Section 4 · Auditing and Monitoring · 10% of exam
4.2 Sensitive data, sandbox controls, and security review
Exam objective
Explain how to set up sensitive data for business, legal, or compliance use cases across production and sandbox environments, and review security settings and pending access changes.
Data protection begins with classification. Identify personal, financial, health, authentication, and confidential business data; document its purpose, owner, retention rule, and approved users. Apply field-level security and record sharing to match that classification, and avoid collecting fields the process does not need. Page layouts, component visibility, and list-view filters improve usability but do not substitute for security. Reports, exports, integrations, analytics, and logs are additional paths that can reveal the same value.
Production access should be based on job needs and reviewed as roles change. Give users the minimum object, field, and record permissions needed to perform their tasks. Use permission sets for additional duties, monitor privileged permissions, and remove assignments when the duty ends. For especially sensitive actions, consider separation of duties, approval, audit evidence, and a controlled deployment process. Access reviews should check not only profiles but also permission sets, groups, roles, sharing rules, territories, teams, and dashboard execution context. Sandboxes serve development, testing, training, and staging purposes, but their data can be as sensitive as production if records are copied. Decide which sandbox type and refresh strategy are needed, then assess what data comes along and who can access the environment. Where supported, use data masking or a synthetic dataset to reduce exposure. Validate integrations, outbound email, scheduled jobs, and credentials so that testing does not contact real customers or production services. Restrict sandbox access and remove stale copies according to policy.
A refresh can replace sandbox content and may alter users, configuration, and test evidence. Coordinate refreshes with release management, communicate the impact to developers and testers, and preserve needed metadata or test data through approved processes. A partial copy or developer sandbox may be sufficient for some tests, while full-volume testing may need a different environment. Choose the environment based on the data and scale needed, not by defaulting to production-like data for every task.
Security review should include pending changes. Before deploying a permission-set or sharing change, identify impacted users, records, and dependent automation. Use a sandbox and representative personas to validate access, but confirm the sandbox does not mask production-specific settings. After deployment, verify effective access and monitor for unintended denial or exposure. If an access recalculation or asynchronous process is pending, track its completion and communicate the temporary state to affected teams.
Compliance does not come from one Salesforce checkbox. It comes from matching platform controls to organizational policy and applicable obligations, documenting the design, and operating it consistently. Salesforce features can support controls but do not by themselves determine legal compliance. Work with the organization's privacy and security owners when retention, consent, data residency, or regulated processing is involved. In study scenarios, identify the control that limits exposure and explain its scope.
Exam focus: distinguish sandbox isolation from data sanitization. A sandbox can contain copied production data. Recognize field security and sharing as security controls, and identify reports, exports, logs, integrations, and dashboards as secondary exposure paths. For a pending security change, check the deployment or recalculation state and validate the intended access before broadening permissions.
Include a masked-data or synthetic-data plan in the sandbox design before the first production refresh, not after test data has already been exposed.
When evaluating a control, document its coverage and gaps. A field hidden in one Lightning page may still appear in an export or integration; a masked sandbox may retain sensitive values in logs; a deactivated user may still own records or scheduled work. Create a data-flow inventory that includes where information is collected, transformed, copied, displayed, exported, and deleted. Assign owners for access reviews and sandbox masking, and make the evidence easy to verify. These practices help administrators turn a configuration into an operating control.
Practical scenario
A team needs realistic test records for a full-copy sandbox, but the records include private customer details. Approve the environment only after limiting sandbox access, selecting a documented masking approach, disabling or redirecting outbound integrations and email, and defining refresh and deletion controls. Test that the masked fields remain useful for the workflow without exposing original values.
Extra concept
Encryption and masking solve different problems. Encryption protects data in specified storage or transmission contexts; masking substitutes or obscures values in a copy. Neither removes the need for least-privilege access, retention controls, and review of exports or logs.
Key takeaways
- Classify sensitive data and restrict it across fields, records, exports, and logs.
- Sandbox copies can retain production data and require masking and access controls.
- Review pending changes and validate intended access after they complete.
- Official references: https://help.salesforce.com/s/articleView?id=005298969&language=en_US&type=1 https://help.salesforce.com/s/articleView?id=sf.data_sandbox.htm&type=5 https://help.salesforce.com/s/articleView?id=sf.security_overview.htm&type=5 https://help.salesforce.com/s/articleView?id=sf.admin_monitorsetup.htm&type=5
