Given the experimental nature of the sites in question, I was intrigued this morning when I was doing my regular rounds of the Internet and came across this:
What happened? Did the little blog, whose mantra is thus, suddenly become twice as interesting overnight? Since its content is updated infrequently - once weekly at best - I doubted it. I had a fairly good idea as to the source of this extra traffic, and I was right. StumbleUpon, my fair weather friend, had returned.
This is not a huge increase in traffic by any means - when a really big StumbleUpon day happens, the chart looks a lot different. However, the interesting thing is not the fact that StumbleUpon happened, it's that it keeps coming back to the same page and yet that page's reviews never change.
While Google Images has a long-term love affair going with the site in question, StumbleUpon visits about once every few months, and always to the same URL. Funnily enough, Google Images prefers the pictures on this particular page as well. The really strange thing about the on-again-off-again relationship that StumbleUpon has with this piece of content is that the reviews don't change and, as far as I can tell, few people are particularly impressed with the page. According to its reviews, only three people like the site and it has only been written about twice. And yet, every once in a while, StumbleUpon is back outside, throwing stones at the window and asking us out.
I know that the StumbleUpon reviews pages are supposedly far from accurate: the service has always been a bit cute with its numbers and I've never seen a review page that has accurately portrayed how much traffic the site in question will receive. However, it amuses and surprises me that StumbleUpon does not work like most other social media sites, which will not "re-promote" content after it has been popular. So far, StumbleUpon has revisited this page four times.
It's great for us - writers, publishers, linkbaiters, and SEOs. The do-or-die aspect of Digg is negated by the notion that not only is StumbleUpon popularity and traffic not a ticking clock, but that even once the first wave of SU traffic dies away, it may return again. I am not entirely sure what sets off the subsequent influxes of stumbles (I assume it is a tag or a thumbs-up from a prominent account), but there is one theme in common with each wave of Stumble traffic: every time it comes back, it brings with it more visitors.
However, my analysis of StumbleUpon traffic has been really interesting for more than just the fact that it can't leave my one page alone. A common idea I've heard from the Internet marketing community - and one that I used to spread - is that StumbleUpon is the everyman's social media site. "Regular" people - people who don't get FAIL or Rick Rolls and who squint at me when I go out in this tee-shirt - were thought to use StumbleUpon, at least in higher volumes than they do any other social media mainstay.
I can tell you from my little blog's stat counter that this really isn't as true we we'd like to believe. This morning, I went through the stats and analysed my StumbleUpon audience in comparison to my regular visitors.
Firstly, the graph of browsers is telling. I check this graph relatively often and rarely does Firefox 2.0.0 come out on top. In fact, the last time Firefox eclipsed Internet Explorer 7 was the last time StumbleUpon showered us with affection. Needless to say, the MSIE 5.5 users (yes, omg wtf, etc) were not Stumblers.
This is not a huge increase in traffic by any means - when a really big StumbleUpon day happens, the chart looks a lot different. However, the interesting thing is not the fact that StumbleUpon happened, it's that it keeps coming back to the same page and yet that page's reviews never change.
While Google Images has a long-term love affair going with the site in question, StumbleUpon visits about once every few months, and always to the same URL. Funnily enough, Google Images prefers the pictures on this particular page as well. The really strange thing about the on-again-off-again relationship that StumbleUpon has with this piece of content is that the reviews don't change and, as far as I can tell, few people are particularly impressed with the page. According to its reviews, only three people like the site and it has only been written about twice. And yet, every once in a while, StumbleUpon is back outside, throwing stones at the window and asking us out.
I know that the StumbleUpon reviews pages are supposedly far from accurate: the service has always been a bit cute with its numbers and I've never seen a review page that has accurately portrayed how much traffic the site in question will receive. However, it amuses and surprises me that StumbleUpon does not work like most other social media sites, which will not "re-promote" content after it has been popular. So far, StumbleUpon has revisited this page four times.
It's great for us - writers, publishers, linkbaiters, and SEOs. The do-or-die aspect of Digg is negated by the notion that not only is StumbleUpon popularity and traffic not a ticking clock, but that even once the first wave of SU traffic dies away, it may return again. I am not entirely sure what sets off the subsequent influxes of stumbles (I assume it is a tag or a thumbs-up from a prominent account), but there is one theme in common with each wave of Stumble traffic: every time it comes back, it brings with it more visitors.
