To download the complete version of "The Story So Far: What We Know About the Business of Digital Journalism," a new report on digital news economics from the Columbia University Graduate School of Journalism, click here.
One way to counteract that is to try to sell more “branding” advertising on the web—that is, ads designed to draw positive attention to a company’s name and image, rather than to trigger a direct response to an offer for a product or service. Brand advertising has long been profitable for traditional media, especially television and magazines. But because publishers have struggled to convince advertisers that the web is a good platform for branded ads, they’ve often missed out on this lucrative revenue stream. “We had been told ad frequency of more than two or three exposures is a bad thing,” said Pettigrew. “But for brand advertising, it’s different. Six to twelve exposures [of an ad] increased the persuasiveness. And this is the most useful segment to our advertisers.”
Because Gawker Media has an array of sites on different topics, it can sell engagement across its network. Thus, the same readers who click on gadget stories at Gizmodo can be served with branded ads at Lifehacker or Gawker. And if Gawker Media can demonstrate to advertisers that its readers are loyal, it can charge higher ad rates, Pettigrew said.
At the other end of cultural spectrum, Public Broadcasting System’s web strategists are also using engagement metrics to increase revenue.
Amy Sample, director of web analytics for PBS Interactive, says she and others at the site modified a formula created by web analysts Eric Peterson and Joseph Carrabis to get a better sense of which readers were most devoted. They came up with their own criteria to determine PBS.org’s most loyal audience, based on the number of pages a reader views, the amount of time a reader spends on the site, and how often and how recently readers have come.
As it turned out, less than 5 percent of the visits on the site came from users who met all of PBS’s engagement standards. But those people are a critical group, says Sample. She found that they stay on PBS.org for 13.5 minutes per visit (compared with a three-minute average for everyone else) and click on nine pages per visit (versus three for other users). PBS saw economic benefits from this audience. Such users were 38 percent more likely to donate money to PBS than less engaged users; they were also more prone to encourage others to use the site. And when PBS saw the usage patterns, executives decided that video, a favorite platform for frequent users, should be promoted more prominently. That translated into revenue, because the site’s video ads get healthy $30 CPMs, Sample said, or about three times as high as other ads on the site.
Engagement correlates with editorial content. To see how the relationship plays out at a large site, we can examine some numbers for dallasnews.com, the main site for The Dallas Morning News. (These metrics are from the full year of 2010, a few months before the publisher began charging for access to much of the site.)
For the year, the site averaged around 40 million page views a month, driven by 5 million visitors who visit, on average, about twice a month and click on about four pages per visit. Those numbers are fairly typical for a site the size of Dallas’s and provided the publisher, James Moroney, some of the figures he used to calculate the rationale for instituting the pay-for-access plan (see Chapter Five).
But the broad numbers tell only part of the story. In fact, dallasnews.com, like many big online organizations, is many sites rolled into one. To analyze its data, the company sorts its traffic statistics into various categories, including by content areas: news, entertainment, sports, weather, and blogs.
News gets the most traffic, in terms of total visitors and visits. News visitors average around two visits a month and click on an average of about 1.5 pages per visit. Their habits are typical of those found at many other news sites—not particularly engaged.