Data modeling and management in Salesforce are all about defining how data is stored, structured, and related. This includes objects, fields, relationships, and the tools used for data import, export, and validation.
Standard Objects:
Custom Objects:
External Objects:
Standard Data Types:
Formula Fields:
Quantity * Unit Price.Required Fields:
Default Values:
Enforce specific business rules to ensure data accuracy.
Example:
ISPICKVAL(Status, "Open") && ISBLANK(Assigned_To__c)
Lookup Relationship:
Master-Detail Relationship:
Hierarchical Relationship:
Data Import Wizard:
Data Loader:
Explore Salesforce Objects:
Work with Fields:
Experiment with Relationships:
Use Data Tools:
Data Modeling and Management in Salesforce involve defining how data is structured, related, and managed within the platform.
Salesforce Objects are database tables that store specific data. While standard and custom objects are commonly used, Big Objects provide unique capabilities for handling large-scale data.
Big Objects are designed to handle massive data volumes (billions of records) while maintaining performance and scalability. Unlike standard and custom objects, Big Objects:
Salesforce Fields store data within objects. Two specialized field types—Roll-Up Summary Fields and Picklists—offer unique capabilities for data aggregation and structured selections.
Roll-Up Summary Fields are only available in Master-Detail relationships and allow the parent object to summarize child object data.
A company wants to track the total revenue of an Account based on related Opportunities:
Picklists restrict field values to predefined choices, ensuring data consistency.
Single Picklist:
"Pending", "Shipped", "Delivered".Multi-Select Picklist:
"Electronics" and "Home Appliances" at the same time.INCLUDES operator must be used) and do not support dependent picklists in reports.Salesforce Relationships define how objects are connected. In addition to Lookup and Master-Detail relationships, Self-Relationships and Junction Objects allow for hierarchical and many-to-many relationships.
A Self-Relationship is a Lookup relationship where an object references itself.
Junction Objects facilitate many-to-many relationships between two objects by using two Master-Detail relationships.
In an education system:
"Enrollment" is created:Salesforce provides built-in tools for data import and export, but third-party ETL tools offer additional flexibility.
ETL (Extract, Transform, Load) tools help integrate Salesforce data with external systems (e.g., databases, ERP, CRM). Popular ETL tools include:
Salesforce Data Masking protects sensitive data in non-production environments (sandboxes).
| Feature | Data Masking | Encryption |
|---|---|---|
| Purpose | Obfuscates sensitive data in sandboxes | Protects live production data |
| Use Case | Testing & development | Compliance with security standards |
| Data Retrieval | Data is irreversibly modified | Encrypted data can be decrypted with the right keys |
| Tools | Salesforce Shield, third-party ETL | Shield Platform Encryption |
The additional topics covered in this section enhance our understanding of Data Modeling and Management in Salesforce:
When should I choose master-detail instead of lookup?
Choose master-detail when the child should depend tightly on the parent for ownership, security inheritance, and roll-up summary behavior. Choose lookup when the relationship should be looser, optional, or independently owned.
This is a core modeling decision. Master-detail is best when the child record is conceptually part of the parent and should inherit the parent’s sharing and lifecycle. It also enables roll-up summary fields, which is often the deciding exam clue. Lookup is better when you need flexibility: the child may exist without the parent, needs its own owner, or the business relationship is more referential than dependent. Junction-object scenarios are also built from master-detail relationships when modeling many-to-many between two objects. A common exam trap is selecting master-detail just because the records are related. Relationship strength matters more than relationship existence. If deletion behavior, inherited access, or roll-up reporting is required, master-detail is favored; if those constraints would be too rigid, lookup is safer.
Demand Score: 86
Exam Relevance Score: 94
How should I model a many-to-many relationship in Salesforce?
Use a junction object that has two master-detail relationships, one to each parent object. That gives Salesforce a native many-to-many structure and enables parent-child reporting and roll-up options where appropriate.
Salesforce does not give you a direct checkbox called “many-to-many.” Instead, you create it by introducing a third object that represents the relationship itself. This junction object stores one record per pairing and can also hold extra attributes about that relationship, such as role, status, or effective date. The pattern is common in both exam questions and real org design because it is scalable and explicit. Community questions around AccountContactRelation and similar structures show that many-to-many often appears when one person or item must relate to multiple business records. A common mistake is trying to fake many-to-many with text fields or multiple unrelated lookups. That weakens reporting and integrity. On the exam, when the relationship itself needs data, that is a strong clue that a junction object is the right design.
Demand Score: 84
Exam Relevance Score: 91
Can I safely change a field’s data type later, or should I assume that is risky?
Assume it is risky until you verify supported conversions, dependencies, and possible data loss. Some conversions are blocked outright, and others fail because formulas, flows, code, layouts, or existing data make the change unsafe.
Field-type changes look simple but can have a large blast radius. Community posts repeatedly show that admins hit errors not because the target type is inherently invalid, but because the field is referenced elsewhere or the conversion would damage existing values. Relationship fields are especially restrictive; for example, converting a lookup into plain text is not a normal in-place conversion path. Even seemingly harmless changes can be blocked if Salesforce detects possible data loss or active dependencies. The safer pattern is to inventory references, test in a sandbox, back up data, and sometimes create a new field instead of forcing a conversion. On the exam, when a scenario mentions preserving data and minimizing risk, “create a new field and migrate values” is often stronger than “edit the existing field type.”
Demand Score: 76
Exam Relevance Score: 88
What is the biggest design consideration with external objects?
Treat external objects as data that Salesforce references, not as fully native custom objects. Their behavior depends on the external system, the adapter, and feature limitations around write support, formulas, relationships, and performance.
External objects are powerful when you need near-real-time access to off-platform data without storing everything in Salesforce. But community demand shows that builders often overestimate how “native” they behave. Official documentation emphasizes that each external object maps to an external table and that writable behavior has special considerations. Community posts ask about transactional behavior, sync issues, and cross-system constraints because DML, filtering, and formula-style expectations do not always match standard objects. On the exam, external objects are a good answer when storage duplication is undesirable and the source system remains authoritative. They are a poor answer when the requirement assumes all standard object features, full transactional consistency, or unrestricted automation support. A common mistake is choosing them only because “integration” appears in the prompt, without checking feature limits.
Demand Score: 80
Exam Relevance Score: 86