On 15/01/10 I stumbled my article “My First Recurring Income Online …”. I also delicious bookmarked it, shortened it with su.pr and posted it to twitter. Every day, for a whole month, I stumbled and rated websites around the internet for at least 20 minutes per day. Since that month I’ve been an active user of StumbleUpon and the results of this experiment have blown me away. This is a review of what happened.
At it’s most basic element, my experiment was to see what effect – if any – my consistent participation in the service StumbleUpon has on how well my articles go on StumbleUpon when I stumble them myself.
When I started this experiment I was actually quite pessimistic about how successful it was going to be. But like most things, it just kind of came to me one day and I decided to roll with it.
I’m not usually in the habit of stumbling my own articles. I read somewhere that stumbling your own articles too often is bad and actually inhibits your effectiveness in the StumbleUpon system, eventually leading to no traffic at all if you do it all the time.
Now that I have some perspective, I think it has something to do with the ratio of how many links your favourites have to one particular domain as to what your vote weight is for that site (or any site you favourite perhaps).
I’ve actually been a member of StumbleUpon for quite a while. I have buttons on my blog, but I rarely promoted my own stuff, or even used the service as it was intended. After about a month or so of use back when I signed up, my account was rarely used.
Nowadays however, I’ve become a bit more of an active user of the service (here’s my StumbleUpon ID if you’re interested in connecting).
Traffic Activity (Before Actively Stumbling)
As far as traffic goes, my articles don’t really get stumbled all that often by my blog’s visitors so the ticker in Google Analytics hardly moves at all for traffic from the StumbleUpon domain. Here is my traffic stats for visits from StumbleUpon for Dec/Jan:
It had been a while since I’d stumbled anything from my own site so I guess I was starting from a pretty blank state. If I had my chance again in this experiment I would have created a whole new StumbleUpon account to get a more accurate representation.
Because I was starting from such a blank slate, I thought it would be interesting to stumble one of my blog posts as an experiment in how much traffic it brings in over the next couple of days.
My idea was to then participate in the stumbling process for a predetermined period of time per day, stumbling and rating articles for my enjoyment for around 20 minutes. At some point in the future I would then submit another article or two of my own to StumbleUpon and see what the difference was.
Needless to say (by the picture above) submitting that first article made absolutely no difference to my stats. There was no rush of visitors, not even a small one. Confidence was high!
How Long Did I Actually Stumble For?
I stumbled on average for 20-30 minutes per day for that first month (pretty much all of February).
After that I got a little lazy, but my stumbles were now more about discovering new content (stuff that hasn’t been stumbled or reviewed yet) or rating things that I knew had already been stumbled (but not actively using the stumbling interface to bounce around the web).
For new discoveries I would write a small 20-50 word review and be on my way. I’ve done this sporadically over Feb and March.
All in all since the first stumble on the 15th of Jan, 2010 I’ve been actively using StumbleUpon for around 2 and a half months (today being the 02 April).
Latest Stumbles Of My Own Content
Over the last few weeks I thought it would be a good idea to start testing stumbles on my own content again to see if there were any results.
I’ve made the following Stumbles of my own content recently:
I’m not sure how accurate the numbers are here, but the last two were the most surprising because in Google Analytics they created two very large spikes (large for my site, anyway).
I’m sure we can all agree the spike in traffic is interesting, but what I find most interesting is that the quality of StumbleUpon traffic compared to what most people assume it to be: rubbish.
Unfortunately, I think the generalisations are actually pretty close to the mark. It’s not the kind of traffic that lingers around, or even clicks around.
In addition I didn’t see any similar spikes to my feed count or any other statistics that I’m monitoring, it was pretty much just the pages that were successful that received the traffic injection and that was that.
I’m going to monitor how much traffic is sent to me via StumbleUpon over the next few weeks, but in general it’s been increasing ever so slowly over the last two months. Not including the spikes, the traffic has roughly doubled, although that could just be from the added articles. We’ll see how that one goes in a follow up article maybe.
So what do you think of StumbleUpon traffic? For a lot of people it’s one of their biggest traffic sources, but then what are your rates of adoption like? Do stumblers people subscribe for example? Do they buy your stuff? Let us know in the comments below.
I’d love for this experiment to put some factual data against some of the preconceptions people have about StumbleUpon and the quality of it’s traffic. Personally, I’ll be playing a much more active role in StumbleUpon. I think that for a small site StumbleUpon remains one of the best ways to grow your traffic streams.
If you liked this article, maybe you could stumble it? 😉