Follow a fictional access request from item identification through least-privilege review and safe testing. Workday controls the exact secured items and available policies in each tenant.
1. Build a security model from the item outward
Start with the report, task or process action the user needs—not a broad role label.
- Secured item: Identify whether the need concerns a task, report, data source or business-process action.
- Functional area and policy: Describe the difference between a domain policy and a business-process policy.
- Access chain: Trace user membership to a security group, the applicable policy and the intended item.
Practice output: For a fictional HR analyst who cannot view a staffing report, draw the possible access chain and list what an authorized security owner would verify before changing any policy.
2. Compare group types and scope
The same permission can have very different reach depending on group membership and constraints.
- User-based groups: Explain why a group assigned directly to a user may grant broad, unconstrained access in its permitted area.
- Role-based groups: Explore how an assignable role and organization context can constrain access to relevant workers or transactions.
- Membership review: Ask whether membership follows a job, organization or explicit assignment, and how it changes after transfer or exit.
Practice output: Compare a global HR operations analyst with a regional HR partner. Write the business reason, target population and review owner for each proposed group membership.
3. Separate data visibility from process authority
Being able to see a record is not the same as initiating, approving or correcting an event.
- Domain permissions: Distinguish a need to view a secured report from a need to modify an item in the same domain.
- Process permissions: Identify the groups that may initiate, approve, rescind or correct a particular business-process type.
- Segregation of duties: Flag any proposal giving one actor both sensitive data change and unreviewed approval power.
Practice output: Build a fictional manager-change matrix: reporter, initiator, approver and auditor. Show which action each needs, which it does not, and who approves the access request.
4. Investigate denied or excessive access
Use evidence before adding permissions, and check the opposite risk of overexposure.
- Item-first investigation: Identify the report or task and the policy securing it; an error message alone does not identify the missing group.
- Account context: Compare expected membership with the actor's actual role and organizational assignment.
- Overexposure test: Consider whether a grant exposes other populations, fields or actions beyond the original request.
Practice output: Write three hypotheses for why a team lead cannot approve one transfer. For each, list evidence, a safe test and the risk of an overly broad fix.
5. Govern policy changes and verify outcomes
A change is not complete when a setting is saved; it must be reviewed, activated and tested under the organization's process.
- Change request: Document business need, affected group, secured item, policy, population and expected behavior.
- Activation boundary: Recognize that security-policy changes can remain pending until an authorized activation step is completed.
- Acceptance and rollback: Test allowed and disallowed actors with synthetic records in an authorized non-production tenant, record evidence and define rollback ownership.
Practice output: Draft a fictional access-change record with requestor, independent reviewer, activation owner, positive and negative tests, and a rollback trigger.