• Cert++
  • Practice
  • Certle
  • Review
  • Tracks
  • Checklist
  • Guides
  • Upgrade
Cert++
  1. Home
  3. Platform Data Architect

Platform Data Architect

Checklist progress

0/157Learned

Platform Data Architect

Study Checklist

  • Platform Administrator
  • Platform App Builder
  • Platform Foundations
  • Platform Developer
  • Platform Administrator II
  • Agentforce Sales Consultant
  • Agentforce Service Consultant
  • Platform Data Architect
  • Platform Development Lifecycle and Deployment Architect
  • Platform Identity and Access Management Architect
  • Platform Integration Architect
  • Platform Sharing and Visibility Architect
  • Heroku Architect
  • B2C Solution Architect
  • Experience Cloud Consultant
  • Agentforce Field Service and Operations Consultant
  • Agentforce Nonprofit Consultant
  • Data 360 Consultant
  • Omnistudio Consultant
  • CRM Analytics and Einstein Discovery Consultant
  • Platform User Experience Designer
  • Platform Strategy Designer
  • B2C Commerce Developer
  • JavaScript Developer
  • Omnistudio Developer
  • Platform Developer II
  • Marketing Cloud Engagement Administrator
  • Marketing Cloud Engagement Specialist
  • Marketing Cloud Engagement Consultant
  • Agentforce Sales Foundations
  • Business Analyst
  • Marketing Cloud Engagement Developer
  • Marketing Cloud Engagement Foundations
  • Agentforce Specialist
  • Agentforce Life Sciences Consultant
  • B2B Commerce Administrator AP
  • B2B Commerce Developer AP
  • Agentforce Consumer Goods AP
  • Agentforce Financial Services AP
  • Agentforce Health AP
  • Agentforce Manufacturing AP
  • MuleSoft Integration Foundations
  • MuleSoft Developer
  • MuleSoft Developer II
  • MuleSoft Platform Integration Architect
  • MuleSoft Platform Architect
  • Tableau Desktop Foundations
  • Tableau Data Analyst
  • Tableau Consultant
  • Tableau Server Administrator
  • Tableau Architect

