AOL gains, Google loses in Nielsen metrics change

A new ranking methodology at Nielsen/NetRatings gives AOL a boost while disadvantaging Google.

Nielsen/NetRatings this week made a change to its metrics for ranking the most popular Web sites. It will now focus on the amount of time people spend on a site instead of just the unique audience. This move benefits sites with services like instant messaging and e-mail over sites that offer quicker activities such as search.

Using the new total-minutes calculation, AOL is ranked as the most popular Web site in May in the U.S., followed by Yahoo; MSN/Windows Live; Fox Interactive, which owns MySpace.com; Google; eBay; Microsoft; Electronic Arts; Apple and YouTube, which is owned by Google.

Under the unique-audience calculation, the top 10 list looks like this: Google Yahoo, MSN/Windows Live, Microsoft, AOL, Fox, eBay, YouTube, Wikipedia and Apple.

Fox's MySpace, Google's YouTube and eBay have loyal and engaged users, while Microsoft's Windows Media Player and Apple's iTunes helped those rankings, Nielsen/NetRatings said. Electronic Arts broke into the top 10 based on total minutes because of the popularity of its Pogo.com gaming site.

Web sites are increasingly offering services that keep a user on a site for longer, and measurement firms are looking at user "engagement" metrics more and more. The widespread adoption of streaming audio and video, which provides dynamically changing content within a single page or media player, and applications like Ajax, have diminished the importance of measuring page views.

On Thursday, I'll take a broader look at what industry insiders think about this change and what impact it may have on advertising.

[NWLB]

Oh sure, but....

...does their "method" account for a larger percentage of AOL users accessing the sites with Dial-up services. Are they starting to count from the point the site is accessed, or when it finishes loading? Or what about when the next site visited finishes loading? Give or take the technical side of the method, it would seem that a service reliant on slow dial-up speeds would possibly give the illusion of spending more time on a site.

And if people access that site within AOL's pathetic software, as opposed to a pure browser, is AOL doing anything that might effect those stats? Those "high speed" boosting gimicks for Dial-up could effect people reloading pages, or accessing them as well.

Not knocking anything, just curious how their method really works.

[jshindl]

Also

Here's what I think part of this is:

Aol's site attracts less savvy users and is harder to navigate. Google's site attracts more tech savvy people and is easier to use, thus is takes less time to get what you are looking for. Under these numbers, Google gets penalized for the users it attracts and for the power of its Web site.

[jfmezei]

Time spent on a web site ?

Is Nielsen clueless on the nature of HTTP transactions ? How can then reliably calculate the amount of time someone spends on a page/web site ?

[Thomas_74]

There is more to metrics

there is much more to metrics today then just time spent.. with multi-tabbed browsing, i might open a page and leave it for the whole day.. or a user might just open a page in iPhone or something and never visit it again for a while.. doesn't mean that use actually spent that much time on one page..

And if you are watching video, how does this metric stand ? i will watch the video for its length.. and move on, does not mean i spend 10 mins on the page.. it just means that the video was 10mins long.. duh Neilsen.. you can do better...