• by India White , Columnist, December 13, 2016

Brands need to have an intimate knowledge of their consumers to reach and connect with today’s consumers. Buyer personas have been a way to help marketers humanize their target customers for decades. While these archetypes may have a traditional feel to them, enhancing consumer profiles with data-driven insights brings a new life to them.

How Data Science helps you understand your target audience

Humans love to label. We have created a codified language for large groups of people based on age, gender, location, income, etc. How often do you hear the term Millennial when discussing valuable target audiences? When creating personas, we tend to turn this information into peripheral observations: “Tends toward upscale shopping habits” or “Decisions motivated by the need for adventure.” We’ve fallen into a comfort zone of demographic data transformed into gut feelings and broad generalizations about what our consumers want.

New methods of leveraging massive data sets to finding real, conclusive insights, are driving broad demographic and label-driven personas to obsolescence. As a more recent source of insights, big data is differentiated from other data sets as it requires more robust systems to query the data and identify trends.

As an example, let's explore what network data can tell us about the Moms in our audience. Think of your Twitter followers; they follow you because they find you interesting. By looking at who else they follow, you can learn more about the other things that they are interested in and passionate about. However, manually sifting through your entire audience’s following isn’t a realistic task. 

Data science and machine learning allow for algorithms to sift through the millions, if not billions, of network connections within your audience to understand what makes people similar, and what makes them different. This is what we call interest-based segmentation analysis. 

What a data science-enhanced Mom persona could look like

As an example, I ran an interest-based segmentation analysis on a prominent outdoor apparel brand. The like-minded clusters that were discovered in the audience included many outdoor activity-related interests such as Skiing Enthusiasts, Surfers, Runners, Hunting and Fishing, but also included adjacent interests like Fashionable Yogi Moms, and Sustainability and Conservancy.

Within each of these clusters, we can learn a whole lot more about their interests and what resonates with them. For instance, if we explore deeper into the Fashionable Yogi Mom cluster, we find the following:

Top Bio Keywords Include: Fashion, Yoga, Mom, Wife, Fitness, Health, Food, Music, Family, and Lover.

Top Brands Include: Patagonia, The North Face, REI, Target, Nordstrom, Amazon, Starbucks, GAP, Whole Foods, and Calvin Klein.

Trusted Media Outlets Include: Huffington Post, The New York Times, Wall Street Journal, Vogue Magazine, USA Today, Outside Magazine, Time, People Magazine, Pinterest, and CNN.

Favorite Celebrities Include: Ellen Degeneres, Oprah Winfrey, Jimmy Fallon, Anderson Cooper, Tom Hanks, Bill Gates, Jimmy Kimmel, Justin Timberlake, Ashton Kutcher, and Conan O’Brien. 

Without any demographic data, or inherent bias and assumptions, network graph analysis was able to give us a good sense of who these people are and what they care about. They are Moms who value the outdoors, fashionable gear and equipment, comedy and feel good celebrities, fitness and health, and education and news. 

While this is just a brief example, it begins to demonstrate the audience-first approach that you can take to creating personas using big data and machine learning. These insight-based personas allow brands to make data-driven decisions around content, influencer identification, ad placement, media buys, brand partnerships, etc. Knowing who your consumers are and what makes them tick allows you to make better, faster, and smarter business decisions.