Checklist progress

0/157Learned

  • When designing a data model for a multi-cloud Salesforce org, what are the key considerations for choosing between lookup relationships and master-detail relationships?
  • What are the trade-offs between denormalized and normalized data models in Salesforce, particularly in relation to SOQL query complexity and governor limits?
  • What is the maximum number of master-detail relationships per object and how does it constrain data model design for complex domain models?
  • What is the maximum number of lookup relationships per object and how does it constrain data model design for complex domain models?
  • What are the limits and behaviors of roll-up summary fields in Salesforce?
  • How do roll-up summary fields influence decisions about parent-child object relationships in a high-volume environment?
  • How does the choice between formula fields and custom metadata types affect data model maintainability at scale?
  • How does the choice between formula fields and custom metadata types affect governor limits at scale?
  • What are the implications of using a junction object versus a many-to-many relationship with multiple lookup fields for cascade delete behavior?
  • What are the implications of using a junction object versus a many-to-many relationship with multiple lookup fields for the sharing model?
  • When should an architect use a polymorphic lookup (e.g., WhatId on Task) versus dedicated lookup fields to multiple objects, given cross-object query limitations?
  • When should a Person Account model be chosen over a standard Account/Contact model for a B2C business?
  • What data model trade-offs does enabling Person Accounts introduce for integrations and reporting?
  • What are the constraints on External ID fields (field type limitations, maximum per object, indexed behavior)?
  • How do External ID field constraints shape identity strategy in a data model?
  • How do compound fields (Address and Geolocation) affect SOQL query patterns, indexing capabilities, and data migration compared to individual custom fields storing the same information?
  • How does implicit record sharing through master-detail relationships differ from explicit sharing through sharing rules?
  • What are the architectural implications of master-detail versus sharing rules for sensitive child records?
  • When designing account and opportunity team models, how do Account Teams and Opportunity Teams differ in their effect on record sharing?
  • What are the data model implications of Account Teams and Opportunity Teams for high-volume team membership?
  • What are the differences between Hierarchy Custom Settings and List Custom Settings in terms of data access pattern, governor limits, and appropriate use cases in a data model?
  • How does the use of external objects and Salesforce Connect change data modeling decisions compared to persisted custom objects for integrating external system data?
  • When designing a data model that spans Sales Cloud and Service Cloud, what field-level and object-level features should be used to ensure proper cross-cloud data visibility?
  • How does Field History Tracking (limited to 20 fields per object, 18-month retention) compare to Salesforce Shield Field Audit Trail (up to 60 fields, 10-year retention) for meeting long-term audit requirements in a data model?
  • How does defining OWD as Private versus Public Read Only versus Public Read/Write affect the scalability of sharing recalculation in high-volume environments?
  • What are the differences between criteria-based sharing rules and ownership-based sharing rules?
  • How do criteria-based and ownership-based sharing rules interact with OWD settings to produce the effective access level for a record?
  • How does the Grant Access Using Hierarchies setting on an object interact with the role hierarchy, and what data model design considerations arise when it must be disabled for sensitive objects?
  • How does the depth of parent-child hierarchies in a data model affect sharing rule applicability and inherited access?
  • A company needs to restrict visibility of financial records so that sales reps see only their own accounts but managers see their entire team's. Which combination of OWD settings and sharing rules best supports this while keeping the data model simple?
  • A scenario requires that certain sensitive fields are only accessible to users in a specific role. What field-level security and object model techniques should be combined to enforce this without duplicating objects?
  • When a data model requires record-level security that cannot be achieved through standard sharing rules alone, what are the implications of using a custom sharing object versus Apex managed sharing?
  • When designing a data model for a community (Experience Cloud) with millions of external users, how should object OWD settings be configured to avoid sharing performance degradation?
  • How should a business data dictionary be maintained in relation to a Salesforce org's object and field metadata to ensure shared understanding across business and technical teams?
  • How do custom metadata types serve as technical metadata containers for configuration-driven behavior?
  • When are custom metadata types preferred over custom settings or custom objects for metadata management?
  • What approaches are available for implementing data classification (e.g., PII, sensitive, public) in Salesforce field metadata?
  • How does data classification in Salesforce field metadata support downstream governance processes?
  • When designing a taxonomy for a Salesforce data model shared across multiple business units, what techniques help enforce consistent categorization without restricting future extensibility?
  • How does data lineage tracking in a Salesforce data model help with regulatory compliance and impact analysis when object schemas evolve?
