Banner
    The Data: Social Media Stinks At Driving Traffic
    By Hank Campbell | August 15th 2011 02:52 PM | 11 comments | Print | E-mail | Track Comments
    About Hank

    I'm the founder of Science 2.0®.

    A wise man once said Darwin had the greatest idea anyone ever had. Others may prefer Newton or Archimedes...

    View Hank's Profile
    Over time, a variety of people have asked me where they should get the word out that they are writing on Science 2.0 and I am happy to tell them - but they are surprised by the answer.   Social media darlings like Twitter and Facebook will accomplish very little beyond being another place to follow comments.  Old technology is what matters in getting eyeballs to read your work, and that means a place like Stumbleupon.com, Reddit.com, Digg.com or Slashdot.org.  They are not social media, they are social news and that is where the science readers are(1), outside your own base and search engines, of course. 

    A group called Outbrain did an analysis of recent traffic trends and say the same thing - on social media, people react to the title, they click the link and read less often.   If you have a large Twitter or Facebook group, it's fine to get a lot of repeats of your posts, but you are getting little traffic, and if someone is paying for a social media marketing campaign, they are insane.  

    Overall, Facebook may drive nearly as much traffic as The Drudge Report, for example, but it is a million submissions to do it - if you get an article on The Drudge Report it will crash your server with the traffic and you will only notice traffic from Facebook if you are really looking.  And many people look at linked Tweets as bordering on spam.

    Breakdown of external traffic sources by type

    Are their numbers reliable for science?  The results will look a little different.  While our overall traffic from something like Reddit is quite low, your individual traffic if your article makes their front page is quite high.   I have had people with 500,000 followers on Twitter post Science 2.0 articles and it is maybe 500 readers to the article but a solid article on Reddit will bring you 30,000 readers or more.   On Twitter, though, if Ashton Kutcher tells a million followers to buy tequila, maybe that works; Reddit followers would make goat noises and hack his computer if he tried that over there.

    Drudge Report does not link to non-corporate science sites so there is no hope of you getting there but social news sites without editors can help you get a new chunk of audience, it just takes patience and respecting the community (i.e. not showing up and spamming them with your stuff and expecting great results) in each individual place.  It's a lot of work but the biggest names in science media either do it themselves or have marketing groups doing it for them.

    Google is far and away the leader in driving traffic via search.  Being on the Yahoo!  front page might be Drudge levels of traffic but, like Drudge, they only use certain sites picked by editors and never independent media - what is achievable for an individual, Yahoo search, is almost nonexistent which means having a quality article that shows up in Google search is still the best way to go over the long haul.

    NOTE:

    (1) As in anything successful, those social news sites are heavily populated by marketing people for other science sites, so don't be discouraged if your submissions get clobbered. Over time, the audience will catch on to tricks.   Stumbleupon is mostly valuable only once you get a decent group following you so dirty tricks by competitors will happen less there.

    Comments

    vongehr
    Thanks for the tips, Hank. Renders me happy never to have cared about twitter or facebook at all.
    like Drudge, they only use certain sites picked by editors and never independent media
    Didn't Drudge report once start out as some sort of alternative thingy? Well, success corrupts everything I guess. Another one my intuition about ignoring was right about, ha ha.

    Yes, reddit is not bad, if one selects the right categories (never put interesting physics into the physics category for example). Stumbleupon is another one I am interested in, but I still do not understand it. You say
    Stumbleupon is mostly valuable only once you get a decent group following
    Why would those be following you on stumble not just follow you on Science2.0? This surely can only add traffic if you are at least initially quite dedicated to what they are doing at stumble, namely stumbling, and one does not "stumble" upon one's own stuff. Isn't such time better invested just writing more and better content on Science2.0 to let other people stumble upon it? I would really appreciate more practical suggestions [you know, I stumble often upon your stuff actually ;-) ]
    Hank
    I appreciate that!  What I meant was, like us, those communities are inhabited by people who can smell someone exploiting them for marketing rather quickly so it would do little good for me to get a Stumbleupon account and put 10 of my own articles - it would go nowhere because I have no trust there.    But if I spend time and post interesting stuff from all over the Internet on stumbleupon and build trust with their internal system, like Bente and Andrea have done, then when they do contribute something from here, it has value and gets readers.

    It is the same on reddit or digg but you can get some readers just putting stuff on those two - the downside is those two sites have experts in marketing for other sites who have become trusted internally there, so when they vote something down, it means more also.   None of us are spending the time to get big on social news sites so we basically rely on a reader to like an article and put it there.

    Michael Martinez
    There is a nepotistic effect that cannot be avoided in these service provider studies of Internet referral traffic.  That effect stems from the structure of the Web itself more than from the methodologies used to produce these studies.  The methodologies are fine for the groups of sites being studied by their findings don't shed much insight for the larger Web.

