The truth about Facebook attribution
With Facebook’s announcement this April that their Q1 2014 revenue totaled $2.5 billion (up 72 percent vs. Q1 2013) and revenue from advertising totaling $2.3 billion (up 82 percent vs. Q1 2013), one assumption has now been undoubtedly proven: online marketers believe in and trust Facebook advertising. So much so, they’re flocking their online ad dollars to Facebook en masse. What still remains unclear, however, is exactly how Facebook marketers are measuring and detailing their “much improved” impact.
I believe full transparency is the key to successful online marketing. In recent months, online marketers are asking more and more relevant and much-needed questions on how to best and most accurately document their Facebook ad efforts. And so, I think it’s time we marketers have an honest conversation and get on board with all the ways and models of attributing Facebook’s sales.
Perhaps you’ve found yourself asking, or have been asked, questions such as these: How can I best attribute Facebook ad sales? Do I count both view-based and click-based conversions? Should I use just Google Analytics? Should I use just Facebook’s Conversion Tracking? Or should I use a combination of the two? Argh, what do I DO?!?
Honestly, there’s no perfect answer for everyone, but there might be a perfect answer for you and your business. In other words, there isn’t just one, simple right way to implement Facebook attribution models. Marketers should be creative, entrepreneurial, and innovative in trying out different recipes for success.
Seasoned Facebook marketers often say they implement various different combinations of attribution models mainly because they want to document the difference between view-based and click-based conversions. This objective is important, yes, but to measure sales effectiveness while also considering the ways in which consumers use various desktop and mobile devices, I recommend online marketers utilize Facebook’s Conversion Tracking as the main point of optimization—while still implementing a combination of other success indicators. Here’s why:
Facebook’s Conversion Tracking counts conversions from people using Facebook that have seen or clicked on a Facebook ad, regardless of device. ‘People’ is the key word here, since this term is dramatically different with Google Analytics. For example, if a person sees and clicks on a Facebook ad on Wednesday on their mobile device, but then visits the seller’s website directly on Friday from their desktop computer AND makes a purchase, Facebook still attributes the conversion back to that person when they clicked on the ad using their mobile device. In fact, 67 percent of users begin shopping on one device and continue on another.
Google Analytics, however, would attribute the same scenario as two unique site visitors—one from Facebook on Wednesday who didn’t convert, and one directly on Friday who did, thus counting two visits instead of one person. I believe Google Analytics provides an imperfect view by measuring purely linear, or click-based, conversions. In other words, only when a user sees an ad, clicks on it, and immediately makes a purchase. In truth, the path to purchase is much more complex with mobile, desktop, laptop, and tablet options all playing possible roles. Google Analytics has attempted to solve this with “sessions,” but doesn’t paint a full picture as well as Facebook does.
The solution to this somewhat murky scenario isn’t implementing fancy software; rather it’s just being honest and transparent with your clients (and yourself) about many different data points. For example, measuring pure and simple Facebook conversions is great, but marketers should also measure other important features such as Facebook’s website clicks (how many people actually got to your site with Facebook ads), Facebook total post consumptions (post impressions divided by post consumptions, showing how many people actually consumed the post when it was shown), Facebook engagement (could be a bunch of different things), all of the Facebook referrals via Google Analytics, and average time on site someone spends coming from Facebook. While continuously measuring and sharing all of these factors as a whole, you will paint a fuller picture of an ad’s impact while still focusing on sales as the main goal.
This may not be the simple, clean cut answer you’re looking for on Facebook attribution questions, but I strongly believe it’s our duty as online marketers to tell a full and complete story—including data and metrics that may not always make us look like rockstars. To build client trust and to continue to implement creative ad techniques, I think we all need to get on board with open, transparent reporting through these various models.
Photo by Praneendra Kuver for Inside Facebook.