Nielsen/Netratings has just started ranking Web brands. It does this by measuring “engagement” by total minutes spent with the brand and unique audience. Brands that had a large unique audience spending a long time engaged had the highest score. AOL came in first, followed by Yahoo!, then MSN/Windows Live, with Google coming in fifth.
If that sounds backwards to you, it’s because it is. In fact, it’s basically the exact opposite of what we tell our clients. Anyone in online advertising knows that Google has the largest, most valuable audience, period. But aside from the obvious, there are three additional reasons why this new ranking system is potentially dangerous. First, the measurements aren’t necessarily going to be accurate, second, even if they were accurate, it won’t reflect the value, and third, it might empower high-ranking publisher sites to up their rates without adding value.
Let’s start with reasons why the measurements won’t be accurate. First, there’s the ongoing problem of cookie deletion. According to Nielsen/Netratings, AOL had an audience of 91.6 million unique visitors in May 2007. According to a cookie deletion study by comScore, 31% of users delete their cookies once per month. If a user visits AOL, deletes AOL’s cookie, and then arrives at the site again, AOL erroneously counts that second visit as a unique user. Simple mathematics tells us that AOL’s numbers are inflated by 28 million visitors. True, all brands ranked should be affected equally by cookie deletion, but the true value to advertisers is what suffers here.
Additionally, is time where the user minimizes the browser window counted as time spent “engaged with the brand?” What about time where the user opens a new tab in a browser like Mozilla Firefox, or opens a new window and maximizes it? Or what if I leave a page open and get up to go to the bathroom? Or I could leave my office for a three hour meeting. You get the idea. If Nielsen’s clock is ticking during situations like these, I’m not sure how that time adds value to any brand. What if I have two windows open but minimized, like AOL IM and AOL News. Does that count as twice the time?
But even if the measurements are accurate, it’s hard to see how some of that is relevant to the power of the brand. Suppose I’m on AOL IM for ten hours a day. I have a hard time seeing how that helps me feel connected to the brand beyond the first three minutes of that session. The same goes for the iTunes store and Apple, or Windows Media Player and Microsoft.
Similarly, I find it difficult to believe that there’s any additional value created for advertisers by increasing the length of the impression. Perhaps with a video ad, you’d want to make sure that the viewer sat through the whole thing, but with a static banner ad, or any type of rich media outside of video, a long impression does not equal a more valuable impression. My AOL instant messenger buddy list window may have had a rich media ad for T-Mobile going all day, but, if I notice it at all, I’m unlikely to pay attention to it more than once during the period in which it is open.
This begins to get worrisome because I meet with publishers all the time who charge arbitrary rates. When asked what the value is, they respond, “that’s just our rate card.” The big concern with this is that the publisher sites that rank well will be able to say, “according to Nielsen/Netratings, we’re one of the top one hundred brands on the web, so we’re charging a $20 CPM.”
Personally, I never want to see another rate card again, and the next time I hear a sales guy tell me if I spend more, I will get a discounted rate, he had better be prepared to show me exactly what I’m getting for my money. Time spent on sites and unique users do not guarantee us anything.
