Going Mobile: An Introduction to Cross-Device Targeting

As marketers have increasingly taken an audience-centric approach to marketing, cross-device targeting has become an essential part of their marketing strategy. The promise of delivering a consistent and cohesive experience to customers across channels and devices is a key part of the value proposition.

In this post, we’ll go through some of the basics of cross-device targeting: What is it? How does it work? In subsequent posts, we’ll discuss the importance of ad delivery and also cover some best practices.

An Introduction to Cross-Device Targeting

You can break cross-device targeting into two distinct functions:

1. The first function is the “cross-device” piece – the ability to match data and identify a single user across multiple devices. The desktop-to-mobile scenario is one of the more popular use cases, but cross-device can also encompass desktop-to-desktop, mobile-to-mobile, and mobile-to-desktop (not to mention non-desktop/mobile devices, like TVs).

2. The second function represents the “targeting” aspect – the ability to deliver a targeted ad to that user across their different devices.

In this post, we’ll focus on the cross-device matching piece.

Matching Device IDs: Probabilistic vs. Deterministic

Accurately matching users across different devices is challenging. For example, a user accesses her phone from a coffeeshop in Hoboken on Monday, her office computer from Manhattan on Tuesday, and her tablet while on the road from an airport in Duluth on Wednesday. How can an advertiser know these three instances all reflect a single user?

There are two models for figuring this out:

1. The first method relies on identity data like login information or device IDs. This is known as deterministic matching. For example, if a user signs onto Facebook on their computer and their phone, an advertiser can be assured that both devices belong to the same the user.
2. The second method relies on statistical modeling to piece together a single identity based on multiple non-personally identifiable data points such as cookies, device data or browser data. This is known as probabilistic matching.

While deterministic matching is more accurate than probabilistic matching, it comes at the cost of data ownership, as the major owners of that login data – Facebook and Google – prevent advertisers from using it outside of their ecosystems. In contrast, while probabilistic matching trades off on some accuracy, advertisers can leverage their proprietary data however they want.

In practice, advertisers have to use a combination of both approaches to balance scale, performance and control.

In the next post, we’ll focus on the “targeting” part of the “cross-device targeting” equation, and how ad exchanges play an essential role in the process.