  • What are the primary use cases that justify implementing a Big Object over a custom object, specifically regarding storage capacity, query behavior, and data archiving needs?
  • What are the limitations of Big Objects compared to standard custom objects in terms of supported SOQL features, triggers, workflow, and process automation?
  • How does the index-based query model of Big Objects (requiring an exact match on the first index field) affect how an architect designs the composite index for anticipated access patterns?
  • What are the data ingestion mechanisms available for populating Big Objects (Apex, CSV, Bulk API) and what constraints apply to each?
  • Given a scenario with billions of historical interaction records that must be retained for audit purposes but rarely queried, when is a Big Object the right architectural choice versus using external archival storage?
  • What is the difference between record-locking skew (caused by DML contention on a parent record with many children) and sharing calculation skew (caused by ownership concentration)?
  • How does having more than 10,000 child records under a single parent account affect DML performance and sharing recalculation?
  • How does having many records with the same lookup value contribute to record locking contention during concurrent DML operations?
  • How does ownership skew (too many records owned by a single user or group) impact sharing rule performance, and what design patterns mitigate this?
  • How does the mitigation strategy differ for record-locking skew versus sharing calculation skew?
  • When a customer's data model results in excessive child-to-parent relationships through roll-up summary fields, what approaches mitigate lock contention while preserving aggregate visibility?
  • A large enterprise org has a single corporate account with 50,000 child opportunity records. What architectural techniques should be recommended to mitigate data skew and sharing calculation delays?
  • How does using a private OWD with many criteria-based sharing rules on a skewed object compare to using a public read/write OWD in terms of sharing recalculation performance?
  • What are the four primary MDM implementation styles (Registry, Consolidation, Coexistence, Centralized)?
  • How does a canonical data model differ from a source-specific data model, and why is it important for harmonizing data from multiple systems into a Salesforce org?
  • When choosing between a Coexistence MDM style and a Centralized Hub style for a company that needs Salesforce as its system of record while keeping existing ERP customer data authoritative for financials, what are the key trade-offs?
  • A company integrates customer data from CRM, ERP, and e-commerce. What survivorship rules and weighting strategies should be applied to determine the authoritative attribute value when sources conflict?
  • What matching and de-duplication strategies are available in Salesforce (Duplicate Rules, Matching Rules, third-party MDM) and when is each appropriate for consolidating customer master data?
  • How should hierarchy management (account hierarchies, organizational hierarchies) be modeled in Salesforce to support MDM use cases such as roll-up reporting and targeted marketing?
  • When leveraging external reference data (e.g., D&B firmographic data) for account enrichment in Salesforce, what approaches ensure the enrichment does not overwrite cleaner local data?
  • How does integrating a third-party MDM platform (e.g., Informatica MDM, Reltio) with Salesforce via API differ architecturally from implementing MDM natively within Salesforce, and when is the external MDM approach preferred?
  • What are the trade-offs between using a Registry-style MDM pattern (Salesforce as index) versus a Consolidation-style pattern (Salesforce as golden record store) for a customer with real-time data update requirements?
  • When implementing a golden record strategy in Salesforce, how should external IDs from source systems be stored to maintain traceability while the master record serves as the single system of truth?
  • How do Salesforce Duplicate Management features (matching rules, duplicate rules, duplicate jobs) support the creation and maintenance of a golden record within a single org?
  • When Salesforce merges duplicate Account records using the standard merge UI or Merge API, what happens to the child records, related files, and custom sharing grants of the non-surviving records, and how does this affect golden record maintenance?
  • A single Salesforce org receives customer updates from both a marketing automation platform and a call center CRM. How should a golden record be established and maintained when these systems have conflicting address data?
  • When consolidating contact data from three source systems, how would you define attribute-level survivorship criteria such as most recent, most complete, most trusted source, and frequency-wins?
  • What role does source system confidence scoring play in multi-source attribute consolidation, and how can this be implemented using Salesforce field values and metadata?
  • A customer has phone numbers from 5 different systems with varying data quality. What methodology determines the winning phone number attribute when survivorship rules produce ties?
  • How should source system cross-references (cross-reference IDs linking Salesforce records to originating system records) be stored and maintained to support bidirectional traceability?
  • What approaches allow business rules that depend on customer segment metadata (e.g., enterprise vs. SMB) to remain consistent across multiple systems when the customer's classification is managed in Salesforce?
  • When a customer record in Salesforce was created by merging records from multiple systems, how should the provenance metadata be preserved to support audit requirements and future re-processing?
  • What are the data access differences between Salesforce Platform licenses and full Salesforce licenses, and how does this affect which standard objects can be used in a data model?