    Web traffic buzzes around communities of sites like bubbles that get less dense (less traffic from fewer sources) as you move away from their centers (closely aligned Websites and personalities).

    You say Yahoo! is not very helpful for individual sites.   That's a common complaint, and yet I have seen reliable statistics for niche Websites that receive as much traffic from Yahoo! as from Google or Bing.  I have also reviewed referral statistics for a small number of sites that receive more traffic from Yahoo! than from Google.

    There are niche markets out there where the Web traffic from Twitter is quite substantial.  There are also moving markets like John Scoble.  In January Matthew Ingram wrote on GigaOm that Scoble's 100,000+ loyal readers follow him from social media service to social media service, trampling other users' visibility in the process.  He is currently crushing Google+ users (Rocky Agarwal announced on TechCrunch in July that he had blocked Scoble on Google+).

    Whenever someone starts citing statistics about how much traffic such-and-such sends, I can usually find counter examples.  It's not because the other person is wrong -- I have rarely found anyone's statistics to be wrong.  It's just because we're all seeing relativistic traffic bubbles (if I may bastardize the terminology somewhat).
    Hank
    What you cite are isolated instances that deviate from the norm but it doesn't disprove anything - somewhere out there someone may claim Bing is better for them than Google too.  But there is zero chance that Yahoo or anyone else drives more traffic than Google.

    Does Google+ currently drive more traffic per link than Facebook and Twitter?  Early results here say yes.  Maybe Google+ has more science people per capita.  Twitter is a non-entity and Facebook is okay, just not worth spending time building.
    Michael Martinez
    There is no norm.  There probably never has been.   It's a myth that Google is the primary source of traffic for every Website.  In fact, since the Google Panda update in February (and the iterations that have followed), a large number of Websites lost most of their Google traffic.  Many of them now receive more traffic from Bing and Yahoo!.  I have seen the data for hundreds of such Websites.

    Social media is also a primary source of traffic for some Websites.  I reviewed social media case studies where the site operators set out to attract traffic ONLY from social media.  They ignored the search engines.  The data is very real.

    In 2008 I was brought in to analyze traffic patterns for a very large events directory.  Their Google traffic was less than their Yahoo! traffic.  After reviewing their server logs I was able to point out issues that were impeding Google's crawler.  That Yahoo! was able to crawl their content better was probably due to Yahoo! having direct access to their data.  I have seen this on other sites.

    Google is also very aggressive in banning or penalizing Websites that continue to function on the basis of traffic they draw from other services.

    And then there is internal search or site search.  Major retail sites like Amazon, eBay, Hewlett-Packard, Wal-mart, et. al. serve 10s of millions, even 100s of millions of queries across their own content every month.  The amount of traffic that site search generates is immense.  Studies like Outbrain's are interesting but they are not representative of what is actually happening on the Web.
    Hank
    There is no norm. There probably never has been. It's a myth that Google is the primary source of traffic for every Website. 
    Again, my article never said that was the case, nor did I say every website is ill served by social media.  I'm not sure why you keep going back to that.   Social media does stink at driving traffic - Wal-Mart is not part of social media nor is Yahoo.  If any website is really getting a lot of traffic from Facebook, that's wonderful, but I would like to see data if anyone claims any content site (not content farms, images, whatever) of decent size is getting more from Facebook per article than they get from something like Stumbleupon - it doesn't exist.  You keep telling me I am wrong in saying that everyone else also sees a failure of social media to do anything good for writers - this site included - but have you to show a single instance other than saying 'some websites'.
    Michael Martinez
    It was only my intention to point out that the Outbrain study is not representative of the entire Web.  It's easy enough for me to provide a few counter-examples.  I was trying to point out that there are different demographics at work, and these studies really don't take those demographics into account.

    In 2010 the San Francisco Chronicle, among other sites, cited a Compete study that said Facebook was driving more traffic to portals like Yahoo! and MSN than Google.

    In 2009, GigaOm cited another Compete study that said Facebook was driving more traffic than Google to Perez Hilton's site.
    And that makes sense.  Technical audiences don't participate in social media as much as non-technical audiences.

    This PDF of a presentation from the UK in 2010 has an interesting chart on Slide 18 ("USA Educational Level") that shows there is less participation in social media from people with graduate degrees (and of course it's a smaller demgraphic).

    That said, it's easy enough to find conflicting studies that say many scientists do use social media (2008) and scientists don't use social media (2011) so it's not surprising your own referral data for Science 2.0 should discount social media as a major source of traffic.

    There have been some creative attempts to map the social media audience but none of them really explain what is going on.

