Driving traffic from social media
I think one of the goals of using social media is driving traffic to your site, so I’ve been keeping track of how many hits we get on www.oc.edu from Twitter and Facebook using Google Analytics.
A few things to point out. First being that this is not a perfect science. Twitter clients (Tweetdeck, Seesmic, etc) make tracking click throughs tricky. Some apparently count their clicks as a direct hit to the site (no referring website).
I do use bit.ly, a service that tracks clicks, but I haven’t quite perfected the art of dissecting through all the graphs to get to the meat of the data. I have also heard you can use a Google Analytics campaign to track click throughs, but the truth is that I don’t spend that much time preparing what to put on there. So perhaps that’s my problem.
Okay, so back to what I do know.
I searched from January 1, 2009 until December 4, 2009. I went through Traffic Sources > Referring Sites and searched for anything with Twitter and Facebook. I did this for both of our servers, which use different analytics codes.
Results-
Twitter- 2500
Facebook- 7550
So, I think 10,000 hits in one year (so far) is pretty good! I’m happy with it. It’s a nice bench mark to measure against.
Of course this is a drop in the bucket to our total number of hits, but I choose to think of it as “wow, those are all deliberate hits that we wouldn’t get without social media.” Hooray!
I also have hope for click throughs that are getting lost, not to mention links to third party sites or video, pictures, etc.
Interesting note- I did a test to see how click throughs on Seesmic (my desktop client of choice) show up in Analytics. For some odd reason,they showed up as through Facebook. Why is that? I have no idea. Does that totally mess up my findings? Maybe.
Giovanni Gallucci did some research that concludes that bit.ly doesn’t accurately register in Google Analytics. However, budurl.com does track uniques and pass them on to Google properly. It’s a paid service, but it might be worth it to get accurate metrics.
I track clickthroughs by looking at bit.ly to see who clicks through instead of looking at Google Analytics on the site side to see where people come from.
I need to do this because sometimes I’m linking to a page on our site, sometimes I’m linking to a news story that features our faculty or programs, sometimes I’m linking to a page on the central news archives at the main campus. One way or another, it’s all content about us and I track it all.
I’ve been measuring pretty obsessively for over a year and comparing social media to other spots where we plant links (like in a news release to route someone back to a Web page).
For us, with a highly targeted effort to build an identity on Twitter as a campus focused on health sciences/health care research and education, with graduate design studies and some other smaller programs, I’d say it’s been effective in drawing–as you say–DELIBERATE visits by people who know where they’re going.
I’m on Twitter in part because it’s a chance to connect with journalists, bloggers and other thought leaders I’d like to make aware of our research, which I can do more directly than by optimizing our site and hoping they’ll put in the right search term (although we also work on that).
We’ve had over 12,000 click-throughs to date and the most popular items are all exactly what I’d want them to be, as far as ones that are our strongest programs for which we want to be known.
Case study on SNCR site (link is to PDF of all winning entries in the microblogging category–we won an Award of Excellence in the Academic division): http://bit.ly/4wPxxF
Tweets also live on, in the form of archived feeds on web pages all over the place, so those clicks keep coming long after the original tweet.
Now that tweets and unprotected Facebook updates will show up in Google results, they have even more value in driving traffic.
I haven’t yet put the same energy into Facebook. That along with more work on our YouTube channel and a new blog are on the list for 2010. The way I look at it, we can’t NOT be there.
@BarbChamberlain
Director of Communications and Public Affairs
Washington State University Spokane
@WSUSpokane
http://www.spokane.wsu.edu
http://www.facebook.com/WSUSpokane
it’s surprising there is no in-built “social media element” in Google Analytics yet. If that becomes available, then the most important data would be unlocked. No more going from site to site to compare metrics. Hope it happens!