Pageviews vs Visitors: Do people really just sit there and reload all day?

April 28, 2012 2:31 PM

On the Lifehacker/Gawker websites, what do the views/viewer numbers really mean? How can articles with 100,000 pageviews only have 10,000 visitors?

I wrote a popular post on Lifehacker recently and am trying to discuss its significance for a book proposal, and could use a hand figuring out what these numbers mean.

On this post (not mine), for example, it shows around 33,000 pageviews and 3700 visitors. Does that imply that each visitor reloaded the page an average of 9 times? This article is not really a collection of links that a viewer would return to multiple times, so what's going on here?

posted by sdis to Computers & Internet (9 answers total)

Could the number be inflated by people coming back to check the comments thread? Or, posting a comment, and then having the page reload afterward?
posted by synecdoche at 3:31 PM on April 28, 2012

Does it differentiate between bots (search engine crawlers, etc.) and whatever "visitor" means? That's the first question that came to mind.
posted by circular at 3:35 PM on April 28, 2012

I have no idea... They just changed their commenting system, for the worse imho, and their view count technique could be messed up too. Can't Gawker tell you? And then you can tell us.
posted by caclwmr4 at 5:21 PM on April 28, 2012

I have wondered myself how sites account for people who have lots of tabs open in their browser, that are restored after the browser closes and restarts, such as for security updates, installing add-ons, machine reboots, etc. I have tabs open right now. That may not be it, though.
posted by forthright at 5:57 PM on April 28, 2012

How many times a page is viewed. This includes reloading a page, going to another link and then going back to that page (think how many times you click on back - you're not necessarily reading anything, you're just there).

The comments would cause people to reload a lot and Gawker has high click-around-ability (totally a real word) with its navigation/general layout.
posted by mleigh at 7:30 PM on April 28, 2012

Reload = refresh.
posted by mleigh at 7:31 PM on April 28, 2012

Response by poster: I've sent Gawker an email asking about it. Comment threads are on the same page, so that shouldn't cause new pageviews (that, and most people aren't allowed to comment anyways, so that's going to reduce the number of people actually going down to comments).

I could see people going to a linked website and then clicking the back button, but that's only going to double the viewer count...I'm a bit at a loss as to what would make it go up by a factor of 9-10x.
posted by sdis at 2:30 AM on April 29, 2012

Best answer: Ah! Finally a question I'm equipped to answer.

That is because Gawker Media properties use the metric of unique new visitors, not unique visitors. So the two numbers you have are pageviews and "first-time" visitors (which, given the state of cookies, probably means anything from first-time to "people who haven't been there in a month). This is dumb to publicly display, and unique to them. But it is Gawker Media's obsession, on the principle that first-time visitors are the sign of growth, not return visitors, which is actually correct.
posted by RJ Reynolds at 6:53 AM on April 29, 2012

Response by poster: So if an article without many links has 100,000 pageviews and 10,000 unique visitors, then it's probably been seen by 50,000-90,000 people, 10,000 of which came to the Gawker site for the "first time" to read that article?
posted by sdis at 2:24 PM on April 29, 2012