The Ultimate Guide to Identity Resolution

How Does Identity Resolution Work?

The specific steps involved in identity resolution can vary depending on the data that is available and the desired level of accuracy. However, the general process typically involves the following steps:

1. Collect data about the customer from different sources. This data can include things like email addresses, phone numbers, device IDs, IP addresses, and purchase history.
2. Clean and normalize the data. This means removing duplicate data, correcting errors, and standardizing the format of the data.
3. Identify potential matches between records. This can be done by comparing the data points in each record, such as email addresses, phone numbers, or IP addresses.
4. Score the matches. This means assigning a score to each potential match based on how likely it is that the two records refer to the same person.
5. Merge the matches. This means creating a single record for each customer, even if the customer has been identified using different data points.

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

Deterministic Matching

This type is when fields associated with a specific profile are consistent from one system to the next. For example, if your CRM includes a profile for “Zach” and your enterprise resource planning tool does as well, you can combine them into a single profile with 100% confidence.

Probabilistic Matching

This type is when a specific profile in one system may possibly match a different profile in that system or in other systems.

Key Terms in Identity Resolution

Identity Stitching

Identity stitching is the process of ingesting, matching and deduplicating profiles from among multiple data sources. These could include anything from web and mobile streams to customer relationship management (CRM) tools.

Identity Graph

Identity graphs are the primary output of identity resolution. They act as the underlying data structure and contain all the elements that make up a marketing profile.

This includes event-specific behaviors, attributes derived from those behaviors, second-and-third party enrichment information, models based on all of the above and identifiers that correlate with a specific profile.

Put simply, an identity graph is the backbone of a CDP’s ability to activate a personalized customer experience.

For example, say “Zach” visits a brand’s website, opens an email from the brand and then purchases one of its products. Within such a session there are events that are defined as behaviors (i.e., email open) and identity signals that correspond to those events (i.e., email, first name, last name).

These behaviors and identifiers come together as part of an identity graph that comprises Zach’s profile. This enables marketers to address Zach more effectively by tailoring experiences to be more relevant. There can be hundreds of other identifiers associated with Zach as he moves through varying systems in the customer journey.

360 Customer View

A 360 customer view — also referred to as a customer 360 or single view of customer — is often erroneously conflated with identity graphs.

Instead, it should be viewed as the visual representation of an identity graph’s underlying data structure. It’s what you’ll see when looking at a specific marketing profile.

The term “golden record” is often bandied about when talking about 360 customer views and CDPs, but it shouldn’t be. A golden record is a master, authoritative truth set that drives critical business functions.

For example, in a retail context, the golden record may include financial, product and customer records that operations teams can access. Customer 360s and CDPs don’t serve this purpose, although they often connect to the systems that do.

Identity Signals

Identity signals, also known as identity bread crumbs, are a type of attribute or piece of metadata that can be associated with a specific action or event and point back to a specific profile. Signals can be both highly determinate (i.e., hard signals), such as emails if a person signed up for a newsletter, and probable (i.e., soft signals), such as a device ID that could aid in the process of creating a persistent, addressable profile.

As this profile goes through a defined customer journey, from awareness, consideration, purchase, retention and beyond, these hard and soft identity signals are a key element in building up an identity graph. (Read our post about customer journey stages.)

What is Happening With Identity Resolution Now?

How have things changed since third-party cookie deprecation first sent shockwaves across the advertising landscape? Have you defined a new identity strategy?

Third-party Cookie Deprecation

A little more than a year after Google announced it would phase out third-party cookies from Chrome by 2023, the technology giant extended its deadline for cookie deprecation until 2024.

But Google isn’t the only game in town. Apple has been making moves toward a privacy-first future, having already removed third-party cookies from Safari. And, with its iOS 14.4 update, the company made it much harder for brands to rely on Apple’s Identifier for Advertisers.

New Innovations Like Composable Identity Resolution

So where are things moving in the future with identity resolution? The first scenario, which is already happening, is that identity will move down the stack and closer to the data. If identity is part of a Composable CDP offering, that means the rules/criteria/models that define the identity resolution process would be defined in the CDP, but the processing would happen in the Cloud Data Warehouse, with the resolved customer identity table(s) living there and being available for consumption for all downstream applications, including the CDP. ActionIQ’s Identity resolution product, as well as newer companies such as Truelty, follows this Composable architecture.

How Can Brands Use a CDP For Identity Resolution?

With the cookiepocalypse changing the way brands resolve identities, Customer Data Platforms offer a solution that uses first-party data as a foundation for growth.

With the expanding capabilities of Customer Data Platforms, brands are taking advantage of resolving identities with their CDP. Here are a few examples of brands that have supercharged their identity strategy with a CDP.

Bloomberg Slashes Acquisition Costs by 76% with a First-Party Strategy

With a CDP, Bloomberg expanded access to multiple acquisition channels, and created an audience layer with anonymous web visitors, registered users and paying subscribers in one central hub accessible to the company’s business users.