However, my analysis of StumbleUpon traffic has been really interesting for more than just the fact that it can't leave my one page alone. A common idea I've heard from the Internet marketing community - and one that I used to spread - is that StumbleUpon is the everyman's social media site. "Regular" people - people who don't get FAIL or Rick Rolls and who squint at me when I go out in this tee-shirt - were thought to use StumbleUpon, at least in higher volumes than they do any other social media mainstay.
I can tell you from my little blog's stat counter that this really isn't as true we we'd like to believe. This morning, I went through the stats and analysed my StumbleUpon audience in comparison to my regular visitors.
Firstly, the graph of browsers is telling. I check this graph relatively often and rarely does Firefox 2.0.0 come out on top. In fact, the last time Firefox eclipsed Internet Explorer 7 was the last time StumbleUpon showered us with affection. Needless to say, the MSIE 5.5 users (yes, omg wtf, etc) were not Stumblers.
That search traffic, visitors referred from non-StumbleUpon sites and people accessing the site directly or via bookmark aren't using Firefox is plainly obvious. Normally, Internet Explorers 6 and 7 battle it out for supremacy. I broke the last two days' worth of visitors down by browser further, showing Stumblers next to their "regular" peers.
StumbleUpon makes a toolbar for Internet Explorer browsers. Either no one uses it, or StumbleUpon serves different content to people based upon their browser. I highly doubt the likeliness of the latter scenario.
People who didn't arrive via StumbleUpon show the regular distribution of browsers for the site. In my opinion, my Firefox visitor count is higher than normal as well, since I've linked to the site from SEOmoz and most of my SEOmoz traffic comes in using this browser.
Without making too many sweeping generalisations, I'd say that the Firefox dominance from StumbleUpon suggests that StumbleUpon is still catering to a very web-savvy audience - a far more savvy audience than I'd previously thought. This isn't such good news, because the content on StumbleUpon always suggested that more and more regular people were using the tool. Since it's "regular people" we're often trying to reach, it's a bit disconcerting to realise that StumbleUpon isn't as diverse as we'd like to think. The blog I'm citing in this post isn't in the least bit "geeky" and I'm now thinking that the Stumblers who've found it interesting are simply typical social media users who happen to also have an interest in my non-geeky content.
I don't want you to think this is a new lesson to me, as anyone who's seen social media traffic has seen this type of thing before. What strikes me as interesting during the latest couple of rounds of StumbleUpon traffic is the complete lack of diversity in users' browsers. Even previous Digg and Reddit outings didn't result in this complete Firefox domination.
Do forgive the dual topics of this post: it can be said that I've investigated two entirely different StumbleUpon phenomena in the recent past. The things I've learned in playing with Google Images recently are a different post entirely.
Without making too many sweeping generalisations, I'd say that the Firefox dominance from StumbleUpon suggests that StumbleUpon is still catering to a very web-savvy audience - a far more savvy audience than I'd previously thought. This isn't such good news, because the content on StumbleUpon always suggested that more and more regular people were using the tool. Since it's "regular people" we're often trying to reach, it's a bit disconcerting to realise that StumbleUpon isn't as diverse as we'd like to think. The blog I'm citing in this post isn't in the least bit "geeky" and I'm now thinking that the Stumblers who've found it interesting are simply typical social media users who happen to also have an interest in my non-geeky content.
I don't want you to think this is a new lesson to me, as anyone who's seen social media traffic has seen this type of thing before. What strikes me as interesting during the latest couple of rounds of StumbleUpon traffic is the complete lack of diversity in users' browsers. Even previous Digg and Reddit outings didn't result in this complete Firefox domination.
Do forgive the dual topics of this post: it can be said that I've investigated two entirely different StumbleUpon phenomena in the recent past. The things I've learned in playing with Google Images recently are a different post entirely.
Bizarrely, I was discussing this exact phenomenon with a client yesterday.
One of his internal stakeholders was doubting their entire analytics package (HBX, so nothing to sniff at) as it suggested that 90% of visitors from SU were on Firefox. I have to say that my client & I thought that this made perfect sense.
You see (or at least I think I do), SU is only for normal people in relation to other Web 2.0 sites. The really normal Web 2.0 sites are YouTube, Facebook & MySpace (Bebo/Orkut etc.. depending on location); everything else is for geeks.
But because the users of SU tend to have experienced puberty, and can handle more concepts at once than Apple vs Microsoft, or Rickroll vs FAIL, we assume that they must be normal.
No, we're all freaks here in SEO/SMM land, but some of us are freakier (and geekier) than others....
