio (3%).2
Other survey research suggests that as the dollars in online ads rise, so too does the array of advertisers. According to Outsell, a California-based research firm, fully 8 in 10 marketers say they are advertising or planned to do so on the Web in 2006. That number is expected to reach 90% by 2008. Moreover, the same study also reported that smaller companies with ad budgets of less than $1 million are allocating more of their budgets online.3
For the moment, however, the largest Web companies, particularly the most popular Web portals — are reaping nearly most of the benefit. According to Pricewatershouse Coopers, the biggest 50 Web companies, including Yahoo and AOL, are attracting 96% of the ad spending, the Wall Street Journal reported in late 2005.4
But when all is said and done, only a small share of total advertising dollars are channeled to the Web. In 2005, advertisers spent roughly $11 billion on pure-play Internet companies, a relatively small number compared to television ($65 billion), newspapers ($55 billion), and even radio ($20 billion), according to figures compiled by Veronis Suhler Stevenson.5
Market analysts have also looked at all this another way. According to TNS Media Intelligence, the top 50 advertisers are spending just 3.8% of their budgets on online ads. While the figure is higher for smaller advertisers, 6.8%, the data emphasize the relative immaturity of the online ad industry, but also its potential for growth.6
To some analysts, the amount of advertising on the Web does not correlate to how much time people spend there. A Veronis Suhler Stevenson study found that consumers were spending 17% of their media consumption time with the Web. But the medium made up just 8% of total ad expenditures.7 According to Pam Horan, president of the Online Publishers Association, this represents a “misalignment,” and more money could be steered to the Web once it is corrected.
That gap may suggest more reason for growth, but at least some analysts think the pace can’t continue. As the base being measured gets bigger, growth rates of 30% a year, they reason, just aren’t sustainable.
The market research firm eMarketer, for instance, projects online ad spending to grow 15% in 2007, 18% in 2008, below 10% in 2009, and then 7% in 2010.
According to Emily Riley, an analyst with Jupiter Research, “The growth is trending downward. We’re no longer seeing hyper growth rates.”
If that continues, it creates a difficult scenario for journalism. The finances of the old media platforms are expected to flatten. But it is not clear when, if ever, the new media will compensate.
New Economic Model
That points to the trillion-dollar question about the Web: Can journalism ever figure out a way to “monetize” the new platform?
A year ago in the newspaper chapter of this report, we estimated that at current growth rates, it would take more than a decade for the Internet to match the revenue of newspapers.
The new estimates about advertising growth cast serious doubt about whether that can happen anytime soon. Certainly, no one imagines the Web can sustain 31% annual growth rates in advertising for a dozen more years.
Now even more questions exist about the nature of advertising on the Web.
Historically in journalism, audiences come for news and are exposed to advertising incidentally. Online, hunting through advertising is an activity unto itself. Advertising here works more like yellow pages, or classified advertising in the newspaper. It is a form of research, often targeted.
In the new platform, in other words, journalism and advertising are more loosely connected. And display advertising, which grabs the eye, creates an image, reinforces a brand, and imparts a little information, is only a small part of the Web advertising experience.
That raises doubts about whether journalism can survive in a form we are familiar with if the Web remains an advertising-based environment for news. Increasingly, the math on Internet advertising suggests that the news business must find a new economic model, one that gets consumers to begin to pay for content.
One possibility, the one companies are experimenting with, involves trying to persuade consumers to begin to pay for premium content. The New York Times, with its Times Select program, asks consumers to pay for its editorials, columnists and archives. But the success here has been limited, and the most expensive aspect of journalism—the basic news gathering, is not covered.
Another scenario that has had even less success is trying to get consumer to buy subscriptions for all content online.
Increasingly, a different model, one that the industry has not pursued as publicly, or even perhaps as aggressively, seems to have more merit.
In this model, rather than charge consumers directly, news providers would charge Internet providers and aggregators licensing fees for content. The Internet providers would likely turn around and add those fees to the bills consumers pay for Internet access. Aggregators could also be involved in the fee system, sharing revenue with news content providers.
News organizations may have to create consortiums to make this happen, which is a complicated but not necessarily difficult step. But trying to negotiate among a dozen or so internet access providers is far more workable than trying to win over individual consumers, one at a time, to pay for something they currently get for free. Internet access providers also have an incentive to make this happen. They recognize that if journalism shrivels, there will be less for people to want to access online. Their business depends on the news. Aggregators are even more aware of this.
And the notion that the Internet is free is already patently false. Consumers pay every month to get access.
