Cleaning House - Print+Promo

By Elise Hacking Carr

Painting a Picture

The world is made up of complex individuals, which can make predictive analytics tricky. In general, marketers should look at their target audience from the following three angles as outlined by Stephen H. Yu, president and chief consultant for Willow Data Strategy, Old Tappan, N.J.:

• Demographic: What do they “look” like? Think age, income, number of children, family composition, education level, religion, race, housing, etc.

• Behavioral: Also known as “transaction data,” behavioral data refers to your target audience’s purchasing habits. Questions to consider include: What did they buy? When? For how much? Where? Through what channel, and with what type of payment method? What did they click and browse? Where do they frequent?

• Attitudinal: This is a separate dimension from the previous two categories, while it can be estimated with demographic and behavioral elements to a certain degree. But not all rich people who frequently purchase golf equipment are politically conservative. At times, we all must stop and ask the audience. Survey data and social media data fall under this category.

Move over, synergy, there's a new buzzword in town: big data. Okay, so maybe the idea of "big data" isn't new, per se (its roots date back decades), but marketers have taken renewed interest in recent years. A Google search of "big data" produces thousands of results and varying definitions. So, what exactly is it?

In simple terms, big data refers to massive data sets that are analyzed to reveal patterns and trends intended to inform better business decisions. Think Amazon and its ability to recommend products based on a customer's purchasing habits.

Not everyone, however, believes bigger is better when it comes to data analytics. Stephen H. Yu, president and chief consultant for Willow Data Strategy LLC, Old Tappan, N.J., is pushing for smart data instead. "Big data in itself should never be the goal, as the size of the data doesn't mean anything for decision-makers and marketers," he said. "What they need are smart answers in bite-sizes that they can use without having to have a degree in statistics."

Take the weather, for example. "A weather forecast is based on tons of data, but reads like: '70 percent chance of snow showers Friday afternoon.' Anyone can act on such information, and marketing answers should be like that, too," he contended. "When dealing with customers and prospects in near real-time, there is no second to waste to go through thousands of data points."

The only way to achieve smart data is through smart practices. This means having a proper strategy in place, employing best practices and prioritizing time. To make sure you don't miss the mark, Print+Promo asked Yu and Summer Gould, president of Eye/Comm Inc., Santee, Calif., for five of their best tips. Here is what they had to say:

1. Outline a Proper Data Strategy