Pwned: The StumbleUpon Digg Experiment Initial Results

First, a disclaimer about this experiment and the analysis.

In retrospect, this experiment was flawed.  The subject matter and style of delivery was very clearly geared in favor of one of the combatants.  When it was initially conceived, it was decided that the experiment would best be delivered through a post that announced itself.  By checking traffic statistics on a post titled: “The StumbleUpon Digg Experiment”, there would be equal billing, equal exposure, and most importantly, equal chances through the delivery methods to give both sides a chance.

We were wrong.

In retrospect, it is clear that the title and subject were more geared towards a social experience.  While Digg offers ways to share stories with friends, it isn’t nearly as suited for this as StumbleUpon is.  The very nature of SU is designed to where members passively share stories they like by stumbling and reviewing them.  Digg requires active participation by people to see the story.  Stumble requires nothing other than clicking and waiting for websites to be served to them.

$8.95 Domain Names Transfers from Dotster
The Digg traffic of lore says that websites will get tens of thousands of visits in a short period of time.  The problem is, not all stories get this kind of traffic when they hit the front page.  Because they have to be actively clicked on, people have to have an interest to click.  This story, while more popular than past stories that have hit the front page of Digg from this blog, still didn’t have the universal appeal to get a ton of traffic.

Stumblers are, by their nature, more interested in stories about SU than Digg users are interested in stories about Digg.  This and other factors make this experiment somewhat flawed.  Thus, the results were tremendously in favor of StumbleUpon.

Detailed statistics are being compiled.  We are breaking it down hour by hour, noting traffic spikes and comparing them to specific times of reviews on SU.  The data so far is interesting, but for those interested in the general results, here they are:

Story posted early in the morning, PST, December 14th.  It was Dugg shortly after being posted by ThinkingSerious. Then, it was Stumbled by hockeyguru around 7:00 am PST.

December 14th, 2007:

  • Traffic from StumbleUpon: 15,694
  • Traffic from Digg: 8,463

Surprised?  So was I.  Traditional thought is that Digg wins the short race and SU catches up over time.  Again, the subject matter/headline were geared to get Diggs, but not actual visits.  Another thing to note as that this story made it to the “Top in All Categories” box on the front page.  Less than a minute later, it was buried off that page after getting 24 hits.  Stories that do can get a huge bump in traffic for that day.  Some stories whose statistics I have seen have gotten twice as much traffic or more from being there than they did when they were initially on the front page.

December 15th, 2007

  • Traffic from StumbleUpon: 4,761
  • Traffic from Digg: 2,357

This is actually a very encouraging number from Digg.  Considering most of the traffic came from people when the story had to be found on deep inner pages, this is a huge number relative to the original day’s response.  Stumble traffic, if anything, was a little disappointing, as some stories get 80% of their initial day traffic on day two.  Despite the positive reviews still rolling in on day two, the percentage compared to the first day was low.

December 16th, 2007

  • Traffic from StumbleUpon: 3,689
  • Traffic from Digg: 349

There is the drop that happens with Digg.  There is also the famed residual traffic from Stumble.

Again, let me be clear.  This story was unintentionally geared to for SU.  The results are not a real representation, as stories that aren’t about Social Media will still have the normal results of Digg wins the first day and SU may or may not catch up over time.  The subject matter appealed to SU’s manner of delivery.  One digg is one digg and helps very little.  One stumble and especially a positive review by the right person can generate a chain reaction, a viral effect that can snowball into more and more stumbles.

Thank you to all who participated by reading, sharing, stumbling, and digging this story.  We want more data.  If you have statistics from posts other than social media themed ones, please contact us.  We want to explore, analyze, and share the data.

* * *

Read more social news stories on this blog.

Leave a Reply


  1. Very interesting results indeed. Its almost impossible in such a case to isolate all variables, but I still find the results really useful. Thanks!

  2. Interesting results, but i guess they’re not that far off to begin with.
    My personal dealings with SU and Digg on my blog have produced the same results. And only one story I ever wrote got to the front page of Digg, granted, for only about 5 mins or so. But many more of my stories get popular on SU then on Digg.

    Anyway, this is a nice experiment, but i think it should be conducted again, for “scientific” purposes.

  3. The Big difference between Stumble and Digg traffic wise is that if your article is truly interesting, StumbleUpon traffic can take weeks or months to die off. I’m not sure the exact algorithm, but as people continue to stumble, the article keeps getting shown to people.

    I’ve had pages where the stumble traffic died off almost entirely in about a week and others where 200,000-500,000 views later is still going strong. I think Stumble much more than digg really lives or dies on your content depending on its quality and ability to illicit additional thumbs up in very quick viewings…

  4. JASON

    I quote.. ” . One digg is one digg and helps very little. One stumble and especially a positive review by the right person can generate a chain reaction, a viral effect that can snowball into more and more stumbles. ”
    I was wondering if you could tell me, how does one become “the right person” on SU where he’s/she’s positive comment does wonders for the article eh thumbed-up??
    (hope that made sense)
    Thanks, Jay

  5. Koliedrus

    I’d like to suggest a repeat of the experiment but with the content page that’s to be used to gather statistical data being obscured. Perhaps publishing on a website that doesn’t immediately inform the visitor that they’re participating in an experiment simply because of its name.

    Content, of course, will matter once the visitor reaches the page.

    I expect that I’ll learn about “Round 2” when I learn about Round 2.

  6. Yep, your findings, scientific or not are right on the money. I think Stumblers are more prone to sign up as subscribers to your feed as well. Diggers are not as reliable in terms of traffic. Hope you don’t mind, I reprinted this article with your link back to the original. Thanks. Really great info.

  7. Jay

    good experiment… SU has helped a lot with my traffic, while digg has helped a little… so this is some good backup…


  8. Great to see stumble come out tops as I knew it would! You will see a second wave and a third wave of traffic repeat from stumble in a week or 2 weeks.

  9. Linked to this article from here:

    Why StumbleUpon “Owns” Digg

  10. Chris’s comment about the longevity of SU is dead on.. I’ve had SU stuff go on for years and years..

    Digg is just a big flash and then nothing.. frankly, I stopped caring about Digg a long time ago – I don’t event WANT to be Dugg – it just tied up my site with idiotic comments and valueless traffic.

    I think Digg is going to slide into well deserved oblivion soon enough..

  11. I love SU, DIGG is no fun and my traffic from SU is amazing.

  12. Ok, a rush of traffic comes either from SU or Digg wow! But does that traffic have any value other than my stats show X amount of visitors. I am a popular blogger…

    More info would be needed for this type of experiment. Did visitors read other articles or leave after visitng one post? Did they clcik on ads? Did they subscribe to a feed or newsletter?

  13. Cool post

  14. StumbleGenius Stumbling on thru, StumbleUpon thumbs up to you, hope to see you in more clicky stumble trips soon. Happy day merry seasons greetings an all that jazz.

    Read you later or tweet me on Twitter @StumbleGenius … tweet you soon ???

  15. It is really difficult to analyze the different variables though especially when it pertains to psychology and human experience, it is a highly subjective science. However, I think that the experiment results are nothing to be scoffed at, we all learn from experiments whether they succeed or fail, a fact of science endeavors 🙂