Breaking Down The StumbleUpon Algorithm

Filed as Features on September 20, 2007 4:32 am

StumbleUpon is a tool that allows its users to discover web sites, videos, and pictures that have been previously found and labeled by other users. Its a lot of fun, and perhaps just as important, a great source of traffic for your blog.

But other than doing the obvious, such as becoming a member of StumbleUpon, finding “friends”, and otherwise rating sites and being a good Stumbler, what else can be done — and understood — about StumbleUpon to maximize your efforts?

Turns out one Stumbler, Tim Nash, has tried to break down the mechanics of the StumbleUpon algorithm based on his own observations. While it has yet to be fully vetted by any of the StumbleUpon community it does provide some interesting variables to chew on.

Check his thoughts out over here, but Tim basically tries to break things down into understanding an Audience Score, where he attempts to break down some variables that lead to a “power” stumbler; several models that might explain how any given site might get popular through StumbleUpon (taking into account, for example, the first stumbler); and finally some bonus factors that might influence things in ways that deserve a lot more working out.

Just like we’ll never know the Google algorithm, we’ll probably also never know the one for Digg, or *any* of these algorithm driven sites. With time, we’ll probably derive our own ideas of how they work, and Tim’s ideas are probably just the beginning with StumbleUpon.

Having said that, there are some experienced users who feel that breaking down the algorithms to boost your rank doesn’t really make a lot of sense after all. You can read a lengthy review of that over at DoshDosh.

Tags:

This post was written by

You can visit the for a short bio, more posts, and other information about the author.


Submissions & Subscriptions

Submit the post to Reddit, StumbleUpon, Digg or Del.icio.us.

Did you like it? Then subscribe to our RSS feed!



  1. By Tim Nash posted on September 20, 2007 at 6:14 am
    Want an avatar? Get a gravatar! • You can link to this comment

    Thanks for the write up, I wrote the article in a hope to not only inspire some discussion on the algorithm its self but also to demonstrate just how many factors are involved. Even my “big model” doesn’t fit exactly how stumbles occur but I think it at least begins to give people some idea of what “appears” to be happening

    Reply

  2. By Tony Hung posted on September 20, 2007 at 6:30 am
    Want an avatar? Get a gravatar! • You can link to this comment

    Hi Tim,

    No worries — I don’t think we’ll ever know for sure, but I think its always interesting to have the discussion, and posts like yours are always good launching points for that.

    Reply

  3. By Tim Nash posted on September 20, 2007 at 10:50 am
    Want an avatar? Get a gravatar! • You can link to this comment

    Not wishing to be picky but any chance you can correct my name, I just got a pingback from another post which have decided I’m called steve :D

    Check his thoughts out over here, but Steve basically tries to break things down into understanding an Audience Score,

    Fussy I know.

    Reply

  4. By Tony Hung posted on September 20, 2007 at 12:50 pm
    Want an avatar? Get a gravatar! • You can link to this comment

    Done — with a name like “Nash”, its easily to confuse your appellation, particularly to a Canadian basketball fan. :)

    Reply

  5. By Tibi Puiu posted on September 21, 2007 at 10:19 am
    Want an avatar? Get a gravatar! • You can link to this comment

    Interesting points,with a lot of things covered very well, not very sure about his view on the algorithm though.

    Reply

  6. StumbleUpon How it might work | TechWagSeptember 21, 2007 at 3:53 pm
  7. What’s Your StumbleUpon Rank? Introducing StumbleRank : The Blog HeraldNovember 4, 2007 at 4:28 pm
  8. http://www.blogherald.com/2007/09/20/breaking-down-the-stumbleupon-algorithm/March 21, 2008 at 12:22 am

    Your words are your own, so be nice and helpful if you can. If this is the first time you're posting a comment, it might go into moderation. Don't worry, it's not lost, so there's no need to repost it! We accept clean XHTML in comments, but don't overdo it please.

    Current day month ye@r *