I certainly see that more now than I did before. If I'd been asked to guess (without having looked at any stats or analytics), I'd have guessed about 75 - 80% of SU users to fit our definition of geek. I'll now stick with 90-99% :)
The aspect of StumbleUpon traffic that drives me crazy is that, while every metric looks good, subscription rates (conversion) are terrible. My theory is that, while Stumbling, people get stuck in that mode. That means that they're much less likely to bookmark or subscribe than they would during normal browsing, assuming that StumbleUpon will eventually pull up other good content from that site.
Oh yes, the "time spent on site" stats aren't all that impressive (enough time for people to browse the images, usually). I'm guilty of a short attention span whilst stumbling as well. Not immediately enthralled? Stumble on...
"The aspect of StumbleUpon traffic that drives me crazy is that, while every metric looks good, subscription rates (conversion) are terrible."
Pete -
Big thumbs up to that brilliant observation. The very name of the activity "StumbleUpon", I believe - places the surfer in the conscious, and I would venture to guess - subconscious state of "stumbling".
I suspect that the likelihood of a conversion under this mindset is perhaps an order of magnitude less than under "normal" circumstances.
That's pretty troubling when contemplating the value of StumbleUpon traffic. I know from personal experience that when I Stumble, I keep going; quickly passing sites that I might otherwise stick to.
Conversely, maybe it evens out since Stumble seems to send an order of magnitude more traffic than other social media sites.
Regardless, it would make for a great case study to compare the conversion rates of various social media type traffic.
Unfortunately, you would probably need an expert in "usereffect" to prepare that type of study. As we all know, those guys are extremely difficult to find. :)
If I were brilliant, I'd be able to parse my StumbleUpon traffic and make sense of it, but no luck on that so far. It's all just guesswork at this point. It is interesting, though, that, especially with the toolbar, SU is almost like meta-browsing.
Exactly... Information consumers are not necessarily product consumers.
The act of "stumble" carries a non commitment conotation...
There will be many new tools and ways to bring people to your site, but if you broaden the road you will be diluting the results...
Even iTunes free downloads get bounce rates.
I think most "regular people" (using the above definition of FAIL, Rickroll, etc.) don't use any social bookmarking sites when they browse. They either bookmark it as a favorite, or e-mail it to their friends if they want to share.
But I do agree that StumbleUpon is more Web Everyman than any of the other sites.
And it makes sense when you say this: StumbleUpon makes a toolbar for Internet Explorer browsers. Either no one uses it, or StumbleUpon serves different content to people based upon their browser. I highly doubt the likeliness of the latter scenario.
Based on anecdotal evidence entirely, I think people who use Firefox, use it because they CAN add things to their browser. And people who use IE don't even think about adding things to it.
Unless they're in the web business, they might not even know they can add toolbars (Alexa, Stumble, Google, etc.) ot IE. So to me, it would make sense that more users of StumbleUpon toolbars use Firefox.
And I think one reason that you get a flood of people - is from people who actually use the "Stumble!" button in the toolbar. I know when I'm bored I'll just log into one of my Stumble accounts and click through and rate stuff all night long.
The same things tend to come up after awhile and I think it might have something to do with the algorithm on how Stumble chooses to present information (a combination of recency and quantity of votes).
So in a sense, stumbles begat stumbles. The more you get, the more you're going to get until something changes and the SU visitors drop off.
But you hit the nail on the head - the greatest thing about StumbleUpon is the longevity of traffic it brings. Unlike Sphinn, Digg, Reddit, etc - Stumble Traffic keeps coming back long after the page first got attention.
That's actually a great idea! I never thought of that that way. But I tend to agree...
This is where all of them are coming from, and my theory is that SU will send one or two people my way every now and again, and if one of them gives it a thumbs-up, the flood gates are opened again.
I'd second that theory. I'm amazed how few thumbs it takes to re-grab StumbleUpon's interest. I've had some spikes that seem to come from a very low number of votes, relative to other social sites.
"...the greatest thing about StumbleUpon is the longevity of traffic it brings."
Yeah, but if it doesn't convert, it's like getting a bad case of the crabs. They just keep on coming.
Wow, thanks for that tidbit!
That's certainly an interesting simile, Sean.
Not sure about browser distinction (but tend to agree with Vinny on that: FF users are more plugin savvy) but SU traffic is really something I can't stop trying to figure (and most of the time fail to). It drops and rises without any visible reason and sometimes just can't be explained...
