When Google Analytics announced their new social reports at SES New York last month there was a lot of online buzz declaring Google was “squaring the circle” of social media ROI by putting in place a more robust collection and filtering mechanism for more than 400 sources of social media data (including the most familiar suspects along with several more obscure sources most of us wouldn’t have thought to specify in an advanced traffic filter).
I could have joined the fold of people speculating about the new features, but I prefer to write about platforms I actually work with, and the insights coming out of using those platforms to solve real world problems, and didn’t yet have access to the new features. Now I do. I reasoned that an ounce of my own personal experience is worth much more than many hundreds of pounds of someone else’s.
Platform Speak vs. Reality
After examining the new features of Google Analytics on my own blog, I found the platform speak, as explained in Google Analytics Social Overview page, (“where social traffic sources and pages identify communities interested in my content”), doesn’t match up with my experience of what the platform provides.
Even the measure of a social “assisted conversion” only tells us that a social media takes part in certain conversion events, but doesn’t tell us by how much – at least, not yet.
What Google Analytics is actually providing with these new reports is similar to a “faceted mirror.” There are a lot more bits and pieces of information, some of it useful, much of it not, providing me with more information, but not telling a story. Perhaps only an analyst, particularly, someone who is a subject area expert, can leverage the information that Google Analytics provides us.
Let’s look a few examples from the new social analytics reports Google Analytics now provides.
Social Network Sources
For example, networks such as Twitter and Facebook are communities in much the same way that Manhattan and Brooklyn, two places I hang out at, often, but most of those people there do not act together in any communistic sense that relate to my content, so I’m not getting any useful information about people who visited my site due to social media really are.
Traffic networks are mixed up with traffic tools. For example, Twitter and Facebook are termed networks while HootSuite, which is a tool, and WordPress, which is a blogging platform, are treated identically by Google, as being the same type of traffic – social media. While technically true, it’s incorrect to place these sources on equal footing. They should be further subdivided so as to better represent what they really are.
Six percent of the conversions I set up on my site are connected to social media, and 4 percent (2/3 of the 6 percent) show social media what the last interaction before converting – and I have no idea if that is a good or bad percentage, though we should always strive to improve it. Other profiles I have access to where Goals were set up (without setting Goals in GA, conversions cannot be calculated) show 2 percent of a large CPG site has social conversion activity while a small restaurant chain in New York City has a 3 percent.
In every case, there’s a lot of information and context a person must add over the Analytics to make sense of what Google provides, suggesting a user of these tools must first frame their activities and marketing goals in terms of how the platforms are designed and structured, align the tools to measure those goals, and then decide what it all means.
One could argue, we, as marketers, should take the time to structure our communications medium along the line of what, and how the platforms measure. If we did that, our Google Analytics readouts would be much better and more insightful.
The traffic to my site that came from my LinkedIn profile indicates people who viewed my profile from within LinkedIn’s own URL structure were double (10 visits) those who used the my LinkedIn personal profile URL (5 visits), but unless I better understand why someone would prefer one view over the other, the distinctions are probably meaningless, even though people who viewed my personal profile URL spent almost 7 times longer on my site, than those from LinkedIn who didn’t.
Pages Getting Traffic from Social
They say that content is king in social media and in SEO, and it so happens one of my most popular recent posts is on the subject of Big Data. According to the sharing widget on my site, my content was shared 30 times.
Yet Google Analytics records only 21 visits
What happened to the other 7 shares? Does a share necessarily result in a visit back to my site? Probably not! I suppose I could interpret the 21 visits as a representation that 3 out of 4 times that my content on Big Data was shared, it was clicked on and a visitor viewed my blog post).
While Twitter drove the lion’s share of the social traffic to my blog post, clicking on it shows me a bunch of shortened URLs, most of which don’t make much sense unless I deliberately create and apply these URLs to different situations beforehand, something that is very difficult to foresee and structure effectively.
In the case of t.co/GM24ACUU, the average visit lasted over 22 minutes, far longer than any other referrer, and yet, unless I can tie the t.co url back to the individual account on Twitter who shared that URL, it’s unlikely to do me any good.
Clearly, GNIP’s new service that provides 30 days of historical information on Twitter (and may eventually extend as far back as 2 years), for a price, might give me the answer I seek, but it would cost me a small fortune to get it (and I’d need a programmer with access to the GNIP API, first).
New & Noteworthy Features
Google Analytics provides detailed information about Google+, and if you’re heavily invested in Google Plus, the readouts can be very helpful, otherwise, not so much.
The Social Visitors Flow charts are the most interesting visualization the new reporting provides, because it can be segmented in many different ways, including geo-location (such as in the image above).
Imagine my surprise to discover that there were 51 visits to my site from Madrid. That fact alone might give me some ideas about cultivating Madrid in my marketing efforts, but unless I can find those people and reach out to them, it may be difficult for me to do much useful with the information (similar to the analogy of the faceted mirror mentioned earlier).
The new social reporting in Google Analytics provides a marketer and site owner a lot of additional information, but as tools such as Google Analytics continue to evolve, they typically requires a lot more configuration and planning, beforehand, in order to leverage the new capabilities.
Twitter Canada MD Kirstine Stewart to Keynote Toronto
ClickZ Live Toronto (May 14-16) is a new event addressing the rapidly changing landscape that digital marketers face. The agenda focuses on customer engagement and attaining maximum ROI through online marketing efforts across paid, owned & earned media. Register now and save!