This is the model that exists in cable. The cable companies pay a fee to the channels that they carry, and they pass the cost of that fee along to the consumer.
There are differences between cable TV and the Web — far more content producers — but relatively few of them actually are in the business of producing content that is expensive to produce and passed along free. Journalism is at the center of that group.
This would be a major shift in emphasis for the business side, which is one reason perhaps it has not been aggressively pursued. But the mandate for a new economic model now seems clearer than ever before. And the idea that news producers can be passive in this increasingly seems like a fatal error. If one were inventing the financial system for the Internet a decade ago, instead of it evolving by happenstance, the notion that those who provided the funds.
The Challenges of Measuring Online Traffic
In addition to the complications of who will pay for the Web and the future of advertising, in 2006 another set of questions came into clearer relief — growing doubts about how to measure the audience.
For years, the Web’s appeal to advertisers has been built in part around a greater sense of accountability than in older media about who is actually paying attention to the advertising content.
But observations that there were holes in this claim have begun to grow over the last few years.
Roughly a dozen firms track online use, but there is no standard method for doing so. The result, according to Rich Gordon, a professor at the Medill School of Journalism at Northwestern University, is that some metrics inflate the absolute number of unique visitors, or unduplicated visitors to a Web site, while others undercount it.
Such discrepancies began to attract more attention, and more concern, as 2006 wore on.
As an example of the problems, two of the most important firms measuring online audience, comScore and Nielsen, determine a Web site’s traffic by assembling a panel of online users considered representative of the overall online audience. But there are questions about how representative the samples are. Panel members must install specific software on their computers. But since some corporate offices and government agencies do not allow their staff members to install it, those segments of the population may not be fully represented in the samples. That is also true for many universities.
There are also concerns about something called click fraud, which occurs when a user or computer-generated program clicks on an online ad without the intention of actually viewing the ad. By one estimate, as much as $800 million could be wasted by advertisers on fraudulent clicks.8
Finally, one other potential curve ball is the consumer tendency toward media multitasking. Some worry that online consumers may not be as engaged as they are with other mass media. According to the Internet marketing research company Big Research, nearly 7 in 10 Web users report watching TV while they surf the Internet. “Suddenly a once captive TV audience has now defined itself as a moving target,” said Phil Rist, the vice president of strategic initiatives for the company.
As a result, some marketers, including Ford, Visa, and Colgate-Palmolive, have started asking for audits of online audience figures. In the words Dave Morgan, CEO of Tacoda, a behavioral and targeting ad network with headquarters in New York, the success of an advertisement placed on a Web site will more likely be “decided by the accountants, not the chief marketing officers.” If such a system is put in place, how could it affect the overall economics of the Web?9
The issue is unlikely to be resolved quickly, or easily. The fact that the technology of Web pages can change so quickly means that how it is measured will have a hard time keeping up.
Main Categories of Online Advertising
To better understand the economics of the Web, it is useful to know the different kinds of advertising it attracts. Three types of advertising account for roughly 90% of the ads on the Web — search, display and classifieds.
Search ads are more targeted than display or classified because they appear only when a reader has searched for a relevant topic. In the first half of 2006, search advertising surged, increasing 37%, to $3.2 billion, according to data released by the Interactive Advertising Bureau and PricewaterhouseCoopers.10
The surge kept it at pace with the first half of 2005, accounting again this year for 40% of all online advertising. What’s more, the gains are expected to continue. Jupiter Research, another forecasting agency, projects that search advertising will continue to be the most lucrative area online, forecasting a compound annual growth rate of 12%.11
Classified advertising is the second major category is. In dollars, the category grew 56%, reaching $1.6 billion in the first half of 2006. Classified has slowly increased its share of all online advertising for several years and now accounts for 20% of online advertising, up from 18% in 2005.12
Display advertising, involving graphical banners placed on a Web site, is the third category. Display ads are less targeted than search ads because their placement is not influenced by the online viewer’s behavior or search criteria. Their dollar amount increased 21% in the first half of 2006 to $2.4 billion. The increase, while large, was below those of search and classified ads. Display ads now made up 31% of all online advertising, down from 34% in the first half of 2005.13
The Interactive Advertising Bureau, an association that publishes the most widely quoted data on the online advertising industry, has traditionally categorized rich media as a subcategory of display. Rich media is a broad term that refers to digital, interactive media and can include video, graphics, text, animation or audio. Almost all discussion about rich video relates to video, which has moved the online industry from a largely text-based medium just five years ago to a more visual one now.