    People find Web niches based on their careers and personal interests and some niches are more closely aligned with social link sharing than others.  Hence, the Outbrain study (based on data they have access to through their own clients/partners) isn't representative of what is actually happening on the Web.  I have yet to find any public studies that really attempt to document or map social demographics and influence.  I've seen some private data that is pretty interesting but even then I was cautioned that it was only representative of the data that could be collected.

    No one is wrong.  It's just that we don't yet have a reliable picture of the lines of influence across the Web.  And those lines change as new Websites become popular and influential.  I can point to my own Website statistics and say that Google+ referrals appear to be increasing as Twitter referrals appear to be decreasing -- but that only reflects the demographic to which my Websites are known and appeal.


    Hank
    cited a Compete study that said 
    Some calibration is in order.   Anyone using data from a free, panel-based service is automatically considered to be using easy-to-find, completely wrong information - SF Gate and a journalist who has to put out a story in an hour is a fine example of that.  Alexa, for example, finally had to hardcode Google in as the #1 site because they famously had them as #2 for the longest time.   Three years ago, when we were 1/3rd the size we are now, Compete and Quantcast had us at over a million uniques.  Now they have us at 100,000 or whatever, 1/10th of what we actually have  - the difference is 1 or 2 people who read us were in their sample then and are not now because they extrapolate their panel out to the mass market.  Comscore is the only traffic site that actually buys ISP data - but they charge $50K a year to use it so they rarely show up in Web articles by journalists, much less web bloggers.

    Again, overall, you are missing the point.  You linked to an article saying scientists use social media, for example, but I never said they didn't. I said social media drives no traffic worth talking about compared to search and social news sites - and that is unquestionably true, even if we both agree these guys have a flawed methodology.   I've still yet to see a content site that considers social media, even as an aggregate much less Twitter or Facebook individually, worth anything at all in traffic.  Google+ may change that but data talks and there is no data (unless your sites have 500K and up in traffic, the data is too volatile to count) that shows it has done so yet.   
    Michael Martinez
    I fear we have reached an impass so this will be my last comment in this topic lest I wear out my welcome.
    I do understand your point quite well, though.  I've been working with Web metrics for over 10 years and I know who buys ISP data (more than just comScore, I assure you).  All the third-party services are using statistical projections that are hardly reliable when compared to individual sites' data.

    All of these studies (and I have read many) fall short because they represent limited points of view, or specific demographics.  None of them is sufficiently accurate enough to be a sole authority on the topic of Web traffic, referral patterns, etc.
    Again, each community sees different things in their analytics.  I have reviewed data from publishers with several similar sites where each site had a distinct demographic and unique referrer profile.  Yes, they shared some percentage of referrers, but they also had significant differences.  Often these networks engage in a lot of mutual promotion between sites.

    Entertainment Websites receive a great deal of traffic from social media -- sometimes more than they receive from search engines, sometimes not (according to analytics I have reviewed).

    Publishers syndicating articles on major portals like Yahoo! and MSN receive a great deal of traffic from those feeds -- sometimes more than they receive from search engines, sometimes not (according to analytics I have reviewed).

    There are "consultants" who specialize in driving traffic to client sites from services like DIGG, REDDIT, Facebook, Twitter, and other social media platforms.  The quality of their traffic is questionable but they do know how to shift the source trends in analytics.

    Nonetheless, I have always maintained that no one has a complete set of data, and therefore none of us has a full picture of what is happening.

    If we can agree that search provides a natural, rich channel of traffic to Websites, then let me suggest that marketing succeeds in creating other, semi-natural channels of traffic that are no less valuable or important.

    You are using Outbrain for your source of information? I'm sorry but that's like asking McDonalds's for information on staying healthy.

    Outbrain is a cancer on the internet. They have somehow managed to stick their stupid news links on every site I visit.

    All I see is the same stupid NewsMax story about how 1 cool trick will change my life by warning of the impending doom on wall street.

    Really it amazes me that site owners are such whores that they'd put this crap on their site.

    Here's 1 cool trick that will get you viewers and retain those viewers and even get those viewers to allow your ads.

    Treat your visitors the way YOU would like to be treated when visiting a site.

    Hank
    You are using Outbrain for your source of information? I'm sorry but that's like asking McDonalds's for information on staying healthy.
    It doesn't sound like they are wrong, it sounds like you don't like them. You said you see something from them on every site you visit (except this one) which means they do have an idea of web metrics and whether or not being on social media helps. The company behind those darn Zwinky ads is annoying too (maybe the same group, I suppose) but they have to know what works by now.

    Do you have opposing data to this?  Obviously it is only anecdotal but social media is truly pointless in traffic here.  We include them in our toolbar for convenience but if we took out all the traffic from social media each day, we would go down basically nothing. Facebook, Twitter and GooglePlus could disappear and no one would notice in number of readers. They're fun, of course, but this is about generating traffic, not interacting with an audience.