Measuring the ROI of Social Media Traffic to Facebook

The bane of Social Media is a constant state of developing metrics.  While still the industries favorite buzzword, social media must constantly prove its value and worth by climbing the slippery slope of ROI.

It’s a hard enough task, but many clients are spoiled by the relatively clean value of PPC and even the measurable results of SERP rankings that SEO can provide. They want the same kinds of proven value from social media, but like a rebellious teen, social media isn’t fitting so nicely into that prescribed ROI box.

Developing Social Media Metrics with Google Analytics

A few weeks ago, I went on a quest for a new Holy Grail of Social Media. Overheard at a webinar that was being used for training at the DragonSearch offices, my ears perked up when one of the participants mentioned “Google analytics on a Facebook Fan page”. Could I finially have a clear method of measuring true Social Media ROI? .

While Facebook does have, of course, their own internal Insights to a fan page, my immediate excitement caused me to think that, like GA for websites, GA on FB would allow me to see where the FB traffic was coming from. I wanted long juicy lists of referrals from other social media venues like Twitter, and I wanted to see traffic flowing from blog posts and comments. I had hoped to see links coming from Foursquare applications and other social links sprinkled all over the web. I want to measure my results from various social media campaigns and to be able to see what is working!

After a few failed attempts, I joyously did a little happy dance and watched as GA was, successfully, beginning to pull data off of a few of the Facebook Fan pages we manage for our social media clients.

Google Analytics Thwarted by the Walled Garden

I waited. I hoped. And then, I had to let out a disappointing sign. Even the awesome power of Google Analytics cannot weed its way past the Facebook blockade. Alas, no matter how far I drill down into GA for a Facebook fan page, the numbers say very little.

Yes, the numbers of visits to a Facebook page is shown as is some interesting numbers relating to visitor loyalty, bounce rates, trends, location etc. but that was not what I really wanted to see as much of that can come from the internal Facebook Insights.

Measuring the Traffic to a Fan Page

Of a larger interest was the Traffic Sources.  Here GA results showed a smidgeon of traffic coming directly to the Fan page, a huge zero from search ( ?)  and over 99% coming into Facebook from referring sites.  Here’s the plum in the pudding I think!

But no.

Alas, Facebook traffic seems to become nothing but Facebook traffic once anyone comes upon Facebook. All other previous traffic history is erased as if a link only originated in Facebook and Facebook only.  For instance, while the Facebook general referral does get broken down into smaller individual pages, like so:

Most of the referring traffic comes from the first blacked out page (personal names and accounts protected per privacy, etc) which is the name of the actual page we are measuring itself! So it comes from there to there? Self referrals? Can I please be like a medical HMO and deny Facebook based on its self referral?

Following a link to say, brings one to their OWN profile page.  Not very helpful at all. I know that all this traffic did not come from ME!   However, the one small bit of hope I do see is that IF another Facebook page or even a user sends traffic, then the URL of that account or page WILL be shown as the second blacked out site is. It is an actual person’s Facebook account page.  This will be useful in identifying the better more active fans and brand ambassadors.

Ok.. Disappointment of social media ROI results is a constant companion, so let’s compare what we CAN see!

Comparing Google Analytics to Facebook Insights

Since the two do not speak the exact same language, we first need to translate what they mean so we can get nice apples to apples comparison.  “Interactions” on a Facebook Fan page seem too nebulous ( and also, too much math) so I am looking at Page Views since that seems like it would be closest to GA’s visitors.  I also set both to measure the same time frame: March 31st when GA began to pull data to April 9th, when Facebook stopped.

Google analytics does the math for me and therefore is my friend:

Facebook, on the other hand, requires addition. Do notice, however, that we do see similar graph lines! Dare I be hopeful yet again?

After checking my math twice, Facebook results for the same time frame say:

1904 unique page views and 4150 total page views

These numbers are are quite different from the 2,309 that Google Analytics produces. Oddly enough, however, and whether this actual means anything.. if one subtracts the Facebook numbers you can get close to the GA number!



2246 which almost equals the 2,309!

I doubt that really is more than an odd coincidence.

In fact, GA’s pageviews and FB pageviews still do not compare at all in a way that makes us feel warm and fuzzy:

4,150 per FB

2,546 per GA

So what have we discovered from this experiment?

The ROI of Social Media shall continue to be the Bane of my Existence.

I can produce number. I can produce charts. I can give a client pretty graphs, but what does it all mean and how confident am I regarding the truth of those numbers?

Not very much..yet.

I have still have hope… after all social media is still so very new and the metrics are still developing. I can relax and observe a while longer while the ability to harness ROI matures.

Related Posts
Related Case Studies

Dragon 360 Logo