The Big Data and Social Media Crossroad

Posted by Taylor Haney on Thu, Aug 28, 2014

We have a few more fantastic guest blogs for you this month before we wrap up our data and analytics theme for the month. Alex Silberman, Digital Strategist at Arnold Worldwide takes a look at the intersection of data and social media and how important it is to maintain to human element throughout. 

Alex SilbermanAlex Silberman’s role at Arnold Worldwide is to be obsessed with digital performance. From creative to media and all of the little gaps in between, his digital performance group looks for appropriate ways to measure and manage advertising campaigns to drive results for clients. Prior to Arnold, Alex honed his skills in media, product, CRM, and ad operations for multiple agencies, vendors, and one failed startup! 

The convergence of “big data” and social media is ongoing and unstoppable. As digital marketers, we’ve been trained (since birth it feels) to measure and optimize everything, and to replace human decisions with algorithms whenever we can, all for the sake of efficiency. And, I think as a philosophy, this is the correct approach to take. I really do.

However, we need to take a step back and approach data and analytics with a bit of humility and self-awareness, and acknowledge upfront that, as an industry, our ability to track, analyze, and optimize is still sub-par. Data-driven advertising is not at, and may never reach, “set it and forget it” functionality. Context is still important. Intuition and creativity still impact campaigns. Many of the inputs we’re measuring – likes, shares, comments, clicks, engaged time – are rough approximations of human behavior. Even with data and analytics, the people still matter.

We’re also limited in what we can effectively analyze. Too often, we trust the analysis of peers without considering their skill set, or our ad tech partners “black box” optimization techniques. And we do this willingly, so long as a campaign “performs!”

But as an industry we rarely acknowledge these limitations. A majority of people will share an article on social channels without even having read it themselves, but it does not feel like anyone has actually considered, as a data point, what this share actually represents in analytics. We see “engagement rate” as a worthy KPI, but why? What impact did it have on a consumer’s thought process or decision journey?

As Facebook’s own “Edgerank” has decimated organic reach, it’s bumped up these engagement rates by deciding which content is most relevant to which consumers. As a theory, this is fascinating! Facebook has essentially told brands, “you can talk all you want, and we’ll make sure the message gets delivered to the right person at the right time. Trust us.” With Twitter’s recent experiments in algorithmically deciding to show consumers things they did not ask for/expect to see, we can see it’s an industry wide practice. As interesting as it may be, I think this is a disturbing trend. Are we really comfortable letting “big data” decide if our content is worthy, and to whom?

As a consumer, I’m constantly worried about the content I see being filtered based on past preference. I don’t want to fall into some kind of content recursion loop. As marketers, we should be worried about the same thing for our audiences. Sometimes a new take on an old product can totally change perception, and that’s a big part of why we’re still important. We need to be able to decide for and with our clients when to ignore the data and go against the machine.

Perhaps the biggest challenge we face with this wealth of data in social and across the digital ecosystem is the lack of true conversion attribution. Viewthroughs, click throughs, assisted conversions, lookback windows, weighted attribution, et al. They’re all incomplete, and each one leads to brand partners hoarding as much credit as possible. Sometimes, your campaign didn’t result in a sale. It may have failed miserably, or it may have contributed to some other very important metric. It takes a lot of very human trial and error to patch together a truly successful integrated campaign.

And so here we stand, at the proverbial crossroads. The deluge of data is here. Across platforms and devices, we’re going to continue to track, target, analyze, and optimize. And we should. We just need to apply a human element to it if we’re going to be successful. And we need to remember that there’s a person on both sides of the table, both creating marketing content and consuming it.