But there are still limits here. Rich media grew just 3% in the first half of 2006, accounting for $475 million during that time. Along with high production costs and questions about how the audience is measured, such sluggish ad revenue growth, especially compared to other online advertising categories, probably explains why rich media now makes up 6% of all online advertising, down from 8% a year earlier.14
To get a clearer picture of online video and video advertising it may be helpful to first look at the market of online video users and then at the ad dollars placed there.
The Video Picture
How many Americans are watching video online? Last year we reported that online video use, particularly regular use, was still relatively small. One of the biggest Internet success stories of 2006, YouTube, suggests that the medium made huge strides. Time Magazine’s “Invention of the Year,” the site was purchased by Google in October 2006 with a reported audience size of 30 million unique visitors.15
But the overall growth in online video use is a matter of debate. According to the Online Publishers Association (OPA), there was no growth and even some decline. In its February 2006 survey of Internet users 12 to 64 years old, 5% of respondents said they watched online video daily, the same as a year earlier. On a weekly basis, the number of Americans reporting they watched video actually fell three percentage points and now stands at 24%. And a smaller percent reported ever watching online video (69% versus 74%). The only positive movement was in the number saying they were “aware” of online video.16
Other audience data, however, suggest that use of video on the Web is not only growing but substantially so. For instance, figures from comScore Networks, which combines both survey research and a panel of select viewers that give the organization permission to record their online behavior, found that the number of Americans viewing video online increased 18% from just October 2005 to March 2006. The study found that Americans consumed 3.7 billion video streams and around 100 minutes of video per viewer on a monthly basis.17
What type of video are people watching? The most popular destination for online video viewers is weather, followed by entertainment and then music videos. News and current events video was considerably lower on the list, according to the OPA survey.
The top video sites are owned by some of the largest media corporations. But much of the content distributed over those sites is user-generated, as is the case with MySpace, which is owned by the News Corp. and YouTube, now owned by Google. Here are the top video properties as of August 2006, according to comScore Video Matrix:
Top 10 Video Properties Ranked by Number of Streams Initiated
Source: comScore Video Metrix, August 2006
Though not ranked in the top ten, video from the AP Online Video Network, which offers up to 40 video clips a day of news, is becoming increasingly available to Americans, particularly on newspaper sites, where online readership has been growing rapidly. One estimate put the total audience for AP video at 76 million just six weeks after the service was launched in March 2006.18 Also, the New York Times Web site, among the top 10 most popular news sites, offers a video channel and announced plans in February 2006 to syndicate video to other Web sites.
As of now, online video remains a relatively small piece of the pie. It currently accounts for just $410 million of the $16 billion online ad market.
Over the next few years, though, some analysts predict strong growth. The market research firm eMarketer, for instance, projects that online video advertising will grow to nearly $3 billion by 2010. If it reaches that number, video will account for roughly 12% of all online advertising by the end of the decade.19
There was considerable speculation that the Google-YouTube deal would be a big boost to both video use and advertising. Because of Google’s ability to organize and aggregate information — in this case video — it would be a way for users to more easily search and find the video content they desire. As Diane Mermigas, a contributing editor to TV Week, wrote:
Oddly enough, the surge in online video advertising is projected despite what appears to be a declining number of Americans who say they have seen online video ads. Two thirds (66%) said they watched an online video ad in 2006, down from 70% who said so in 2005, according to survey data from the Online Publishers Association.21
Local Online Advertising
More strong growth in 2007 is expected in the local online advertising industry. Borrell Associates is projecting local advertising to grow 32% in 2007, to $7.7 billion. In particular, local advertising will become more targeted, Borrell forecasts, with search advertising increasing from 24% of all local advertising in 2007 to 44% by the end of the decade.
Right now, the largest percentage of local online ad dollars goes to newspaper-based Web sites. Forty-one percent of online ad dollars is spent on newspaper sites, followed by pure-plays (25%), Yellow Pages (16%), and paid search companies (9%).22
As a sign that local advertising is becoming even more important to media outlets, survey research from Borrell shows that about half of all local news sites added to their sales forces in 2006, increasing the staffs by about 37%; some of the largest local Web sites now have three to four dozen salespeople dedicated to online sales.
It should be noted, however, that local online advertising remains a very small part of total advertising. In 2005, it accounted for just 1.8% of all advertising dollars and even with strong growth expected this year and next, it will account for just over 2% in 2006.23
Ad Dollars and the News
In the first three editions of the annual report we were not able to report on what percentage of total online ad dollars was being spent specifically on news.