  • How do Experience Cloud license types (Customer Community, Customer Community Plus, Partner Community) differ in their access to standard Salesforce objects, and what does this mean for data model design?
  • A company needs to give 500 external partners read access to selected account and case data. Which license type and object visibility strategy minimizes cost while meeting data access requirements?
  • When a customer wants to use Service Cloud functionality for 200 agents and Salesforce Platform licenses for 1,000 internal users who only need account and custom object access, how should the data model be designed to respect license limitations?
  • What Salesforce platform mechanisms (validation rules, Apex triggers, duplicate rules, required fields) should be layered together to enforce data consistency at the point of entry?
  • What is the role of Salesforce Duplicate Management's Block versus Alert action in enforcing data consistency at load time versus user entry time, and when should each be used?
  • How do transaction control mechanisms in Apex (savepoints, rollbacks) support data consistency when a business process writes to multiple related objects?
  • When multiple integration channels (API, manual UI, bulk load) write to the same Salesforce object, what pattern ensures that validation rules and referential integrity are enforced uniformly regardless of the data entry channel?
  • What are the risks of using the Bulk API with allOrNone=false for maintaining referential integrity when loading related objects such as accounts and their child contacts simultaneously?
  • How does Change Data Capture (CDC) in Salesforce support downstream data consistency by providing a reliable event stream of record changes, and what are its limitations compared to after-insert triggers for real-time consistency enforcement?
  • What architectural patterns (e.g., canonical account model, 360-degree contact view, unified timeline) support a single customer view when data originates from Sales Cloud, Service Cloud, and Marketing Cloud?
  • What are the trade-offs between persisting a single view of the customer as a materialized record in Salesforce versus computing it on-demand via API calls to source systems?
  • When multiple interaction systems each maintain partial customer profiles, how should external IDs be used to link records across systems and maintain a unified view in Salesforce?
  • How does Salesforce Data Cloud (formerly Customer Data Platform) differ architecturally from a custom unified profile built on standard Salesforce objects for creating a single customer view across interaction systems?
  • What are the primary patterns (hub-and-spoke, point-to-point, shared middleware) for consolidating data across multiple Salesforce orgs, and what are the trade-offs of each?
  • What role does a global external ID strategy play when consolidating records from multiple Salesforce orgs into a single master org?
  • How does Salesforce Connect with Cross-Org Adapter differ from real-time API integration for accessing data from another Salesforce org, and when is each preferred?
  • When Salesforce-to-Salesforce (S2S) connection is used to share records between partner orgs, what are its limitations compared to API-based replication in terms of supported objects, field mapping, and bidirectional sync?
  • When a company merges two Salesforce orgs and needs to consolidate Account and Contact records, what strategy prevents duplicate creation and maintains referential integrity for related objects?
  • A global enterprise operates Sales Cloud in EMEA and North America as separate orgs. A new requirement demands a unified customer view. What are the architectural trade-offs between org consolidation versus real-time federation?
  • What Salesforce features (data classification metadata, field audit trail, Shield Platform Encryption) support GDPR compliance?
  • How should data classification labels (e.g., PII, sensitive, confidential) be applied to Salesforce fields, and how does this classification integrate with data masking and access control decisions?
  • What is the difference between Salesforce Shield Platform Encryption and Classic Encryption, and when should each be recommended for protecting PII fields in a GDPR context?
  • How should an architect layer data classification, field audit trail, and Shield Platform Encryption in a GDPR-compliant data model?
  • How does Shield Platform Encryption key management work, including the tenant secret, key derivation, and what happens to data if an org's key is destroyed or rotated?
  • What are the implications of applying Shield Platform Encryption to fields that are used in SOQL WHERE clauses, reports, and list views, and how do these functional restrictions influence the decision to encrypt?
  • How does the Salesforce Privacy Center feature support GDPR compliance workflows such as data subject access requests and automated data retention and deletion?
  • A customer requires that personal data be identifiable and exportable on demand (GDPR data portability). What data model and tooling approach supports this requirement in Salesforce?
  • How should the right-to-erasure (GDPR Article 17) requirement be addressed in a Salesforce data model that has audit trail requirements and cross-object relationships referencing the data subject?
  • When a data subject invokes the right to be forgotten, what is the difference between anonymizing the data versus deleting the record in Salesforce, and which approach better preserves referential integrity and audit history?
  • When designing a GDPR-compliant consent management model in Salesforce, how should consent preferences be stored and linked to contact records to ensure they are propagated to all downstream systems?
  • How should data residency requirements (e.g., EU data must not leave EU jurisdiction) influence Salesforce org topology decisions, and what Salesforce features support geographic data residency?
