We’re living in the age of big data. That’s definitely a good thing. The more data we can capture and make actionable, the more efficient and effective digital advertising becomes. This is why we’re witnessing the rapid rise of programmatic marketing, which makes big data actionable and, in the process, allows for more precise targeting.
Almost everyone in the digital advertising industry now recognizes the importance of using data to deliver targeted messages. But if almost everyone now sees the importance, that doesn’t mean that everyone is taking full advantage of the data that’s now available.
The problem is that many of the retargeting systems currently in place weren’t designed to use all the data we can now capture. Most systems that rely on data, such as Site Retargeting, have been working well enough in the minds of many marketers, and so they haven’t bothered to move on from basic Site Retargeting to Programmatic Site Retargeting, which leverages far more data and can achieve much better results.
Other marketers recognize that basic Site Retargeting could be better and more data intensive, but simply don’t know how to move from point A to point B.
So, how do you get from Site Retargeting to Programmatic Site Retargeting, that is, from data to big data? Here’s where things get a bit counterintuitive. Working with big data requires great technology and great algorithms, but it also requires a great marketing team. In a sense, the more the technology advances, the more important humans become, both because you need a very sharp team to put programmatic systems in place and because the systems only achieve their full potential when smart marketers are making decisions that no algorithm can.
“Set it and forget it,” in other words, is not a winning strategy when it comes to working with big data. A successful programmatic campaign requires a human touch and always will.
To appreciate why the best advertising technology requires thoughtful oversight by smart people, let’s take a closer look at Site Retargeting and Programmatic Site Retargeting (PSR).
PSR is a testament to just how far we’ve come in a few short years. Unlike basic Site Retargeting, PSR doesn’t just look at the pages a user visited. Instead, PSR looks at visitor behavior from a wide range of sources, including customer profiles, shipping addresses, items in cart, and referral data, among others. All of these data points make it possible to determine not just which user to target, but also how much to pay to serve an impression to that user in a real-time bidding market, and which creative will be most effective.
Advertising may never be a perfect science, but Programmatic Site Retargeting gets us awfully close. And Search Retargeting, the practice of targeting display ads to users based on keywords they’ve searched, can be even more effective. Search Retargeting has emerged as the ideal form of programmatic marketing because, like search engine marketing, it leverages the power of intent. And, unlike Site Retargeting, Search Retargeting finds new customers who have never visited your site.
So, why do we still need humans once the number crunchers have these powerful programmatic campaigns up and running?
If you’ve ever browsed the Web and found the same display ads following you from site to site like a crazed stalker, you may already have a sense of how programmatic marketing can go wrong when algorithms are left to their own devices.
The problem isn’t necessarily that such ads are a waste of money — although if you’re seeing them constantly they probably are. The larger problem is that programmatic campaigns are not designed to look at the big picture concerns that brand marketers face. From the perspective of a direct marketer running a display campaign, cheesy creative that’s performing well might seem like a success story. The same might also be said of a great creative that’s performing well on a cheesy site. And if you only look at the numbers it’s hard to disagree.
But the brand perspective can’t be quantified in the same way. Recognizing the harm a cheesy creative can do to a brand in the long term — even as it succeeds in the short term — requires the perspective and insight of a savvy marketer.
Other times, the key human role is in determining which creatives to experiment with based on the incoming data. At Chango, for example, our data showed that one of our clients saw the most conversions from visitors who arrived via Pinterest. In response, the client developed a new creative in the style of Pinterest. The custom creative even included the ability to “pin” an item within the creative. These “pins,” in turn, gave us another important data point for retargeting.
The human touch also comes into play when looking at how different parts of a multi-faceted campaign are impacting one another. Take the case of a display campaign that doesn’t, on the surface, appear to be driving many conversions. Looked at in isolation, the campaign might seem a failure. But when marketers sit down and look at the campaign as a whole, they might begin to see things that the algorithms were never designed to see. A brand’s seemingly unsuccessful display campaign, for example, might be driving a lot of search traffic. Or it might be bringing more foot traffic into a retail store. Or it might be very effective in its own right, only the immediate impact isn’t being detected because the view-through window isn’t properly set.
The point isn’t that programmatic marketing doesn’t work. On the contrary, it has taken off in recent years precisely because it’s so effective. The point is that sometimes we have to think beyond the basic cost-based metrics that the industry obsesses over and look at the big picture metrics, such as lifetime value or brand lift. “Set it and forget it” isn’t the answer. A better, if decidedly less catchy phrase might be: “Set it, monitor it, and look at it from the perspective of both other campaigns and the long term strategy of the brand.”
So, yes, this is the age of big data. But the funny thing about the age of big data is that it’s made great marketers more crucial than ever. The age of big data, in other words, is also the age of big people.