The technology landscape is changing rapidly, with businesses generating and using vast amounts of data from diverse sources. As a result, customers interact with brands across multiple channels and applications, leaving behind digital footprints that can be difficult to connect. This fragmentation of customer data poses significant challenges for organizations seeking to deliver personalized experiences and drive growth.

To navigate this complexity, companies must prioritize identity resolution — a way to locate unique individuals and businesses across disparate data sources and link their related data together.

This blog post looks under the covers of identity resolution in Salesforce Data Cloud to provide a unique understanding of how it works and why it’s important.

The rise of data fragmentation

The past decade has witnessed an explosion of data generation across practically every industry. Different technology challenges require different solutions, and that often leads to tailored data stores being used. Object stores like Amazon S3 help developers host static content for websites or store large datasets for big data analytics. Data warehouses — like Databricks and Google BigQuery — have enabled developers to analyze petabytes of data using SQL. These tools help solve common challenges in creating robust data architectures but have led to a proliferation of data silos.

In a world of personalized experiences, people and businesses may have several records spread across different data stores that contain different attributes and related information. Let’s look at what someone interacting with your brand may be experiencing by interacting across applications, geographies, and data stores.

In this example you can see that the customer, Aisha, has interacted with different technologies and each underlying data store represents the same person slightly differently.

Commerce – Aisha shopped online, browsing and ordering products, with details stored in Commerce Cloud.
Service – Complaints and inquiries are managed in Service Cloud across two different geographies
Marketing – Interactions with SMS, emails, and ad campaigns are sourced from Marketing Cloud
Invoices – Documents sent to Aisha are stored in Amazon S3
Shipping Details – Shipping information for all orders is stored in Databricks
Geospatial Analysis – Shipping tracking and geospatial analysis data is stored in Google BigQuery

So the question becomes — who is Aisha? When you need to contact Aisha, which details do you use? Can you easily view all related information to Aisha in one place? Do you even know they are the same person?

Without these answers, Aisha’s experience with your brand is likely to be a fractured, inconsistent one.

Identity resolution in Data Cloud

The goal of resolving identity in Data Cloud is to create a single, enriched customer profile that allows you to get insights and take action as a customer interacts with your brand.

For Aisha, this means being able to easily view and access all data points across any data source under a unified profile.

Data Cloud makes it easy to federate queries across data sources — like being able to see Aisha’s shopping cart from data sourced from Commerce Cloud — and take action on events — like sending a marketing follow-up when Aisha abandons her shopping cart.

How identity resolution works in Data Cloud

Identity resolution is driven by two key components: data mapping and identity resolution.

Data mapping

Your sources of data will each have a specific schema related to attributes that describe a person or business. These sources will also have unique identifiers.

Data mapping is the process of taking disparate schemas and mapping them to a common data model. This mapping process addresses key challenges with multiple sources of data by:

• Identifying common data points: Pinpointing fields that represent the same information across different systems (for example, email address, phone number, name).
• Handling naming discrepancies: Addressing differences in naming conventions in source schemas.
• Creating custom attributes: Defining additional fields to capture unique information not present in source systems.

Data Cloud provides standardized data guidelines and prebuilt data model objects (DMOs) to map into as part of the Customer 360 Data Model. If needed you can also extend the data model with custom objects. A great place to get an overview of the most common Data Cloud objects and their relationships is the Salesforce Architects Data Model Gallery.

In Data Cloud you can visually map source objects (shown on the left below) to common DMOs (shown on the right).

Using Aisha as the example and looking at Service Cloud and Amazon S3 data sources the data flowing into Data Cloud can be visualized as follows:

Each data source — representing a person — creates a data lake object (DLO) that is mapped to a common schema called Individual. Contact points for the person are mapped to Contact Point objects.

After ingestion, you can query the underlying Individual object to look at the records created. Using Aisha as the example, assuming you ingested records from Amazon S3 and Service Cloud, you can query the Individual object using Query Editor.

Here you can see two records related to Aisha — one from Amazon S3 and one from Service Cloud. Each source system refers to Aisha using a different primary key.

Let’s explore how identity resolution can help create one unified profile that developers can use to query across data sources.

Data Cloud Identity Resolution

Identity resolution creates matching and reconciliation rules to find references to the same person or business in the source data and link them together. Data mapping must be completed before creating your rulesets.

Match rules

Match rules are the algorithms that determine whether two records represent the same individual or business. Data Cloud offers a flexible framework for creating custom match rules based on various criteria:

• Exact matches: Comparing fields with identical values (for example, an exact email match).
• Fuzzy matches: Using probabilistic algorithms to identify similar records based on phonetic or semantic similarity (for example, name variations).

Reconciliation rules

Once matches are identified, the system must reconcile the data to create a unified profile. This involves:

• Resolving conflicts: Handling situations where data from different sources contradicts each other (for example, different addresses or inconsistent preferences).
• Prioritizing data: Determining which source data takes precedence in case of conflicts.

The output from identity resolution is a set of unified objects that link multiple sources for the same person or business together. In Data Cloud the identity resolution job shows a summary of the number of profiles that have been matched and consolidated.

By setting up matching rules for Aisha to match common attributes across Amazon S3 and Service Cloud you can create a unified profile in the Unified Individual object.

To see how the identity resolution job has created a unified profile you can use Query Editor to join together the necessary objects.

When creating an SQL query you can join together Individual, Unified Link Individual, and Unified Individual and use the Service Cloud Contact ID to find the Unified Profile ID.

To verify that you’ve created a unified profile for Aisha you can execute the same query but this time substituting the Amazon S3 ID for the same person. The result is the same Unified Profile ID.

You now have one unified profile that can be used to query all related data that has been ingested across your enterprise.

The benefits of identity resolution

With a unified profile, your organization can get new insights and take action as a customer interacts with your brand. Some of the benefits include:

• Improved customer understanding: Creates a 360-degree view of each customer, enabling a deeper understanding of preferences and interactions.
• Increased efficiency and productivity: Creates a global unique identifier for each individual and business to federate queries across data stores.
• Enhanced marketing effectiveness: Optimizes marketing spend by sending relevant messages to the right audience.
• Better customer service: Improves customer experiences by minimizing inconsistencies in customer data that can lead to errors, an incomplete or inaccurate customer profile, and increased operational costs.

If you want to see identity resolution in action, check out the Data Cloud Fundamentals YouTube playlist, which showcases how to use data mapping and identity resolution in Data Cloud.

Conclusion

In an era defined by data and customer-centric business models, Salesforce Data Cloud and Identity Resolution are indispensable tools for businesses seeking a competitive edge. By bridging the gap between data silos, organizations can unlock the value of having one profile for their customers.

Through data mapping and identity resolution processes, a comprehensive and accurate unified profile is created. This unified view empowers organizations to deliver personalized experiences, optimize marketing efforts, and make data-driven decisions with confidence.

Ultimately, the power to understand, engage, and retain customers lies in the ability to connect the dots across disparate data sources. Salesforce Data Cloud and identity resolution provide the foundation for building lasting customer relationships and driving business growth.

Resources

• Documentation: Unify Source Profiles
• Documentation: Data Mapping
• Trailhead: Data and Identity in Data Cloud
• Video: Identity Resolution in Data Cloud

About the author

Dave Norris is a Developer Advocate at Salesforce. He’s passionate about making technical subjects broadly accessible to a diverse audience. Dave has been with Salesforce for over a decade, has over 35 Salesforce and MuleSoft certifications, and became a Salesforce Certified Technical Architect in 2013.