Media Planning & Buying

A Marketer’s Guide to Using Third-Party Data for Programmatic Advertising

March 11, 2015

Third-party data – the term is a constant in the programmatic world and in the trades, but how many marketers can confidently attest to knowing exactly what it is and what they can do with it?

Programmatic advertisers use third-party data every day to increase campaign efficiency by identifying desired audiences and allotting ad spend accordingly. When properly utilized, third-party data can help marketers get the most out of their programmatic dollars, and play an integral role in success. However, putting budget behind a third-party data provider is not always necessary, depending on what first-party data is already available and what type of data is needed for the campaign. When considering using data to inform a campaign, marketers should first assess their campaign goals and what is needed to accomplish them, and then when appropriate, incorporate third-party data providers into their strategies. This post outlines how programmatic advertisers can use third-party data to create efficiencies by developing data-driven strategies to identify target audiences and allocate ad spend accordingly.

Three Different Types of Data
Before determining that third-party data is best for their efforts, it is important for marketers to familiarize themselves with the different types of data that exist. Knowing what type of data is being used and where it comes from will help marketers determine which data will most accurately support a campaign’s needs.

First-Party Datacomes directly from the client or a publisher and can include users’ contact information, behavior and purchase history. This data type is generally considered to be the most accurate and originates from container tags on a site, a Data Management Platform (DMP), or CRM/offline data. A perk of first-party data is that as this data already belongs to the business, there is no additional cost to marketers in obtaining it.

Third-Party Data is provided to marketers by an external data provider. This partner typically creates audience profiles informed by external data sources, and then charges a CPM for the use of the data. Third-party data can come from a variety of places, including surveys/panels, opt-in online tracking, cookie-based tracking, registration data, public records, and offline transaction data such as loyalty cards.

Second-Party Data, a less commonly discussed form of data, is a hybrid of data between two first-parties. Second-party data is the term for when one first-party shares data with another (e.g., advertiser to publisher, and vice versa). This data often consists of registration data which may be detailed enough on its own, or it may need to be compared to third-party data in order to best categorize the users.

Deciding to Use Third-Party Data
When third-party data has been discussed in the industry, the conversation has at times landed on the controversy surrounding its accuracy. This is exacerbated by the lack of an industry-wide consensus on data gathering methods. However, using third-party data can be a great asset to a campaign if marketers are smart about how they use it. Knowledge is the best defense against risk associated with using inaccurate data. Not all partners are equipped with the same capabilities and can provide the same offerings, so it is a good idea for marketers to meet with multiple partners to determine their strengths and weaknesses. Marketers should look to partner with data providers that best collect and categorize their data. They should also consider their campaign goals and use their best judgment when making the decision to use a partner as an agent of their brand.

Using Third-Party Data
Third-party data can be used in a number of different ways, though it is most often used for targeting particular audiences with ads or to serve bespoke messaging to certain segments. In order to maintain volume, it is best used for middle- to upper-funnel initiatives since the more specifically a marketer targets, the less inventory will be available. These segments can be custom designed by the advertiser, or pre-existing segments can be used.

Imagine an automobile company that wants to run more efficient campaigns – by using third-party data, marketers can ensure ads for the company are served specifically to users who have shown recent intent to purchase a car. Going one step further, marketers could target their ads for say a luxury model to users that fall into a high household income segment. Third-party data can also be useful in helping advertisers find new prospects. For example, a hotel could target its ads to users who have not yet been to its site, but who have searched for flights in the area.

Vetting & Partnering with Third-Party Data Providers
To find the best third-party data partners for a campaign, marketers should ask potential partners the following questions:

  • How do you collect data? Is the data cookie-based only, or do they also have access to users’ offline activity? Is the data extrapolated? How many times does a user have to take an action in order to fall into one of their segments? And how recently does a user have to take those actions for them to be documented and collected?
  • How “fresh” is the data? Marketers should ask how often a provider refreshes their cookie pools. Then, they should inquire how often the provider’s cookie lists would get synched to their Demand Side Platform.

Like almost everything else in media buying, rates can be negotiated – especially when based on spend levels. Third-party data usage is billed on a CPM basis, which adds incremental costs on top of the existing media costs. Something for marketers to keep in mind is that rate regulations vary from DSP to DSP. If a Boolean “or” statement is being used to define which segments to buy, and a user falls into two segments, whether the marketer will have to pay the higher or lower price between the segments depends on the DSP. Similarly, excluding segments can cost more or less depending on the platform.


Determining which segments to use – let alone which type of data and whether third-party partners should be brought in or not – can seem like a daunting task, but there are a few ways marketers can ensure they are making thoughtful evaluations and are on the path to the correct decision. Using the client’s first-party data to inform audience buying is a great first step. If there is a container tag on the client’s site that is already synched to the agency’s DSP, an Audience Composition Report (ACR) can be pulled to see which segments stand out. If not, advertisers can use existing market knowledge or client objectives to decide which consumers to target. Off-the-shelf data segments dedicated to broader topics can also serve as a starting point or as inspiration for such decisions. If using custom audience segments is of interest, marketers should work with third-party data partners to create them.

Using third-party data to buy audience segments allows marketers to extend their programmatic dollars by spending them more efficiently. By understanding the differences between the data types, and the appropriate times to bring in third-party data providers, marketers have the ability to make their campaigns exponentially more successful.

Cover photo via LinkedIn