Finding Insights in the Space Between Traditional and Big Data Research

  by    0   1

Working with large data sets is difficult.

One of the hidden challenges when you’re digging for insight is that in larger, more open data environments the practices that accompany traditional research can be less than helpful.

Think of it this way, you create a web page dialogue with a question and three buttons, one that says yes, one that says no, and one that says maybe. Basically, it’s the smallest possible version of a SurveyMonkey form. If you get 50 people to visit the page and click one of the options, you’ve got 50 discrete pieces of data along with ratios of which they chose, and (if you’re watching closely) how quickly they chose each option.

Now imagine that people who visit the page can say anything they want at anytime, and return to do so again and again. How would you structure that? What would you do with the thousands and millions of different bits of information? It’s the kind of enormous problem that data scientists face constantly.

ArCompany CEO & Founder Hessie Jones deftly framed this sort of challenge in a recent post about trying to fit Big Data (and by extension, social data) into traditional research practices.

Her compare and contrast from above illustrates the edge that a good marketer has to maintain. To some extent the challenge reflects a middle ground between quantitative and qualitative research. But it’s more than that.

That edge requires that you know what you’re looking for going in, but you also have to be flexible enough to expand your definitions of things like your audience, and topic correlation. It’s not exactly a fishing trip, but the itinerary isn’t 100% planned out either.

For most marketers this is a new skill set, and it can be frustrating if you don’t have at least a basic level of experience with algorithms and APIs (essentially, categorically structured data that is called and delivered on). But getting started isn’t as hard as you might think. There are plenty of tools to help gather and dig out insights from a wide range of social data. It’s the mindset and practices that go with it that are critical.

Here’s a quick look at using large scale social data to gather insights in the luxury / corporate travel segment

Let’s say you’re in the jet business. You sell a high quality, well known brand both used and new in some sort of marketplace model. The industry isn’t terribly well defined when it comes to digital communications / marketing, and a lot of your sales still move at physical events and by word of mouth, but you know you need to improve your online presence, and you’d like to justify the spend to your shareholders.

Contrary to popular belief, a majority of mass affluents use social media regularly for things like customer service and to research new products.

But what are they talking about? And who are they?

One way to access this is by tracking hashtag useage on Twitter, Instagram, Facebook, and other networks.

You can easily see which keywords are being used in conjunction with your chosen hashtag, and the people using them. While we use a variety of tools at ArCompany, including some that require intermediate to advanced experience in collecting & interpreting social data, this first step is actually fairly simple.

If you click here, that’ll take you to data for #luxurytravel on SocialMention. It’s the free version so it doesn’t include an advanced/historical database, but it’ll give you an idea of how people think about high end travel. 

Using SocialMention you could quickly see what people are talking about when they use #luxurytravel, identify some influencers, and get a sense of how people think about the topic, and what they’re looking for.

Another slightly more in-depth example would be looking at correlation between hashtags:

The above illustrates how you can quickly look for the when and why behind usage of parallel hashtags.

Here’s another example:

While these tools are convenient, one word of warning is that they do lean on some pretty complex algorithms. The methodology behind them is important, and having someone on your team who understands and can question those methodologies is critical.

Of course, all of this just scratches at the surface. It can get a great deal more sophisticated, like measuring how your content is increasing audience size vs. that of your competitors, or digging into how the ideal user for your platform thinks and interacts. This is what our lead Social Strategist Susan Silver calls “Seeing the Humanity in Data” … and it’s powerful stuff.

 Featured img via Wikipedia.org / Public Domain

Joe is a product/ops guy working with the ArCompany team on content, growth, and analytics. He digs media, design, startups, data, rocanroll, anything science-y, and thinking about how to become a better human.

Leave a Reply

Your email address will not be published.