Back

 Industry News Details

 
How Can You Make Your Big Data "Smarter"? Posted on : Jan 28 - 2015

 Some data is certainly better than no data, and lots of data is better than just some data. But it's not always "the more, the merrier" when it comes to data in the digital marketing world. Instead it's the quality rather than the quantity that matters when using data to drive insightful marketing decisions.

So how can you turn the loads of data in your arsenal into "smart data" you can actually use?

"More data often means more noise and fewer signals," says Kent Moy, director of data and analytics at Dentsu Aegis media agency Vizeum. "[So] we take a consultative approach when using data."

According to Moy, when his team works with brands, they analyze and leverage a number of data sets to drive business results, including business data, targeting data, and econometrics data. "For example, econometrics modeling helps inform our planning process and foundation metrics such as engagement, cost per action (CPA), and viewability to drive optimizations," explains Moy.

With the Nike Fuel Band, for example, which collects data on consumers' activity level patterns, "analyzing such volumes of data will bring entirely new insights [into individual users] for personalization. This is the promise of 'smarter data', which will help build a better consumer experience," he adds.

Data experts at Havas and DigitasLBi also share that when they work with brands, they have access to loads of data. But most of the data available can actually have a negligible impact on business results.

"It's just noise," says Peter Sedlarcik, executive director of analytics and research at Havas Media. "One of the most important rules is to stop chasing headlines and press releases from big data providers, [because] many are exaggerated versions of reality and not all will be relevant to your business."

To separate out the important data, Havas always mines and segments proprietary data first, and then includes big data, explains Killian Schaffer, managing director and customer relationship management (CRM) lead at Havas Worldwide Discovery in Chicago.

Schaffer shares two approaches that can be used to drive business returns on big data. "First, we segment populations who can expand our reach into social media. We identify customers who follow social media call-to-actions to identify populations that have the potential for broader influence. We use those attributes to segment for acquisition and referral messages," he says.

"Another area where we've had success is using big data to define opportunities for marketing receptiveness. We've created an index to determine media attention across Twitter Trends, Digg, Reddit, and other news sources. Through the analysis, we've determined high opportunity time periods where news interest drives relevancy for our message."

Additionally, Carol Chung, senior vice president of media technology at DigitasLBi, suggests that when marketers try to glean valuable data from a big data reservoir, they need to have a holistic view of the data they are able to harness, which may include rich media data, demand side platform data, and data-management-plaftform data.

 

"Don't look at your data and technology as separate components," she says. "Instead of choosing one component at a time, you can test different types of data together. Always look at a larger picture of what you want to get out of everything, how you will be able to access data, and what you want to do with the results."  Source