The other day I was talking to a friend of mine who works for an advertising agency. She told me she was struggling with a new project because, while her clients seemed to know exactly where they wanted to advertise, they had few insights into what the audiences of those channels actually thought about their product. She explained that her team was struggling to tell a rich product story because they felt like they knew nothing about the story’s characters.
It sounded like her clients had plenty of big data at their disposal but weren’t sure how to leverage it properly, a problem that similarly plagues market researchers. While such quantitative research sheds light on mass-scale behavior, researchers are often left wondering about the motivations, effects, and nuances of the numbers. In other words, they’re unable to make reasonable inferences about their consumer’s behavior because they’re lacking the qualitative story.
When we talk about big data, we’re really talking about predictive analytics.
The term ‘big data’ has become such a buzzword that its original meaning is often lost in translation.
Big data refers to massive volumes of raw information gathered by businesses from a variety of sources.
The big datasets that researchers and marketers care about consist of behavioral data mined from consumers’ devices, transactions, and digital activity; the size of these datasets is constantly growing due to the relative ease with which the data is captured. But the act of analyzing such data to extract value and inform decisions falls under the practice of predictive analytics, in which existing figures are used to make inferences about the future. In terms of marketing research, this means analyzing big behavioral data to spot trends, patterns, and habits among consumers, then leveraging the findings for targeted advertising purposes—you might even refer to this strategy as ‘predictive marketing.’
Big data is great for telling marketers where and when to advertise but does little to explain why.
Big data is unparalleled when it comes to finding out what’s happening with your target audience because of its ability to reveal digital behavior in pretty broad terms—what sites they visit, which ads they click, whose emails they read. This is all vital information, since it points out where marketing can have the most impact. But in terms of determining messaging or appeal, such data is still relatively unstructured. While big data gets a general sense of your consumers’ habits, it cannot address your focused research goals. For instance, big data will not tell you why certain copy resonates with your consumers more than others. And unlike advertisers, market researchers still need to pinpoint where consumer motivations, perceptions, and associations need to be better understood; any inferences about audiences made from big data absolutely need to be validated before being treated as fact.
Here’s where qualitative research enters the picture.
So how do you go about validating your big data-based conjectures? With intimate, actionable insights found through shopper observations and interactive engagement—that is to say, with qualitative market research. You can even think of qualitative research as ‘small data’: information gleaned from the actual people behind the devices from which big data is mined. Such qualitative engagement with your audience gives a voice to consumer needs and attitudes that would remain otherwise buried amidst huge piles of quantitative data. So in order to persuasively relate the value of your product through proven, effective outlets, small and big data must be used in tandem: if small data tells you what to say, then big data tells you where to say it.
Use the two together to conquer the (marketing) world.
The relationship between big data and small data is identical to that between quantitative and qualitative research. Big data can’t replace qualitative research due to its lack of consumer intimacy when addressing narrow marketing goals—and we all know the narrower, the better. And small data can’t replace quantitative research due to the latter’s ability to gauge the performance of marketing avenues. The two are at their most effective when they are used together to build a comprehensive view of the consumer, capturing their digital behavior while articulating the motivations behind their actions. So instead of thinking of big data as a substitute for market research, think of it as another asset in the researcher’s ever-expanding toolbox.
For an example of how one of the product marketing teams at Google successfully combines quantitative data and user feedback to optimize their product development, check out the case study below.