Methods of Many Colors: Targeting Audiences Programmatically

Part 4 of the “Programmatic Matters” Series examines how to use data to paint an ideal customer picture and steer advertising choices.

The ability to reach the right audience with the right message at the right time may sound cliché, but it exemplifies the true promise of programmatic. Unlike traditional forms of advertising, programmatic advertising is delivered on an individual level. It enables marketers to find and target users based on individual characteristics, behaviors and affinities no matter where they are on the Web. This is all made possible by data – the insights that help you make the most informed targeting decisions. Without data, targeting would be like trying to find a needle in a hoarder’s house (because who really has haystacks anymore?).

In Part 3 of our "Programmatic Matters" series, we examined the different types of data available to marketers. Now it’s time to put those insights to work. When it comes to programmatic, there are several types of targeting strategies you can choose to employ. While each method varies slightly based on your desired goals, they all have one thing in common: They are focused on the customer and not the context.

If data can paint a picture of what your ideal customer looks like, and lets you spend less money to reach them on a site you know they visit, why wouldn’t you do just that?

Before the advent of programmatic, advertising was more focused on the “where” versus the “who.” Just ask Don Draper. The critically acclaimed drama Mad Men romanticized advertising’s golden age, but it seemed to be all about the creative, not the placement. Not much attention was paid to ensuring that those elaborate, alcohol-infused creative ideas actually reached the right audience. Sure, they had primitive research panels to get a better understanding of their target audience, but at the end of the day, media buys were made based on the most likely places where ad executives thought they could reach those ideal customers. Ads for cars were placed in the pages of men’s magazines, while adverts for laundry machines were typically featured in women’s publications. Hey, I do laundry too, Mr. 1960’s Ad Exec.


Strangely, things did not change much with the introduction of the Internet and the rise of display advertising. While creative went from being high-gloss, full-page spreads to clickable squares and rectangles, media buyers still used Mad Men era tactics, choosing to run their ads on websites they thought reached their desired audience. This method is still quite possible, even through programmatic advertising platforms and channels, though it is best to buy these ads directly through the publisher if you are hung up on having your brand message appear on a particular site; you can probably negotiate better placements. But that’s becoming less prevalent, thanks to the ability to measure and optimize campaigns based on user data and real-time results.

If data can paint a picture of what your ideal customer looks like, and lets you spend less money to reach them on a site you know they visit, why wouldn’t you do just that? And the more data you have, the better chances you have at honing in on the right audience. Data increases in value at an exponential rate when you are able to add additional, related data sets to the user profile. This is when targeting becomes truly powerful and what makes programmatic so valuable.

Just like data, targeting comes in many varieties. You can target users by looking at individual traits or behaviors, or you can target users by looking at external variables that may come into play when defining your ideal customer.

Here is a quick rundown of ad-targeting methods that can be executed in a programmatic environment.


One of the most popular techniques currently being used by digital advertisers, audience targeting involves buying and serving ads to a specific audience segment, whether it’s gender, household income, age group, education level, relationship status or hundreds of other specific demographic attributes. The information is typically derived from a combination of both first-party and third-party data, and can easily be selected when deciding how much to bid on available media inventory.


Where audience targeting takes into account what a person looks like, behavioral targeting focuses on what a person does. This is the process of selecting prospects based on their online activities and specific actions they’ve taken a website. Most commonly, these are measurable Web events like what pages a person has visited, what products they’ve viewed or what conversion events they’ve attempted.

Audience and behavioral targeting are essentially people-based methods of targeting, but they also come into play in other targeting approaches. Using these attributes, marketers can go even deeper to find users based on how and where they are browsing the Web.


Retargeting has earned a less-than-stellar reputation as most people associate it with those ads that follow you across the Web. You know the ones. You looked at a pair of sneakers for a few seconds and now every site you see has ads for those exact same pair of kicks. But in spite of its negative association, retargeting happens to be very effective. In fact, 90% of marketers say it’s as effective as search. With programmatic, you can evaluate and bid on individual users who have already visited your site, letting you effectively buy audiences at scale by matching behavioral cookie data with it.


Just what its name implies, geotargeting is the method of delivering highly specialized and highly targeted messages that are customized based on identity and behavioral profiles layered onto a specific geographic area, down to the ZIP code level. It also often enables targeting by IP address.


Cross-device targeting is the ability to serve targeted advertising to prospects across multiple digital devices based on an understanding of how and when they may be using a specific device. Essentially, it's another flavor of retargeting because it enables you to retarget an ad on one device, knowing buyers have seen an ad or visited your site on another device. Naturally, there are more complexities involved in this form of targeting.


Similar to the traditional method of buying advertising based on editorial relevance, contextual targeting looks at the category or keywords of a website page a customer is viewing and then serves them ads that are highly relevant to that content. The difference is marketers can bid on specific keywords and topics across the Web and have their ad served next to related articles.

Determining which targeting tactic is right for you comes down to defining what your campaign goals are and what your budget is. Each of these methods offers the opportunity to deliver messages at scale without blowing through your ad budget.

In Part 5, we will discuss some of the different methods you can use to advertise programmatically.

Image credit: "Modern Languages at Finger Lakes Community College - Costa Rica 2013" by The LEAF Project, Flickr