  • What are the key components of an enterprise data governance program (data stewardship, data quality metrics, metadata management, data catalog)?
  • How does a data stewardship model (assigning business owners to data domains) integrate with Salesforce org governance to maintain data quality over time?
  • What are the trade-offs between a centralized data governance model (single governance board) versus a federated model (domain-level stewards) when implementing Salesforce across multiple business units?
  • How should an enterprise data governance program define and enforce data quality thresholds for Salesforce records, and what platform features support automated data quality measurement?
  • What role does metadata governance play in managing Salesforce org evolution, and how should an organization track changes to object schemas, field definitions, and automation as part of its governance program?
  • How does Salesforce Shield Event Monitoring support a data governance program by providing visibility into who accessed what data and when, and what are the limitations of Event Monitoring compared to a full SIEM integration?
  • How should a data retention schedule be enforced in Salesforce using automation (flows, Apex batch, Privacy Center retention policies) while preserving records that are subject to a legal hold?
  • What are the governance considerations specific to California Consumer Privacy Act (CCPA) compliance in Salesforce that differ from GDPR, particularly around opt-out of sale and data inventory?
  • What are the SOQL query performance implications of having 100 million records in a custom object, and how should the data model be designed to leverage standard indexes effectively?
  • What is index selectivity in Salesforce, and how do the 10% and 30% thresholds for selective and semi-selective indexes influence query plan decisions?
  • When is a custom index appropriate for a Salesforce field versus relying on the standard indexed fields, and what selectivity thresholds determine whether a custom index will be used by the query optimizer?
  • How does adding a WHERE clause filter on a non-selective field in a SOQL query on an LDV object cause full-table scans, and what design changes prevent this?
  • Why are null values not indexed by the Salesforce query optimizer, and how should a data model or query be designed to avoid unintended full-table scans caused by null-heavy fields used in filters?
  • How does the Salesforce Query Plan tool help diagnose slow SOQL queries on LDV objects, and what output from the tool indicates that a full table scan is occurring?
  • When designing an LDV data model, what is the impact of formula fields on query performance, and when should they be replaced with persisted fields populated via automation?
  • A customer has an object with 200 million records but only 10% are active. What data model and indexing strategies help ensure that active-record queries remain performant without degrading full-table access patterns?
  • What are the LDV-specific implications of using standard Salesforce reports and list views on an object with tens of millions of records, and how should an architect mitigate report timeouts?
  • How does data skew interact with LDV to compound sharing calculation performance issues, and what combined architectural approaches address both problems simultaneously?
  • What is Async SOQL and how does it differ from standard SOQL in terms of execution model, result handling, and appropriate use cases for LDV analytics workloads?
  • What is a skinny table in Salesforce, when should it be requested, and what are its limitations in terms of field types, relationship fields, and the conditions under which it improves query performance?
  • How does the division of Salesforce storage into data storage and file storage affect architectural decisions when designing an LDV data model that includes attachments or documents?
  • How should a data retention policy be implemented in Salesforce to automatically purge records that exceed a defined age, and what automation tools are available for this?
  • What are the data recovery and accessibility trade-offs of soft-delete archiving (moving to a hidden archive object) versus hard-delete purging in Salesforce, and when should each be preferred?
  • What are the trade-offs between archiving Salesforce data to Big Objects versus exporting to an external data lake (e.g., AWS S3) in terms of accessibility, cost, and compliance?
  • A customer needs to retain 7 years of closed case records for legal compliance but wants to reduce active org storage. What archiving strategy best balances retention, accessibility, and cost?
  • How do Salesforce data storage limits (file storage versus data storage) and their associated costs influence the decision between archiving in Big Objects versus using Heroku External Objects for historical data?
  • What is Salesforce Connect and how do external objects differ from custom objects in terms of data residency, query behavior, and storage consumption?
  • When should an architect recommend Salesforce Connect (virtualized data) instead of persisting data in Salesforce, and what are the key limitations such as governor limits on OData callouts and SOSL?
  • What are the supported Salesforce Connect adapter types (OData 2.0, OData 4.0, Cross-Org, Custom Apex), and when should each be used based on the data source characteristics?
  • A customer wants to display real-time inventory data from an SAP system within Salesforce without storing it in Salesforce storage. What virtualization approach should be recommended and what are its performance implications?
  • How do the query limitations of external objects (no SOSL, limited SOQL, no standard relationship queries to non-Salesforce objects) affect the use cases where virtualization is an appropriate architectural choice?

