StumbleUpon LogoLast week, Jiannis wrote a great post called “Why You Shouldn’t Stumble Yourself” in which he brought up the seemingly little-known fact that the more you stumble pages within the same URL, the less effective your stumble becomes. The post stirred up a lot of great discussion and got me thinking a lot more about the Stumble algorithm and the important steps that need to be taken when using Stumble for social media marketing purposes.

A major point of the post was there is a mysterious ratio buried deep within the innards of the SU algorithm that compares your total number of stumbles to the number of stumbles you have given to any specific URL. This was brought to my attention by a comment on the post mentioned above. The comment is from Kimota, and I quote him here with many thanks.

I recently saw my stumble traffic plummet and my submissions from my own domain no longer get accepted. SU didn’t even respond to my enquiries as to why this would be. It wasn’t until days of research had gone by that I discovered a buried little paragraph deep down on one of the SU pages that mentioned the ratio of one domain to other stumbles being a factor and that if this ratio is tipped, SU prevents you from submitting any more from that domain and affects your entire traffic. No warning and no possible way of readjusting the ratio to repair the damage.

The most immediate application of this knowledge is that you have to be very careful when submitting your own content to StumbleUpon. If you submit all of your own stuff over and over again, it is going to get less effective. Kimota knows this first hand.

Of particular–even alarming–interest, though, are the repercussions of “tipping” this magic ratio. To better discuss this question, lets say there are two elements: the Offender (the user) and the URL (the URL that has been submitted too much). Is it simply that StumbeUpon stops counting URL submissions from the Offender, or is the URL blacklisted? In the same comment, Kimota goes on to say, “Even when someone else stumbles me, I’m not getting anywhere near the same figures.”

Could it be that the URL is punished for the sins of one overzealous SU user? Can your frequent self-submission ruin the SU traffic potential of your URL even if you never submit yourself again?

Of course, this is the kind of social media sensationalism that gets rumors started and some of you might think this is tantamount to tabloid journalism. Let me be clear right now, I do not know the answers to the questions. Indeed, when it comes to the mysterious web algorithms out there, we are all making our best guesses based upon research.

The ultimate takeaway here is to be overly careful when it comes to helping your fledgling blog through its baby steps. But I am very interested by the idea that a URL as a whole could be penalized for being submitted by the same users all the time. I can’t imagine that would be the case, because that would actually penalize a blogger for having a rabid fan base. I greatly welcome any input and discussion from those of you who can provide insight (anecdotal or, even better, factual). Have you noticed trends? Take a look at your own stats and share them with the community in the comments. As we get some good stuff together, I will round it all up into a follow up post. Let the learning begin.

Written by Matt Peters
Matt is the Co-Founder and CEO of Pandemic Labs, and enjoys thinking about, writing about, and talking about social media marketing whenever someone will let him.