Few definite logics I managed to find is that (1) SU traffic depends on the submission category (I guess some categories just don't have enough active people interested in them) and (2) SU traffic rise and longevity depends on the reviews/tags/thumbs acquisition rate [if the post gets plenty of them in an hour, it will send you lots of traffic for a long time], (3) thumbs down do not stop the traffic flow (at least thumbs down followed by the review as you can't tell how many of the uncommented thumbs down you actually got).
Excellent article on Stumbleupon and the traffic it generates. Your conclusion is extremely important for a lot of people to understand - that all social media traffic is not created equal. Give me Stumble traffic over Digg or Reddit traffic any day of the week.
PS - Is it bad that I would be more attracted to a girl if she was wearing that tee shirt out and I so happened to pass her?
I'd take SU traffic above Digg and Reddit, too. Visitors are quieter, more polite, and generally better guests than the other two sets. They also sometimes out-number the other two after a period of time.
Re: tee shirt. My closet is slowly filling up with geek shirts and I love them all :)
"Re: tee shirt. My closet is slowly filling up with geek shirts and I love them all :) "
I'm so glad to hear that, because you have another one on the way. :)
Really?! That is awesome - thanks!
So we're bribing mozzers with T-shirts now?
Honestly, there are few better ways :)
Vin,
That's only because I forgot to send Rebecca and Jane flowers on Valentines day.
That said, believe me - there is no form of bribery that is below me. In fact, I'm even considering baking and sending a SEOmoz cake to them - I'm just having a bit of trouble with the decorating. :)
Cool whip in a ziplock bag - I'm telling you.
Very interesting post. I recently got an influx of traffic for a post back in January and had decided it was time to research StumbleUpon a little more.
On first impression, I didn't think it was catering to my target market but the traffic and this post makes me wonder.
Btw, I love the website mantra...but sometimes it feels like it cuts too close to the truth!
The first thing you mentioned, the "re-promoting" has actually kinda been a pain in the butt for me.
I ran a contest last year for a free SMX pass.
What was last year, should stay last year, however, Stumbleupon has been sending thousands to that page.
I am not doing that contest this year, however, last month I had people coming to that page in hoards via SU.
In any case, time sensitive pages can be affected by "re-promotion" in a bad way :(
I hadn't thought of how that could be a pain. 301 to a Rick Roll? Or Gab's more thoughtful idea!
;)
I look forward to reading what you have to say about Google images. I had a flood of visitors to our store blog from Google images, apparently looking at St. Patrick's Day bulletin board sets. Since I had not put "green beer" or "wet T-shirt contest" in the alt tags, I've been mystified by that. Normal or not, classroom leprechauns shouldn't get people that excited.
Great post Jane and I am quite pleased that you enjoy our robots.txt t-shirts.
The nature of Stumble traffic is quite pleasing in that the traffic is spread out over a longer period of time as opposed to Digg's 15 minutes of fame style. I also find Stumble users alot more friendly in general. But I was surprised by some of your findings, which is a good thing..... means I learnded something! :-)
I agree that Stumblers are usually more friendly and polite than many other social media users, specifically those from Reddit and Digg. They're usually very quiet, actually. They barely comment and when they do, they're not infuriating!
Yes Stumbleupon does get you some interesting traffic, but I have always received better conversions and keyword impact using Digg. The impact also last longer.
The one great thing about Stumbleupon is for creating an audience for a blog. As long as your content is worthy you can pick up quite a few new subscribers. The key for me was having a fair amount of link worthy content on a blog before I first started using Stumbleupon.
Great Post
Bill
Small, amusing addendum: the last SU visitor was at six last night and there have only been two Firefox visits since then, one of which originated at Digg.
Jane, thanks for the great post!! StumbleUpon is awesome for blogs, I think. When one of my posts got into StumbleUpon I got a traffic spike that took it from 300 hits per month to 20,000. And after the dust settled it now is getting about 3,000 to 4,000 per month and growing.
My blog has only been up since last September and I've been trying to promote it here and there though directories and digg, etc. but the most effect was from StumbleUpon. It seemed to have made the biggest difference in traffic growth.
I have not seen it spike up since then but now that I read your post and crossing my fingers and hoping for a sequel.
My 2 cents. :)
I wouldn't be surprised if you see the traffic return. Like I said, I'm on Round 4 for identical content... Pat's seen this happen too.
When a person hits that thumbs up button, the "people who like..." section at the top of the stumble page will have new avatars added to it. Since you say that only 3 people "like" your site, my guess would be that StumbleUpon uses the time between stumbles in its algorithm. If a person stops on one site for a longer period of time than others, it may be said that it is more interesting/valuable. If that's the case, it might explain these ghost surges....
a stretch, I know, but I don't think the thumbs up is what has changed here. If not that, then what?
