Ah, The Numbers: Twist Anyway You Want
By David Kaplan - Sun 21 Oct 2007 10:12 PM PST
An NYT (FRB: 066570) piece focuses on the continued use of audience sampling via panels by NetRatings and ComScore, (NSDQ: SCOR) and its variance from server logs, an age-old debate. Among the major disputes the Times catalogs:
-- CondeNast’s internal count of monthly visitors to its Style.com site was 1.8 million last month. That differs significantly from the 421K visitors ComScore counted and the 497K measured by NetRatings.
-- Forbes.com claims it had 11.6 million U.S. visitors last month. That number diverges from both the 7.5 million estimated by NetRatings and the 5.8 million by ComScore.
-- CNN and FT.com also have expressed frustration with the differences between their reports and those of the measurement agencies. Still, Time Warner, (NYSE: TWX) CNN’s parent, is getting its own share of criticism from competing sites who charge that the news channel’s website rating skews audience numbers by rolling up the sites for CNNMoney, Fortune, Time, Sports Illustrated and Golf with CNN.com.
Last spring, Publicis Groupe’s Starcom MediaVest Group and WPP’s MindShare told several large, unnamed web publishers that they will reduce their ad spending on their sites if they did not resolve the discrepancies. Over the past several months, the Interactive Advertising Bureau has campaigned against the panel method. Randall Rothenberg, the IAB’s president and CEO, addressed the two ratings companies in April: “To persist in using panels that potentially undercount or ignore the diverse populations that are the future of consumer marketing is to deny marketers the insights they need to build their businesses.”
Naturally enough, ComScore and NNR feel that it’s a case of blaming the messenger and say it’s not their job to make numbers look better. As Manish Bhatia, NNR’s EVP of global operations and U.S. Sales, put it this way: “If the panel numbers were higher, I don’t think you’d be having this debate,” he said. Still, Scott McDonald, SVP for market research at Conde Nast, said the company has tried discussing its problems with ComScore, but claims the company couldn’t explain its panel methods.
Posted in: Advertising, Information, Metrics
