Five Proven Tactics for Leveraging Big Data

By Laryssa Jardine

It's not how much data you have, it's how you use it. Marketers all know what a cliche that is, but we can also agree that the first-party data we collect has the potential to create great omnichannel experiences for your current and future customers that improve your business's bottom line.

Let's take Amazon. Every night before I hit the pillow, I ask Alexa to play the rain sleep sound. One night out of nowhere Alexa asks me if I want to hear the rain sleep sound. Creepy? Slightly — but it's so spot on.

To be real, Amazon's Alexa team probably has thousands of dedicated employees, unlimited resources, and a decade to come up with these insights. Not everyone has that luxury.

For those looking to identify the right data to drive the biggest impact, keep reading. In my experience, there are five concrete ways to make the most out of big data sets.

Make strategic choices with your data.

Data can be overwhelming. I've found that instead of looking at raw data, focus on identifying the insights that can be extracted from the data. When thinking about insights, I think about what dataset could provide that "Alexa" effect. Some of our biggest discoveries have come from studying "slices" of first-party data that indicate a specific, repeatable use case. Aim to surprise and delight your customers by weaving these insights into their journey or relationship to ultimately lead to higher conversion rates and increased revenue generation.

It's these insights that will lead us to understand how we can directly influence real-time decisions in our technology to be smarter and ultimately one step ahead of the competition. Relying on the foundation of insights versus quantity of data will ensure you are leveraging the right information to intelligently inform the right decisions and predict future patterns of behavior.

Cleanse and augment your data through third party solutions.

Know what's better than good data? More good data. You can make your data more valuable by leveraging additional sources to augment it. For example, third-party solutions can identify location based on IP address or differentiate a home phone number from a mobile one, the insights from the appended data allow us to better target and personalize a customers' experience. Appending data is important, but cleansing it is crucial. You don't want to waste your time on bad data. We know the more email rejects and bounces, the more penalized we get. Attempting to call non-working numbers will eat into your bottom line. Don't forget to clean up that data before you invest in using it.

Bring in stakeholder partners to help understand the insights.

Identify the experts within your own company that know that part of the industry inside-and-out. Become savvier about your business. Collaborate with key stakeholders to understand what drives the purchasing behavior of your clients. Why do they work with you? What are you doing well and what you can do better? Understanding the broader context of the industry will help you identify the right insights you need to extract.

Embrace the proof of concept (PoC).

You've got great data and some big ideas. But executive leadership wants to know if those ideas are technically feasible. That's where a PoC comes in. First identify the data that you think can impact conversion rate and really influence human behavior. I recommend determining small, easily achievable proof of concepts around those various theories. Then think about conveying your goals in a digestible way. Pro tip: Visuals help. If you can't show the power of your idea's impact, you will never get buy-in.

Experiment and then experiment some more.

It's important to build tooling that allows business stakeholders and internal associates to run these PoCs on their end with little to no engineering resources. Strive to empower your internal associates to test by giving them the tools to do so. Once you find a winner (or two), continue to iterate and leverage additional insights around the winning results – and ultimately figure out how to involve engineers to pull in this data real-time to your tech stack so that the results from experimentation are engrained in the consumer funnel moving forward. Remember, when building this data into your tech, you want to continue to make it as customizable as possible internally. Hardcoding is your worst enemy.


The views and opinions expressed are solely those of the contributor and do not necessarily reflect the official position of the ANA or imply endorsement from the ANA.


Laryssa Jardine is the SVP of product development at Centerfield, which delivers outcome-based digital marketing solutions and personalized omnichannel experiences for the world's leading brands. Connect with her on LinkedIn.