Given a customer scenario, recommend appropriate techniques and methods for ensuring high data quality at load time.

0/8

  • What pre-migration data profiling and cleansing steps should be performed before loading data into Salesforce to minimize load failures and post-migration data quality issues?
  • How should an external ID field strategy be designed for a data migration to allow upserts, avoid duplicates, and maintain traceability back to the source system?
  • What techniques ensure referential integrity when loading parent-child object hierarchies (e.g., Accounts then Contacts then Opportunities) using the Bulk API?
  • How should duplicate detection be handled during a large-scale initial data load into Salesforce when Duplicate Rules may create performance bottlenecks at load time?
  • When migrating historical data with relaxed validation requirements (e.g., missing required fields in legacy data), what Salesforce mechanisms allow bypassing validation rules during migration, and what are the governance risks?
  • What role does a data reconciliation process (record counts, field value sampling, checksum comparison) play in ensuring migration completeness and accuracy?
  • What is data masking in the context of Salesforce sandbox seeding, and why is it required when copying production data containing PII to a non-production environment for migration testing?
  • How should a delta migration strategy (loading only records changed since the last extraction) be designed using a watermark field, and what are the risks when source system timestamps are unreliable?

Compare and contrast various techniques for improving performance when migrating LDVs into Salesforce.

0/8

What are the performance trade-offs between using Bulk API 1.0 versus Bulk API 2.0 for loading large data volumes into Salesforce?

Learn this concept
Unseen

How does the choice of batch size in the Bulk API affect load performance and error granularity, and what is the recommended approach for tuning batch sizes for LDV migrations?

Learn this concept
Unseen

How does parallelizing data loads across multiple Bulk API jobs improve throughput, and what constraints (object-level locking, governor limits) limit the degree of parallelism?

Learn this concept
Unseen

When loading data with many lookup relationships, what batch ordering and ID mapping strategy minimizes failed records due to unresolved foreign keys?

Learn this concept
Unseen

How does disabling triggers, workflow rules, and process automation during a bulk data load improve performance, and what data consistency risks does this introduce?

Learn this concept
Unseen

What is the impact of enabling or disabling indexes during an LDV data load, and how should index strategy be sequenced relative to the load process for optimal performance?

Learn this concept
Unseen

What is the impact of sharing recalculation on data load performance for objects with private OWD, and what techniques (e.g., deferred sharing calculation) mitigate this during large migrations?

Learn this concept
Unseen

How should a phased migration cutover plan be structured to minimize downtime and data loss risk when migrating from a legacy system to Salesforce, including rollback strategy if the cutover fails?

Learn this concept
Unseen

Compare and contrast various techniques and considerations for exporting data from Salesforce.

0/6

  • What are the differences between Salesforce Data Export (scheduled backup), the Data Loader, and SOQL-based API exports in terms of completeness, scheduling, and supported data types?
  • How does the Bulk API Query operation differ from REST API SOQL queries when exporting large datasets, and when should each be chosen for an export strategy?
  • When exporting large volumes of data from Salesforce via SOQL, what query optimization techniques (indexed filters, selective WHERE clauses, queryMore pagination) ensure the export completes within API limits?
  • What are the considerations for exporting attachment and file (ContentDocument) data from Salesforce, especially regarding storage, binary file handling, and API limits?
  • What are the governance and security considerations when exporting data from a Salesforce org that uses Shield Platform Encryption, field-level security, and sharing rules, specifically regarding which users can export and what they can see?
  • How does exporting encrypted data (Shield Platform Encryption) from Salesforce affect downstream data usability, and what decryption considerations must be addressed in the export design?

Prepare for the Exam

Play Today's Certle
Back to track

Study Community

Ask questions and get the latest info from other Platform Data Architect studiers.

Go to Discord

How does the choice of batch size in the Bulk API affect load performance and error granularity, and what is the recommended approach for tuning batch sizes for LDV migrations?

Explainer

Learn More

Practice Question

Keep going

Next conceptHow does parallelizing data loads across multiple Bulk API jobs improve throughput, and what constraints (object-level locking, governor limits) limit the degree of parallelism?

Checklist progress

0/157 (0%)

0 of 157 concepts learned

Tip: You can filter concepts by status.

Prepare for the Exam

Play Today's Certle
Back to track

Study Community

Ask questions and get the latest info from other Platform Data Architect studiers.

Go to Discord

Explainer

Bulk API batch sizing determines how records are grouped for asynchronous processing, directly impacting throughput and the risk of hitting daily limits. Architects should start with large batches to minimize count, then reduce size only if processing times indicate a risk of timeouts.

Core information
  • A batch is a set of records sent in an HTTP POST request that the server processes independently, not necessarily in the order received.
More details and nuances
  • Avoid using smaller batches unnecessarily because this increases the total number of batches and risks hitting the 15,000 batch per 24-hour limit.