Data Deluge: A Layered Approach To Hyper-Targeted Marketing Campaigns

Mike Rowan, President, Founder @KPItarget

This article originally appeared on Forbes.com.

How specific do you want to get when targeting the audience for your next marketing campaign? The availability of seemingly endless data is taking targeting to the next level — if you know how to use it.

We don’t need to tell you about the benefits of personalized marketing. By now, you likely know that it not only works, but it’s often expected. It’s also not without its challenges. Personalization can be downright difficult to get right, but layered data can help you come closer than ever.

Assuming you have your messaging and creative under control, layering different categories of data can help you successfully pinpoint a precise audience. How precise? We’ve come a long way from targeting audiences like “millennial mom with a college degree.”

These days, you can get hyper-specific using data on everything from credit card transactions and purchase information to geolocation and app usage. Imagine a Venn diagram, where every category of data represents a circle. The more circles you overlay, the smaller and more specific the overlapping center will become. The center of the diagram, of course, represents the bullseye: your target audience.

Let’s dig a little deeper to see how it works by taking a look at a B2C campaign we recently launched for an antibacterial skin cleanser. In this case, we targeted three main audiences: moms, caregivers and wellness enthusiasts.

Defining The Audience

Before you even think about leveraging data, it’s crucial to spend time defining your audience. In this case, we were trying to reach health-conscious women who are fitness enthusiasts. We wanted to connect with women who are interested in personal care and health and beauty products, and who gravitate toward premium food outlets for their healthier food options. 

Be sure to dedicate appropriate effort to messaging and creative that are tailored to the audience you’re trying to reach. Without personalized messaging that resonates, reaching exactly who you’re trying to target won’t do you much good.

Here are the data sets we used to help define our audience:

• Credit card data: To target our wellness enthusiast, we first looked at credit card purchase data. We targeted women who have shopped at high-end grocery stores such as Whole Foods and the Fresh Market. Then we went deeper. We wanted to know what individuals were purchasing at these outlets so we could zero in on health, beauty and personal care purchases. This data set represented the first circle in our Venn diagram: women who have purchased health, beauty or personal care products at Whole Foods or the Fresh Market.

• App usage: Next, we looked at data for app usage. We wanted to reach women who have apps on their devices dedicated to health, wellness, fitness, healthy recipes and grocery stores. This information made up the second circle that we added to the Venn diagram.

• Geographic data: To get even more specific, we then looked at geographic data to target women who have been active visitors of fitness and yoga studios. This information was based on mobile location data for the last 13 months or less. Fitness and yoga enthusiasts made up the third circle of our diagram, and our target audience was the spot where these three circles overlapped. With this information, we were able to launch our campaign with confidence that we would reach our precise definition of a wellness enthusiast. 

Supercharging Programmatic Ads

If you’re running a programmatic marketing campaign, this is when the fun really starts with layering data. But first, you have a decision to make. How hands-on do you want to be with your campaign?

You have three options:

1. Set it and forget it. Feel pretty good about the audiences you’ve come up with based on the buckets of data that you identified? In that case, you’re done. You can let your campaign run and see what happens.
2. Optimize with machine learning. Once you identify areas of engagement that translate to conversions, you can leverage machine learning to dynamically shift your spending toward ads and properties where you’re seeing the most success. Machine learning loves data, and the more data points you have, the smarter (and more effective) your campaign becomes. You can let the machines do the work within the parameters that you’ve defined via your data buckets. For example, perhaps wellness enthusiasts who shop at Whole Foods are outperforming the Fresh Market set. Or maybe those who purchase health-related products are converting like crazy, while those who buy beauty products don’t engage. You can shift your spending according to what’s working — and what’s not.
3. Use the “man behind the curtain” approach. You can take advantage of the full programmatic suite with a hands-on approach that goes beyond the original parameters of the data you’ve collected. That means you can continuously add new data sets or audiences, along with swapping in new ads or imagery, to see what works. With this approach, your campaign continuously gets smarter and smarter — and ostensibly, ever more effective.

How To Layer Data To Achieve Hyper-Personalization

Layering data can work equally well for B2B companies, though the data sets will look a little different. But how exactly do you pull off this layered data approach? 

You can purchase data sets from second- and third-party sources, which you can layer with your first-party data. You can then manually overlay all of that data in spreadsheets. Or you can build your own platform to manage it all. You also could team up with a demand-side platform or an agency to execute the campaign for you, based on the target audience that you’ve defined. Whichever route you choose, layering data sets is a highly effective way to take full advantage of the promise of personalization.

If you’d like to chat about hyper-targeted marketing campaigns, contact me at mike@kpitarget.com.