Coming a bit late to the party but I suppose the high rate of stumble-ff2 visitors is related to the firefox and stumble-toolbar prefetch mechanism.
In my experience, traffic spikes from Stumble tend to be one-hit wonders, concentrated around a few hours of activity and often returning back to prior levels the very next day. So I'd be interested to know if anyone can tell the difference between a post that is "one and done" versus one like this that gets recurring activity...
I always assumed that SU traffic was a once-off deal as well. I don't know why some seem to be recurring hits, but I'm definitely going to monitor all SU activity from now on to see which content gets re-visited and which doesn't.
Great post Jane! Have noticed the spike from StumbleUpon traffic when reviewing analytics. Recent experience includes a new customer website where the link building process has just been initiated. StumbleUpon returns the most traffic by far, but in surges that really look interesting on the analytics graph. Definitely need to assess conversion rate on this traffic. It is undisputable that StumbleUpon returns the most traffic of any link, but the quality of the traffic definitely needs investigating.
As for IE vs Firefox...Firefox is king, if the general public will just accept this. People are soooo scared of change, and would rather stick with IE than change. Once they've been converted there's no looking back though! Have an exSEOllent day everybody!
Yes Jacques, even I fail to understand why people love IE so much when firefox is 100 times better than IE. May be they don't switch coz they love seeing "waiving window flags" and "rotating earth" on the screen when IE loads the page. Firefox rocks.. for both users n us developers !!
As with many of us I think that you're making the mistake of assuming that we're normal, and therefore that the general public are just like us.
The general public aren't afraid of changing to Firefox; they don't know what the **** it is. Simple as that.
When they get a perfectly good browser (when all you've ever known is IE, why would you expect any different) installed with their PC what would make them think of looking for a new one - as far as most people are concerned, IE & Google = teh internetz (sic)
I know people who still think that this is the internet!
Agreed 100%. We are mostdef not the norm :-)
I'm also pretty sure the reason why few people use Firefox is because they haven't heard of it. The only other reason I've ever heard for not using it is because "it sucks," which was said to me by someone who'd not used it since some time in the early 2000s... when it very well could have sucked.
Have any of you actually installed the StumbleUpon toolbar on IE?
I think this could be a significant part of what you're observing.
My experience could be a minority, but I've tried installing SU on IE twice (about 12 months apart) and both times it resulted in frequent random crashes of the browser (maybe about one an hour - usually when opening a site in a new tab, with or without using the stumble button),uninstalling the toolbar instantly stopped the problem both times.
It can't be an isolated case... I'm pretty sure the SU toolbar in IE has problems, and has had for at least a year... I don't get anything like that happening when using it in Firefox.
My old laptop has the SU toolbar installed in IE and I've never had this happen, but I only use IE solely to test whether or not things work in IE, so it is open very infrequently.
Yes Jane. even I wonder what exactly triggers stubleupon's repeated visits. Secondly the traffic that I get from stumbleupon looks like traffic crackers. On Google Analytics, the daily traffic from SU looks something like this 0 0 1 0 0 '316' 0 0 . Why aint the traffic uniformly distributed for a period of time.. ? Do the number of reviews on a stumbled item effect its stumbling frequency? The best thing I like about SU is like you said -"Do it right now or die" like DIgg. You can hope visits even days after you have thumbed your content. Nice Post Jane. I look forward to read more fromyou on SU.
Why isn't the traffic uniformaly distributed? No idea: it has to be due to the number of thumbs-up or positive reviews going up, although the page I've written about here doesn't have any more reviews that it had the last two times SU sent visitors.
Another theory I have is that when more people express an interest in a page's topic, SU recognises that the page probably has a new audience. They supposedly don't show the same person the same content twice, so if 1000 people have recently joined SU and added "blue widgets" to their interests, my blue widgets page can be brought back into play.
Great post, Jane. Thanks. The IE / Firefox stats are interesting ones, and ones that I rarely pay attention to since our content that's most interesting and best lends itself to linkbait is definitely geared towards a very web-savvy crowd (so, I'd assume they're heavy Firefox users). I love StumbleUpon...not just because it occasionally brings our site a lot of traffic but because of the whole "new site discovery" aspect of it. I recommend it to people ALL the time - non-web-savvy people included - just because it's fun. So, I'd guess that those percentages will mellow out a little over time. Do you agree?
That's one of the things I've found most interesting about this latest StumbleUpon experience: the content isn't geared towards a web-savvy audience at all.