Earlier this year, a study showed 54% of marketers have already invested in big data, with another 30% planning to in the next couple of years. But when Gartner released its Hype Cycle for Emerging Technologies, big data lay smack between its ‘peak of inflated expectations’ and ‘trough of disillusionment’.
We all know big data hasn’t come very far since its inception. So despite the success stories, are the majority of us just putting dollars into a problem- and hype-ridden trend after all?
How we treat big data as a part of our marketing mix will answer that. Just like big data has always been about ‘three Vs’, we see its problems as ‘three Ds’.
Problem 1: Definition
In the first place, the movement arose out of the need to make sense of a lot of data – the keyword being ‘make sense’. It’s common to look at ‘how much data I have’ and ‘how fast I can process it’, but factors like quantity and speed are still just about the data – and big data is not about the data.
Big data for marketing is no different. We don’t need more spreadsheets, we need dashboards.
Big data is really meant to give us intelligence unburdened by under-the-hood complexities. Its true opportunities lie not in access to a vast customer data ecosystem, but in enabling us to curate (and act on) what works for us in all that data. With this in mind, we can look for points and patterns that give a more accurate view of the buying journey, and add value to the sales and nurturing processes.
But with data streaming in from many different, disconnected sources, making it relevant calls for more than number-crunching.
Problem 2: Disparity
As more and more datasets spring up for us to mine, narrowing the focus will only get harder. Even the giants don’t always get it right – like Amazon, whose much-praised recommendation system is, according to Michael Fauscette, ‘personalized’ when it should be ‘individualized’.
This raises an interesting point for B2B marketers. We can finally go beyond the ‘people who did this will likely do that’ approach. Instead of profiling by trends, we can account for intangibles like brand preferences and ‘wants over needs’. One big step closer to the marketing Holy Grail.
All the evidence is pointing to algorithmic marketing and predictive analytics as the enablers – together, these can reconcile data from disparate sources and make it timely and useful. Media buys, retargeting, lead scoring – these and more would become in-the-moment. And with all this on a dashboard, responding to sudden windows of opportunity would become just a matter of how fast the folks responsible can react.
The e-commerce and banking fields have already seen cases of these yielding impressive results; it could be the only way we can make big data actionable. But with it comes another issue: defining ‘actionable’.
Problem 3: Drilling down (or the lack of it)
Big data has always been characterized by volume, velocity and variety. But, again, these are intrinsics. These are what data is – not what it can tell us. It’s like “this is dextromethorphan, ethylmorphine, and codeine” versus “this is cough syrup”.
It’s time for marketers to consider new metrics, because getting actionable intelligence means measuring things that really matter. We can capture data from all across people’s digital lives, but that means little without drilling past their actions (and the results of their actions) into their behavior, opinions, biases, and similar qualitatives. Admittedly, this is a lot of probabilities to deal with, but that’s what we’re for.
The human element is not going anywhere. We’re still the fingers on the triggers – it’s for us to look at all those far-flung datasets through marketing lenses, and isolate what we can use. Knowing what prospects and customers think, and what prospects might do while in the funnel, will help strategic decision-making and add weight and resonance to the stories we tell.
Big data for marketing still holds plenty of promise, despite its present state. It can help us understand the audience in ways we only dreamed of before. But to realize that vision, it’s important to look beyond the data to the insights, and tell our stories with that instead.