Nielsen to rank Web sites by visit lengths

Nielsen/NetRatings, a leading online measurement service, plans to stop ranking sites by number of page views and begin tracking how long visitors spend at the sites. Although Nielsen already measures average time spent and average number of sessions per visitor for each site, it will start reporting total time spent and sessions for all visitors to give advertisers, investors and analysts a broader picture of what sites are most popular.

The move makes sense if you consider the time users spend watching video online and sites with technologies such as Ajax that update data continually, making page views less meaningful. "Based on everything that's going on with the influx of Ajax and streaming, we feel total minutes is the best gauge for site traffic. We're changing our stance on how the data should be" used, said Scott Ross, director of product marketing at Nielsen. Nielsen will still provide page view figures but won't formally rank them. Ross said page view remains a valid gauge of a site's ad inventory, but time spent is better for capturing the level of engagement users have with a site.

News source: MSNBC

Report a problem with article
Previous Story

PowerArchiver 2007 10.2 Beta 1

Next Story

Microsoft's next move is the Xbox 360 Code-name Falcon

18 Comments

Commenting is disabled on this article.

Just wondering, wouldn't this lower ratings of Search Engines for example Google as people on search, click then go? Not sure.

Lets just hope this doesn't gimp shape web content like the bloody Nielsen does for TV content, with "SWEEPS" and all that crap. Maybe they'll find out what the peek hour of surfing is and people will ONLY post the most interesting stuff then. Only bad can come of this.

surely if they did it by daily visitors * average time spent, that would give the best way to list them, taking into account all the problems mentioned?

The biggest DUH bit on this is that people will intentionally get people to leave their web browsers sitting open all day, something you wouldn't really do with a tv so much as with a webpage.

Exactly, I was going to post the same thing. I leave webpages and certain forum topics on all day for Neowin because I don't want to search for it later on.

From my understanding, as I had to sit in on the teleconference and web presentation they had last week announcing this because the company I work for does banner ads among other things, is they will somehow only track the web page and/or tab that is in focus, and when the browser is the active app. Not sure how they can even do this, but this was how it was explained to me. So even if you have the browser open with multiple windows, it will only be counting the window "in-focus," meaning only the one you are actually viewing. If you switch to lets say Word for sake of the example, it will no longer be counting any activity at all.

I would imagine a page view would stop counting after a certain length of time, maybe 30 minutes or an hour - you're not gonna have someone that leaves their browser open for 14 days straight actually counting towards the statistics. Sure there will be applications out there used to emulate browsing but then there are applications out there to generate multiple hits on a page as well. There are too many variables to ever truly rank websites so whatever method they pick is going to be imperfect - however, they can try their best to make the ranking relevant and I believe that this is an improvement (it seems like common sense to me).

theyarecomingforyou sounds reasonable. I still could leave a webpage on the focused tab so after a certain amount of time they should stop counting if anything.

Gamerhomie said,
theyarecomingforyou sounds reasonable. I still could leave a webpage on the focused tab so after a certain amount of time they should stop counting if anything.

What about small unattended visits? I tend to background-open multiple links and read them sequentially. So the farthest tab stays open far longer than the time I actually spend on it, but less than the 30 minutes to trigger the timer to turn off. That will give a skewed result.

DirtyLarry said,
...is they will somehow only track the web page and/or tab that is in focus, and when the browser is the active app.

Are there built in methods in browsers to detemine whether or not a page is in focus?

Time spent on a page? That will show how much "viewing time" each ad is getting (as on most sites the ad changes with each page change)

I don't think minutes viewing a page along is a good gauge. I don't think page views alone is either, there should be a middle ground, a combination of both. Maybe weighting the two, or consider the average length of time a hit stays on the site, or maybe use a weight that too. Possibly a combination of minutes viewing a page, hits, and average time spent per hit. That would be a better gauge of a webpage's popularity, instead of just blanket ranking everything based on one category. Wouldn't be that hard to do either.

Pages hits are pretty irrelevant - the web has moved on from static content and advertisers don't really care about if you visit a site but whether you pay attention to a page. Also, page redirects are used to increase hits artificially, hence why you'll often have to click through an ad page or navigate multiple pages to get where you want - those pages have little to no value so there is no reason they should be used to gauge if a site is popular or not. View time is currently the most effective way to judge a site's popularity and it is good to see Nielsen adopting it to rank websites.

Then damn, porn websites win out, by far.
It takes 5 minutes to read a news article w/ it's comments.
It takes 15 minutes to have a good stare at a girl.