Test cases and evidence collection for validating information architecture readiness for Copilot.
Test Cases
Test 1: Site Inventory Completeness
- Objective: Verify the site architecture inventory accounts for all active SharePoint sites
- Steps:
- Run PowerShell Script 1 to generate the site inventory
- Compare the total count with the Active Sites count in SharePoint Admin Center
- Verify all site templates are represented
- Confirm no active sites are missing from the inventory
- Expected Result: Inventory count matches admin center count with all site types represented
- Evidence: Inventory CSV with admin center comparison
Test 2: Hub Site Alignment
- Objective: Confirm hub site structure aligns with business unit and governance boundaries
- Steps:
- Run Script 2 to generate the hub site relationship map
- Review hub-to-business unit alignment with governance committee
- Identify orphaned sites that should be hub-associated
- Verify no sites are associated with incorrect hubs
- Expected Result: Hub site structure reflects approved organizational structure
- Evidence: Hub site map with governance committee approval
Test 3: Content Type Standardization
- Objective: Verify consistent content type usage across the tenant
- Steps:
- Run Script 3 on a sample of 20+ sites
- Identify content types that are used inconsistently
- Verify FSI-required content types are deployed to relevant sites
- Check for duplicate or conflicting content type definitions
- Expected Result: Core FSI content types are consistently deployed
- Evidence: Content type usage report with standardization analysis
Test 4: Copilot Content Discovery Quality
- Objective: Validate that the information architecture supports accurate Copilot responses
- Steps:
- Identify 5 common Copilot use cases for the organization
- For each use case, query Copilot and evaluate response quality
- Assess whether responses reference appropriate content sources
- Note any cases where architecture gaps result in poor responses
- Expected Result: Copilot responses are grounded on relevant, well-organized content
- Evidence: Copilot response quality assessment with source references
- Objective: Verify AI-generated SharePoint sites, libraries, and metadata are reviewed before use in governance or Copilot grounding decisions
- Steps:
- Identify any SharePoint sites, document libraries, views, or autofill columns created with Copilot in SharePoint
- Review AI-generated site plans, library schemas, column names, and extraction prompts with the site owner or information architect
- Test a sample of documents to confirm autofill metadata values are accurate for the intended governance use
- Verify encrypted files, folders, unsupported column types, and files outside the documented processing limits are handled through manual metadata review
- Verify the tenant's Anthropic model setting and user/group scope in Microsoft 365 admin center > Copilot > Settings > View all > AI providers operating as Microsoft subprocessors
- Expected Result: AI-generated information architecture changes are owner-reviewed, corrected where needed, and documented before production reliance
- Evidence: Review sign-off, sample metadata accuracy results, and tenant setting screenshot
Evidence Collection
| Evidence Item |
Format |
Storage Location |
Retention |
| Site architecture inventory |
CSV |
Compliance evidence repository |
7 years |
| Hub site relationship map |
CSV/PDF |
Compliance evidence repository |
7 years |
| Content type analysis |
CSV |
Compliance evidence repository |
7 years |
| Architecture review findings |
PDF |
Governance document repository |
7 years |
| Copilot in SharePoint metadata review |
PDF/CSV/screenshots |
Governance document repository |
7 years |
Compliance Mapping
| Regulation |
Requirement |
How This Control Supports It |
| FINRA Rule 3110 |
Supervisory systems and WSPs |
Proper information architecture supports supervisory oversight of AI-accessible content |
| SEC Rule 17a-4 |
Records accessibility |
Structured content architecture helps meet records accessibility requirements |
| NIST CSF 2.0 |
ID.AM-5 — Data asset inventory |
Information architecture inventory supports identification and management of data assets |
| NIST AI RMF |
MAP 3.1 — AI system context |
Architecture review documents the AI content access context |
| - Back to Control 1.8: Information Architecture Review |
|
|