Glossary of Terms

People-Based Terms Defined

CRM Onboarding

The process that enables a brand to import customer data into a media platform in order to leverage customer insights for marketing purposes. This process usually involves the removal of personally identifiable information (PII), onboarding the anonymized data into the platform and making these insights available to create audiences for media activation based on their own customer data.

Onboarding CRM data allows marketers to seamlessly target, execute and measure audience segments that are built from actual customer insights.

First-Party Data

This is a brand’s own proprietary data - the consumer data a brand collects via their digital or offline properties (e.g. – websites, apps, social, in-store). This data set might include consumer actions such as views, clicks, interests, surveys, subscriptions, store visits or actual purchases. First-party data is often managed as part of a brand’s Customer Relationship Management (CRM) system.

Marketers can use their first-party data as part of advanced marketing strategies typically at no additional cost. An example of this is onboarding a customer list to a programmatic media platform to power a digital retargeting campaign focused on increasing sales and/or leads while improving ROI.

Second-Party Data

This is when one entity has access to another entity’s first-party data. In other words, if a brand entered into an agreement with another brand to be able to access the other brand’s first-party data. Two parties may enter into an agreement to directly exchange data for payment or other specific terms the two parties negotiate.

An example of this is a marketer that manages a health food brand that enters into an agreement to access a health food restaurant’s first-party data. This would allow the health food brand marketer to identify consumers that dine at these niche health food restaurants so that they can now directly target these known consumers as part of their own digital marketing campaigns.

Third-Party Data

This is the data brands can access via an outside company or platform that provides data for marketing purposes. Third-party data refers to several different types of data sets. Third-party data that is people-based will be sourced from registration-based data (directly linked to a known person or household). Most providers offer aggregated data that is probabilistic and is modeled or predicted. This is usually due to the inability for a platform to have access to registration-based data along with the ability to execute to known audiences at scale.

Marketers can enrich their marketing strategies using third-party data, especially with people-based third-party data. An example of this is a brand that desires to reach specific consumers at a specific time. If an auto marketer wants to target consumers that are likely in market for a specific vehicle, the auto marketer can leverage valuable third-party auto ownership data to create highly accurate targeting campaigns to reach and measure in-market vehicle shoppers.

Deterministic User-Identification

This is the unique cross-device identifier assigned to a user based on registration data collected when a user creates or logs into an account (e.g. social media platform, digital application, newsletter sign up, email account). Platforms with large registration databases have deterministic customer identifiers at scale and are able to build out deterministic identity graphs that connect anonymized IDs with all the known devices for that respective ID. Some examples are Facebook (Facebook User ID), Google (Google Advertising ID), Apple (Apple Advertising Identifier) and Viant (Viant ID).

Marketers are able to leverage the platforms that have deterministic customer identification at scale to connect with consumers on all of their known devices either within a walled garden platform that contains the walled garden’s respective IDs or on the open web using a deterministic platform that promotes scale across multiple premium content providers.

Probabilistic User Identification

A cross-device ID comprised of multiple external data points that are fed into algorithms with the goal of producing a probable match of an existing user profile. The platform’s unique algorithms provide a percentage estimate that the current user matches an existing user profile within the platform. Examples of probabilistic device recognition data points include browser type and behavior, device type, operating system, device ID, location, etc.

Marketers might leverage probabilistic customer identification platforms when looking to achieve less quality scale at a cheaper rate as well as if frequency and accuracy are less important components of the campaign goal.

Proxy (Server)

A proxy (server) is a tool on a computer that can be used to hide a user’s information. It enables the user to view a website and receive content from this website without the website receiving information about the user. In effect, the proxy is a filter between the user’s computer and the website. There are many types of proxies all with different purposes. For example, an ‘anonymizing’ proxy has the sole purpose of anonymizing a user’s web browsing. This proxy would be used to allow a user to access sites without sharing the user’s actual device information.

It is important for marketers to understand that items such as proxies and cookies can be critical factors that skew accuracy or can derail probabilistic marketing tactics. For example, if a probabilistic platform has collected data that included several proxies in their algorithms when calculating estimates for creating targeting audiences, there are likely huge gaps of missing or inaccurate user data that will skew the results of the audience and can greatly reduce targeting effectiveness.

Programmatic Ad Buying

Programmatic media buying refers to the automation of buying digital media. Marketers use programmatic technology to buy digital ad inventory across multiple publishers more efficiently via purchase, placement, and optimization of digital advertising in real time via a bidding system. There is also less direct interaction with people reducing the back and forth of manual steps (e.g. multiple insertion orders).

Programmatic media buying capabilities have revolutionized the way digital advertising is bought and managed in recent years. Rather than placing a booking for advertising directly with a content provider, marketers can manage their activity directly through a technology platform and bid to connect with consumers in real time across multiple publishers based on selected targeting criteria.


The method of determining the geolocation (geographical location) of a consumer via their digital device to deliver customized content to that consumer based on his or her location, such as country, region/state, city, metro code/zip code, organization, IP address, latitude, longitude, time zone, country, region, city, postal/zip code, or other location indicators. Deeper data sets can also determine other parameters such as domain name, connection speed, ISP, language, company name, US DMA/MSA, NAICS codes, and home/business. Now that mobile devices are so ingrained in consumers’ daily lives, there is huge opportunity to connect in new ways via geo-based marketing strategies.

Marketers can use geo-targeting tactics to reach people in the right moment with a highly relevant message. Gain market share by targeting people visiting key places, provide special offers or meaningful content to consumers based on where they currently are, find target audiences who frequently visit certain areas and build loyalty by messaging customers that have visited a physical location in the last month. One example is when a brand utilizes geo-targeting to deliver digital coupons for a desired product available at specific retail locations to all consumers that are in proximity of those stores at that time.

Additional Resources