Now, data from TNS Media Intelligence separate ad dollars according to specific types of Web sites. Their data are based only on display advertising, which as mentioned above accounts for just 31% of the total ad dollars online, but the figures at least allow a first sense of the divisions.
According to the data, news may be the second-largest recipient of online ad dollars, trailing only portals and search engines. In 2005, U.S. marketers spent roughly $763 million on news and current events Web sites, compared to $1.1 billion spent on sites like MSN, Yahoo and AOL.24
Revenues and Profits
In the past, large media companies have struggled to generate revenues and profits from their traditional properties. Though revenues from digital operations are contributing to the bottom line at many companies, there are still a number of questions about the online business model.
For newspapers, the dominant economic model is largely dependent on advertising. The numbers here continue to grow. In the second quarter of 2006, newspaper online ad revenue increased 33% over the second quarter from 2005, according to the Newspaper Association of America.25
A number of public newspaper companies themselves are reporting extraordinary revenue growth in online advertising. At the Washington Post Company, online revenue, primarily from washingtonpost.com, increased 31% to $73 million for the first nine months of 2006.26 At the New York Times Company, whose digital properties include 18 daily newspapers as well as About.com, online revenues for the first nine months of 2006 were $190 million, up from $134 million a year earlier.27
A study from Borrell Research, meanwhile, found that large newspaper sites, defined as those with gross revenues of $5 million or more, now contribute anywhere from $3 million to $14 million to the bottom line.
And more online newspaper sites have become profitable ventures. In 2003, more than 6 in 10 (62%) of newspaper Web sites reported they were not yet profitable. By 2005, that number was just 5%.
Just how profitable? That, of course, is a matter of accounting — or how much cost is attributed to the online site versus the old-media operations. But for those profitable sites, according to Borrell Associates, the average profit margin was 62%.
In 2006, we saw signs that the ad-based model could increasingly be the future for newspapers. First, newspaper companies signed a deal with the other major news aggregator, Google, which would help sell print advertising in many of the country’s largest newspapers.
Then in late November, seven newspaper companies that publish newspapers in 38 states formed a partnership with Yahoo. According to a press release, Yahoo will help newspapers “deliver search, graphical and classified advertising to consumers in the communities where they live and work.”28
Several newspaper analysts applauded the Yahoo deal: “I’m used to the newspapers being very reactive, and here they’re stepping out ahead for a change,” said John Morton. “Here, I think they’ve become much more agile in trying to adapt to things that are happening.”
Even solid growth from online ads, however, can’t be considered a panacea for newspapers’ economic woes, as the industry continues to experience shrinking circulation figures and marginal growth from advertising in its print form. Even when online revenue is used to buttress anemic growth from print ads, combined revenues were almost flat in the second quarter, growing just 1%.
And the contribution of revenue from online properties to total newspaper company revenues will remain in the single digits in both 2006 and 2007, according to Borrell Associates. For major public newspaper companies, online revenue will account for 6% of total revenue in 2006, and then climb to 7.4% in 2007.29
The economic picture for other media industries is not necessarily as promising. The economics for online local television do not appear as successful as those of newspapers, but they are improving. The percentage of local television Web sites that reported making a profit in 2005 rose to 24%, up from 15% a year earlier, according to survey research conducted by the RTNDA and Ball State University.
The number of radio stations reporting a profit, meanwhile, did not change, according to the same study. Just 4% of all radio Web sites reported a profit, the survey found.30
When it comes to the online operations of cable news, most of the data are anecdotal. Few companies break anything out.
Those anecdotes offer only a few hints. Kyoo Kim, MSNBC.com’s vice president of sales, told the New York Observer the Web site may be earning more in monthly ad revenue than the cable channel. That claim was not verified, and it should be noted that the cable channel also earns at least half of its revenues from subscriptions. The Observer estimated that the online display ads alone make tens of millions of dollars for CNN.com and MSNBC.com each month.31
According to Rupert Murdoch, Fox’s Interactive revenue will top $660 million by the end of the year — but serious money is still way, way off and profits may still be elusive. In July, the News Corp. division that includes digital assets reported $1.4 billion in revenue for the previous 12 months (the company does not break out interactive revenue), a figure that is one quarter what the company’s movie studio made over the same period and makes Fox Interactive part of the company’s second-smallest division. Moreover, the division lost $150 million in operating income over the previous year, the only News Corp. division to be in the red. 32