The Long Tail Internet Myth: We are Spending MORE Time on Top 10 Sites than Ever

Image from: nikkytok / Shutterstock

In 2006, Compete was contacted by Read Write Web to provide data for a piece on the Long Tail Theory: The Shrinking Long Tail – Top 10 Web Domains Increasing in Reach (You can see our follow up to the post here). The Long tail theory is another way of describing the democratization of media and business through the Internet by removing some of the institutionalized hegemonic powers of the biggest mainstream outlets. In this post we compared the share of page views held by the Top 10 domains in 2001 to that of the Top 10 in 2006. We found that the top 10 domains weren’t shrinking at all, as alluded to by the long tail theory.

Recently, we were contacted by B.J. Mendelson who is writing a new book entitled “Social Media is Bullshit” to be released in the Fall of 2012. He reached out to us inquiring for an update on the 2006 blog post, and being that it had been five years since we touched the topic, we were due for a 2011 update to determine whether or not our data-driven thoughts about the long tail theory remained true.

In terms of the page views metric, they certainly haven’t.  In fact, the upward trend that alluded to a increasing concentration of power in the hands of the top domains has seemed to stabilize.

Since 2006, the total number of Internet domains has increased by 11% and stretched out the power of the long tail in the lens of page view analysis. The number one domain in September of 2011 (Facebook) resembles the number on domain of 2001 (Yahoo!) in terms of page view dominance. It seems that 2006 was the peak of the power gap in that regard.

In 2006, our analysis stopped here with the assumption that page views were an adequate enough metric to test the long tail theory. Five years later, no one would dispute that the Internet has changed considerably. With the introduction of new coding, like html5, javascript libraries, and flash, pages are becoming less representative of an Internet user’s entire experience.  For example, take the action of viewing pictures on Facebook. In the past, you would have to go to a new page to view photos, which equates to thousands of page views a month from a single person. Now, you can flip through an entire album without changing pages.

Because of this, we wanted to look at Internet usage on the Top 10 domains in terms of our attention metric as well. Attention considers all the time we collectively spend online and then determines what percentage of that time was spent on a given site.

Here we see a different story. We see further support of our 2006 research which disproved the long tail theory.

Here is the breakdown of these results by site in 2001, 2006, and 2011 for the month of September:

As you see here, the top 10 sites are garnering more and more of a user’s entire time spent online. Although the change isn’t as rapid as the growth from 2001 to 2006, we are seeing evidence to disprove the long tail theory with regard to engagement, which we think might currently be a more accurate portrayal of today’s internet population in the context of the long tail theory over page views.

With that said, there are exceptions (Search Engines for example), which is why we don’t present Attention as the king of all metrics. As with any standalone metric, we see Attention as an additional piece of the puzzle. It is important to view multiple metrics in conjunction with each other in order to understand each dimension of internet usage and build a complete perspective.

About Jen Duguay:
Jen Duguay joins Compete to take on all things social media. She comes from a social issue background, most recently having worked for the Social Innovation Forum, the venture philanthropy arm of Root Cause, a nonprofit research and consulting firm. Jen's interests include singing, marketing, running, art, making guacemole, and using social entrepreneurship to tackle world issues. She has spent time in Belize and the Dominican Republic working on microfinance initiatives and recently traveled to Kenya where she studied the public healthcare system. Follow Jen @jenduguay on Twitter.

Categories: Social

Tags: | | | |

Bookmark the permalink.


Your email address will not be published. Required fields are marked *

  1. Luca Zanzi

    Other data intersections that might be interesting to throw in:
    1) How many of these websites are accessed through referrals.
    2) How much traffic is generated from referrals through these sites.
    3) How many are set as browser home pages.

    Fascinating that Bing is still a few steps below Yahoo, despite Yahoo using Bing’s algorithm. Shows the weight of a household name over the years.


    • Jen Duguay

      Luca, thanks for the input! We are looking to release a whitepaper in the near future that relates to some of your commentary. Keep an eye out!


  2. Jaan Kanellis

    Don’t you think these numbers for the top ten sites represent the “type” of sites that are in the top ten, rather than they are top ten in traffic?


  3. Jen Duguay

    Hi Jaan, thanks for your comment! The Top 10 sites are ranked by each particular metric. For instance the first group of Top 10 Sites is based off of number of page views and the second group of Top 10 Sites is according to our attention metric. If you’ll note, the Top 10 Sites in 2011 for page views are different than the 2011 Top 10 Sites with regard to attention. Sorry for the confusion!


    • Luca Zanzi

      Yes, the views vs. attention is just fascinating. No numbers at hand (I’m sure Kantar Media has them), but look at Facebook. People check it, yes, but it’s more important that people go on it and stay a while .
      Throwing a wild guess here: how many Facebook users leave a webpage open just so they can use their chat? Granted there are 3rd party aggregators (Trillian, Digsby), but that’s a small fraction of its 800 million user base.
      Guess it makes sense that Facebook has yet to release their own chat client: loss of active visits.


  4. Pingback: Marketing Day: December 5, 2011

  5. Chris Dowell

    The top 10 sites are gaining I think because people only have so hours a day. People have decided that instead of viewing 20-30 websites, I’ll view their top 10. Just happens to be that many people are choosing similar sites.