Study: Popular Sites Don't Necessarily Attract Valuable Buyers
Page 1 of 1
The most popular sites on the Internet aren't necessarily the ones e-commerce marketers and media buyers ought to be concentrating on, according to a new study from Reston, Va.-based online measurement service comScore Networks.
For instance, the Web site for The Detroit News attracts a class of visitors that spends more on average than visitors to The Wall Street Journal's Web site, according to the company.
The same is true in the portal sector, where market leaders AOL and Yahoo! attract less-valuable customers on average than AltaVista.
The study is based on comScore's measure of what it calls Buying Power, an index calculated by tracking the sites a user visits and the amount they spend online through e-commerce purchases. The statistic is compiled through the firm's netScore panel of 1.5 million Internet users.
A Buying Power Index of 100 correlates to the average Internet user, while a BPI of 367 (like DetNews.com) indicates that the average visitor to The Detroit News' Web site spends 3.67 times more than the average Web surfer.
Ideally, BPI information -- which comScore sells as part of its research offerings -- is aimed at ad agencies and advertisers as a way to enhance their media planning and to publishers to increase sales.
The news could serve as a wake-up call to Web media buyers, since it suggests that constructing an online media plan around high-traffic sites and targeted demographics may not be the best solution. It also could be good news for smaller publishers, which lack the traffic to compete with bigger names.
But panel-based studies have recently come under fire, most recently from the online unit of the Audit Bureau of Circulations, the de-facto standard of measurement among magazine and newspaper publishers.
According to ABCi, panel-based measurement firms often underreport sites' Web traffic. The firm said it analyzed several sites and found traffic 40 to 57 percent below what "leading panel-based companies" were reporting.
"Panel-based measurement is ideal for providing the detailed demographic information that advertisers seek, but it is limited in its ability to measure Web site traffic," said Dick Bennett, senior vice president of audit services at ABC Interactive. This particularly affects sites with "disproportionately large regional, educational, international or workplace audience bias."
"And under-counting traffic equals lost revenue opportunities for Internet publishers," he added.
Of course, ABCi isn't completely free of potential bias in the matter. The firm is pushing its new Reconciliation Analysis service, which uses an audited review of a site's visitor logfile, as well as its own third-party auditing practice.
But the allegation revisits a familiar charge -- that the third-party metrics firms routinely miss sizable portions of sites' traffic. comScore itself charges that competitors like Jupiter Media Metrix and Nielsen//Netratings undercount, while its larger 1.5 million-person panel provides for a more precise sample.
For their part, Jupiter and Netratings routinely contend that their findings are more accurate than comScore because of the way it gathers and incentivizes its panel to participate, although comScore says its smaller calibration panel is aimed at rooting out any errors in its larger sample.