Identity Resolution

Identity resolution is the process of creating an addressable customer profile by analyzing and resolving data across multiple touchpoints, attributes and systems.

Share This Post

Share on facebook
Share on linkedin
Share on twitter
Share on email

What is Identity (ID) Resolution? 

Identity resolution is the process of creating an addressable customer profile by analyzing and resolving data across multiple touchpoints, attributes and systems. Attributes might include email addresses, cookie identifiers, device identifiers, mailing addresses, social media handles and more. 

The attributes can have both personally identifiable information (PII) and anonymous information. The process of stitching together and resolving this data is typically done by a software system, such as a customer data platform (CDP). The stitching process uses algorithms and statistical analysis to create a persistent customer identifier that can be used across systems and campaigns. This customer identifier is then used to further enrich a customer profile as more data becomes available.

Why is Identity Resolution Needed?

Consumers interact with your brand across a variety of devices and platforms. In the morning, a consumer might see an advertisement from your brand in a mobile app on her phone. During her commute to work, she might see your banner ad while visiting a news website on her tablet. When she arrives in her office, she might open an email from your brand on her laptop.

Without identity resolution, the three touchpoints (i.e., from the same person) would appear to come from three different people. With identity resolution, those touchpoints can be stitched into a single customer profile, which gives us the opportunity to create a seamless customer experience to engage this user. By understanding that these three touch points came from the same person, we can better understand her interests and needs and make better decisions on providing her the next best offer.

In addition to consumers’ tendencies to use multiple devices and platforms, data explicitly captured about customers can be stuck inside platform-specific silos. For example, a web marketing system that identifies users by cookie ID doesn’t know that the email address captured in a marketing automation system are actually the same person. Data lives in distributed systems that don’t talk to each other. In addition, the data isn’t connected or reconciled in a way that creates a unified customer profile.

These data silos make it likely that a person will receive different offers and messaging from your brand based on the systems they’re using at the time. This can create confusion and doesn’t reflect well on your brand. A unified customer profile helps you create more personalized and relevant interactions.

What Are The Types Of Identity Resolution?

There are two types of identity resolution: deterministic matching and probabilistic matching.

With deterministic matching, customer records are matched by searching for equality across identifiers such as email, phone number, or username. This approach works best when first-party data is readily available.

With probabilistic matching, profiles are matched through an estimate of the likelihood that two identities are the same customer. The identifiers could be things like an IP address, device type, browser, or OS. Probabilistic matching can be less certain than deterministic, and marketers must decide the level of confidence necessary to determine a positive match. This method can be useful when first-party data is limited, or when reach is a priority.

Additional Benefit Of Identity Stitching

An additional benefit to identity stitching is the ability to do more accurate predictive modeling. This involves producing the “training data” necessary to identify “lookalikes” within other customer sets. With automated predictive modeling built into an enterprise-grade CDP, the model-building engine correlates hundreds of profile attributes to provide a recommended list of the most meaningful profile features. Marketers can adjust the model by adding more attributes to include, or by deleting suggested attributes. To build a reliable predictive model, you first need a large set of known customers to use as training data—which is why identity resolution is a key component.

Learn More: How Marketers Can Use Identity Resolution to Succeed In A Cookieless World

More To Explore