Communications specialist Michael Maslansky talks with TRUE about the current corporate obsession with big data that may get in the way of understanding what motivates customers.
TRUE: Have companies become too enthralled with big data?
Maslansky: Companies of all sizes are indeed pretty caught up in the quest for the best big data and finding the most effective ways to use it. That’s a good thing, and there’s no doubt that big data will play a substantial role in changing the relationship between companies and customers, politicians and the electorate, and a whole host of other institutions and their constituencies. But there is a danger when people begin to believe that all of the answers can be found there — because big data at the end of the day is not a panacea for everything that ails a company. Big data has the answers for many questions companies will ask, but it does not hold all the answers. These days that’s a pretty provocative statement.
TRUE: Can you provide some examples of where big data falls short?
Maslansky: Big data is very good at telling you what is happening, but it’s not as good at telling you why. And that why becomes particularly important whenever you need to adjust your strategy to produce a different behavior in your audience moving forward. For instance, big data could tell you that people aren’t buying a particular product at a given time. There could be any number of reasons why that product is not being purchased. It could be that people have moved on to a different product. It could be that stores didn’t have enough inventory of that product, so more would have been bought if the supplies had been available. It could be that something else was on special or people didn’t like the color the product came in. There are any number of reasons why you might want to know the reasons something’s happening and not just what’s happening, and big data usually isn’t as good at telling you that story.
The other big thing that is important to understand about the difference between big data and small data is that big data is really good at helping you understand how people respond to a given set of stimuli that’s out in the marketplace. So that could be the product that exists today, the marketing that exists today, the messages that exist today, the promotions that exist today. What it’s less equipped to help you understand is how people might respond to a different set of stimuli in the future. So it’s less predictive than it is diagnostic. In those cases where a product or issue is not yet in the public domain and you need to find the answer before going public, it’s tough to find big data that can help you analyze that. For example, the messages that we test are never public yet, so we really need to go out there and test a number of potential approaches that could be taken to address this issue or launch this product — which one is going to be the most effective? In those situations, I don’t know that there’s a big data way to compare the potential of various approaches other than maybe a predictive algorithm that makes a whole bunch of assumptions about how people will behave. And there you better hope you are making the correct assumptions.
Social becomes completely useless in that context because people can’t react to something that doesn’t exist yet. Without the scope of that data, you’ve got really very limited data for qualitative analysis, for understanding the why behind the what. At that point, you’ve got to go out there and talk to people in some form or another.
TRUE: What’s the complement to big data then? Focus groups?
Maslansky: Getting in front of customers and talking to them is something that these days gets dismissed or downplayed in a conversation about the highly sophisticated and scientific approach of big data, and yet it is going out there and talking to customers that often provides the greatest insights and produces our most successful strategies. So whether it’s message testing, traditional focus groups or ethnographies or social analysis, we need this kind of a combination of big data and small data. The more qualitative discussion of what’s in social media conversations is where the insights lie.
And it can be pretty scientific as well. We do highly segmented qualitative research where you’re looking for very specific target audience members as if you had profiled them using big data and you’re trying to look at their demographic and psychographic characteristics — and that’s how you recruit them, either for individual interviews or focus groups or multiperson conversations. The screening has gotten pretty sophisticated. Because there are online tools, you can do low incidence populations much more easily around the country, and then the nature of the conversation continues to evolve where qualitative research has become much less focused specifically on what people say they like and dislike, and then using other tools to assess their emotional responses to the different messages or different ideas or concepts.
TRUE: How did you get attracted to, we’ll call it small data for lack of a better word?
Maslansky: I’ve always had kind of a fascination with communications and persuasion. That’s really what we do is we study the art and science of persuasion: how companies and organizations become more effective at communicating their message in language that people will understand, believe and respond to. We say to our clients, “It’s not what you say that matters, but what your audience hears.” And the only way to know what they hear is to let them interpret the message in their own words. We started out in politics and have moved to the corporate world, but the idea was the same, that if you really think and study the kinds of approaches to communication that are most effective, you can start to distill best practices and apply them in a bunch of situations. You can understand the thinking behind decisions rather than just seeing the decisions.