Publishers leery of audits

By Published on .

Some say it will be difficult for BPA's new Web analytics data to replace current reporting When BPA Worldwide statements were released in June, they included a new section of data: analytics on 445 Web sites from its member companies. Glenn Hansen, president-CEO of BPA, expects that number to double by the next audit, and to be at 1,500 when the company's fiscal year ends next June. BPA announced the plan to partner with Nielsen Online and get into Web metrics last fall, and opened up the program in January. While some publishers are excited to have all their print and online numbers centrally audited, others are concerned about another issue. “Our numbers are going to be lower than any other numbers that you get from any other source,” he said, “whether [they are from] Google or any commercial Web-analytics company.” The words “lower” and “numbers” in the same sentence are generally not what publishers and advertisers want to hear. The lower numbers exist, Hansen said, for two reasons. The first is because BPA has built an extensive search engine spider and robot list. Spiders and bots are automatic Web crawlers that visit sites to collect content information for their searching capabilities. The BPA's spider and bot list is able to deliver more accurate numbers because it delves more deeply and broadly into Web site numbers than typical b-to-b publishers do, Hansen said. The second reason BPA's Web traffic numbers are lower than other Web analysis tools is that the organization eliminates all internal hits from the final numbers. The discrepancy in numbers is creating an issue for some publishers. “I think BPA is a little late getting into this game,” said Debbie Winders, VP-circulation and distribution at IDG. Advertisers have not been asking her for BPA audits of the Web sites, she said. “Until they do, we will continue with things the way they are,” she said. Bobit Business Media has the BPA/Nielsen code on six of its more than 25 Web sites, and the numbers from BPA consistently come in lower, according to Tony Napoleone, the company's audience marketing manager. “As convenient as it would be to have Web traffic on our BPA statements, publishers are not going to be interested in doing so if their traffic figures have to take a big hit,” he said. “No one wants to be out in the market with conflicting numbers,” said Nick Cavnar, VP-circulation at Hanley Wood Business Media. “There's nothing wrong with the BPA numbers. ...[But] most publishers have quite an investment in the reporting they've established.” To help publishers deal with the numbers discrepancy, BPA established the “Quiet Zone,” which allows BPA and the member company to see and analyze Web numbers before releasing them to the public. “At month's end, we can talk about why a particular day's numbers are significantly different from the numbers that are coming in from other sources,” Hansen said. And over a period of time, the publisher and BPA can become more comfortable with the numbers and why they differ. The other possibility, Hansen said, is that the numbers will catch up to whatever the company is currently reporting during the time it is in the Quiet Zone. Napoleone said he does not have a problem entering Bobit's sites into BPA's Quiet Zone, “but we will not report them before we can get a concrete answer on what is causing the discrepancies,” he added. He added it is very difficult to beat Google's “user-friendly interface and the ability to drill down for more details on just about every metric on the screen.” Hansen envisions an increasing number of magazines choosing to participate. As they do, he said, competitive publications will feel the need to be involved as well in order to have similar numbers to show media buyers. “The more that are involved, the more value it has for media buyers who are trying to invest in our businesses,” Hansen said. “The way I see it, you're already paying for this by being a member of BPA. Why not get as much information as you can on your publication?” M
Most Popular
In this article: