More often than not, we hear “What the heck is identity?” Identity is the backbone of customer experience management, allowing brands to deeply understand and engage their customers and prospects more personally.

Merkle’s customer experience (CX) tech stack helps brands organize and align to goals and strategies to drive customer satisfaction and revenue. In this blog post, we’ll walk through the stack to define identity’s immense scope and show that an entanglement of capabilities is necessary to reap its full benefit. I’ll start with the identity layer, which describes the components of identity, then move on to the other layers that facilitate and activate identity across a brand’s entire CX landscape.

Before I dive in, I want to give you a simple metaphor for how identity works within the CX stack − and that is the circulatory system. Identity is the oxygenation of the blood (data) by the lungs. It gives the blood its energy. Your data management infrastructure is the heart and blood vessels, pumping blood to the rest of the body. The remaining CX stack layers are the rest of the body, using the blood to carry out different tasks. 

The key capabilities of identity

The foundation of the customer experience tech stack is identity. This is typically what we think of when we think about “identity resolution”. Components include:

• A first-party identity graph stores identifiers attached to a particular profile, including all the metadata like the history, source, and confidence level of the match. The unique individual ID is owned by the organization.

• Identity resolution is the process of matching identifiers to each other. This can be deterministic, probabilistic, or a combination of both, depending on the maturity and the use cases of the matching. It can include first-party, second-party, or third-party identifiers, and matching digital and terrestrial identifiers to an individual person. It also includes matching identifiers to a vendor’s ID, internal identity resolution and hygiene, and the “rekey” process to activate an experience.

• PII recognition conjoins more permanent terrestrial identifiers like name, address, phone number, email, etc. to a consumer-specific profile. PII recognition typically uses a reference base of publicly available data or data bought from credit bureaus that can help resolve close matches (ex: John A Smith is the same person as John Albert Smith).

• Digital recognition is the ability to identify visitors on digital channels including site, app, CTV, and some media. These digital identifiers include first-party cookies, device IDs, IP address, MAC address, universal media IDs like UID 2.0 or RampID, and others. These identifiers are more ephemeral than PII identifiers, so their matches should be examined for fidelity along with reach (and operationalized, such as attaching a confidence score). 

The data and infrastructure layer

This is the scaffolding of storage, transformation, and data transfer platforms that support identity resolution as well as the activation and measurement of the use cases of identity.

• Data Lake – This is where all the parts that make up the first-party identity graph is stored, including all of the history and metadata associated with identity resolution (such as sources of identity, the confidence level of the match, changes in primary identifier, etc.).

• Consumer Data Warehouse – This is the consumption layer for all systems using identity, where the consumer profile and all their corresponding identifiers (ie. the identity keyring) are created during identity resolution.

• Data Transformation or ETLs – These capabilities carry out identity resolution. The complex rules and strategy that determines how matches will occur is enacted here.

• Data In/Out – The data transfer via SFTP, APIs, and other methods enables incoming identity data to flow into the landing zone where it can be picked up by identity resolution processes. It delivers the resolved identities from the ID graph to activation endpoints and “midpoints” like the unified profile or measurement tools. Speed here is determined by use cases.

• Data Value Optimization and Enrichment – The identity keyring, identifier crosswalks, and/or clean rooms are used to join third-party, second-party, and zero-party attributes to first-party consumer profiles. 

• Privacy, compliance, security, and governance frameworks, platforms, and automation are also housed within this layer and are the responsibility of both the organization and identity vendors. They enable sharing of identity and customer profile data within the organization and externally; prevent improper use; delete identity data as it expires; enable subject access requests and audits; and more. 

As we continue through the tech stack, identity is not enacted as much as acted upon.

The AI and analytics layer

This layer uses the holistic consumer profiles we connected via identity earlier to generate insights and drive marketing tactics and strategy.

• Measurement & Attribution, Data Visualization, Experience Analytics, and Voice of Customer – One of identity’s main benefits is the ability to track, measure, and optimize the consumer journey across channels, time, and devices. These analytical exercises can take place within traditional analytics environments, clean rooms, and/or walled gardens.

• Data Science & Engineering and Predictive Modelling & Segmentation – Analysts use resolved identities and profiles to fuel robust predictive and segmentation models, and the confidence scores of the matches inform their conclusions and the models’ use cases. For example, a retargeting campaign based on a low-confidence match might make sense for media but not for email. Data scientists also own the methodology and the validity of the identity matching processes, including the confidence scores.

The orchestration layer

This is where identity resolution really starts to show its value.

• Real-time Customer Profile – Identity resolution allows all the various consumer-related data to be joined into a single profile for cross-channel analysis and activation. This profile also stores historical matches and ranks them in terms of relevance for activation. The customer profile allows the entire customer base to be matched to incoming digital and/or terrestrial identity signals in real time for recognition and personalization through the website and other channels including the call center, text, and email.

• Audience Management – Identity provides accurate audience sizing and broadens reach across channels, both within owned identity graphs as well as vendor-supplied (onboarders and walled gardens).

• Decision & Offer Management, Experience & Messaging Orchestration, and Loyalty Platform – These capabilities rely on the data generated by identity resolution and unified customer profiles to create tailored experiences like offers, journeys, messages, and rewards. Identity resolution also enables accurate delivery of those experiences and journeys; for example, sending an email to the most relevant email address in a consumer’s profile. 

The content and commerce layer

Here, brands can deliver personalized CX based on the results of the identity resolution process and orchestration directions. They work to deliver personalized content and experiences (such as abandoned cart and product recommendations via a first-party cookie) and optimize those experiences over time.

The experience layer

Personalization comes alive in this layer and includes authentication and the collection of identifiers.

• Physical – Real-time identity recognition and customer profile recall to physical environments is the standard for engaging with consumers (think entering your phone or email into a keypad in store and having the option to immediately use rewards). Other, less-common types of recognition are facial recognition and location-based/geofencing recognition. 

• Outbound – Identity allows for accurate personalization in the preferred channel of a consumer. It also reduces waste in outbound messaging by narrowing targeting.

• Media – This is where the addressability and extended reach from ad tech comes into play. Internal and vendor-based identity resolution meet here to deliver personalized ads and extend reach through multiple identifiers per individual consumer. To make this happen, brands need to match first-party data to the identity platform, then connect to publishers to gain even more insights from third-party data. 

• Virtual – In most virtual environments, authentication is nearly guaranteed, which makes real-time identity resolution and personalization via the unified profile table stakes. 

• Inbound – Many inbound channels require authentication, with the outlier being websites. Digital recognition tags onsite and in-app are essential for personalization in real-time. 

Identity is the foundation for accurate, measurable, and deeply personalized consumer experiences, and the capabilities of the tech stack enable this.

Where we see identity going

I’ll leave you with some predictions for the future. Obviously with the third-party cookie going away in Chrome, media activation and measurement will suffer. But I foresee a few things happening. Addressable media vendors will battle it out and we will likely see if contextual or other non-addressable alternatives are good enough for brand-level ads. Walled gardens will consolidate their power even further. And advertisers will have a blank slate to rethink how to best measure and activate to achieve their goals. 

Finally, with new federal legislation potentially on the way, improving internal identity management and consent management will become a higher priority for many organizations.

Want to learn more? Contact our team of experts here.