Significant Change Notifications¶

FedRAMP MAY require providers to delay significant changes beyond the standard Significant Change Notification period and/or submit significant changes for approval in advance as a condition of a formal FedRAMP Corrective Action Plan or other agreement.

---

Terms: Significant change

Providers MUST evaluate all potential significant changes to determine the type of significant change and apply the appropriate Significant Change Notification requirements and recommendations.

1. Is it a significant change? --> Continue evaluation and follow the Significant Change Notification process.

2. If it is, is it an impact categorization change? --> This requires a new assessment and cannot be done under the Significant Change Notification process.

3. If it is not, is it a routine recurring change? --> Follow the Routine Recurring Change process (SCN-CSO-RTR).

4. If it is not, is it a transformative change? --> Follow the Transformative Change process (SCN-CSO-TRF).

5. If it is not, then it is an adaptive change --> Follow the Adaptive Change process (SCN-CSO-ADP).

---

Terms: Adaptive, Impact Categorization, Routine Recurring, Significant change, Transformative

Providers MUST maintain auditable records of significant change evaluation activities and make them available to all necessary parties.

---

Terms: All Necessary Parties, Significant change

Providers MUST include at least the following information in Significant Change Notifications:

1. Service Offering FedRAMP ID

2. Assessor Name (if applicable)

3. Related POA&M (if applicable)

4. Significant Change type and explanation of categorization

5. Short description of change

6. Reason for change

7. Summary of customer impact, including changes to services and customer configuration responsibilities

8. Plan and timeline for the change, including for the verification, assessment, and/or validation of impacted Key Security Indicators or controls

9. Copy of the business or security impact analysis

10. Name and title of approver

---

Terms: Persistent Validation, Significant change

Providers MUST keep historical Significant Change Notifications available to all necessary parties at least until the service completes its next annual assessment.

---

Terms: All Necessary Parties, Significant change

Providers MUST make ALL Significant Change Notifications and related audit records available in similar human-readable and compatible machine-readable formats.

---

Terms: Machine-Readable, Significant change

Providers MAY include additional relevant information in Significant Change Notifications.

---

Terms: Significant change

Providers MAY notify necessary parties in a variety of ways as long as the mechanism for notification is clearly documented and easily accessible.

Providers MAY execute significant changes (including transformative changes) during an emergency or incident without meeting Significant Change Notification requirements in advance ONLY if absolutely necessary. In such emergencies, providers MUST follow all relevant procedures, notify all necessary parties, retroactively provide all Significant Change Notification materials, and complete appropriate assessment after the incident.

---

Terms: All Necessary Parties, Incident, Significant change, Transformative

Providers SHOULD NOT make formal Significant Change Notifications for routine recurring changes; this type of change is exempted from the notification requirements of this process.

---

Notes:

• Activities that match the routine recurring significant change type are performed regularly and routinely by cloud service providers to address flaws or vulnerabilities, address incidents, and generally perform the typical maintenance and service delivery changes expected during day-to-day operations.
• These changes leverage mature processes and capabilities to identify, mitigate, and remediate risks as part of the change. They are often entirely automated and may occur without human intervention, even though they have an impact on security of the service.
• If the activity does not occur regularly and routinely then it cannot be a significant change of this type (e.g., replacing all physical firewalls to remediate a vulnerability is obviously not regular or routine).

Tips on ongoing operations

Key Tests:

• Routine care and feeding by staff during normal duties
• No major impact to service availability
• Does not require executive approval

Examples:

• Provisioning or deprovisioning capacity to support service elasticity
• Changing or tuning performance configurations for instances or services
• Updating and maintaining operational handling of information flows and protection across physical and logical networks (e.g., updating firewall rules)
• Generating or refreshing API or access tokens

Tips on vulnerability management

Key Tests:

• Minor, incremental patching or updates
• Significant refactoring or migration process NOT required
• No breaking changes

Examples:

• Updating security service or endpoint signatures
• Routine patching of devices, operating systems, software or libraries
• Updating and deploying code that applies normal fixes and improvements as part of a regular development cycle
• Vulnerability remediation activity that simply replaces a known-bad component(s) with a better version of the exact same thing, running in the exact same way with no changes to processes

---

Terms: Routine Recurring, Significant change

Providers MUST notify all necessary parties within 10 business days after finishing adaptive changes, also including the following information:

1. Summary of any new risks identified and/or POA&Ms resulting from the change (if applicable)

---

Notes:

• Activities that match the adaptive significant change type are a frequent and normal part of iteratively improving a service by deploying new functionality or modifying existing functionality in a way that is typically transparent to customers and does not introduce significant new security risks.
• In general, most changes that do not happen regularly will be adaptive changes. This change type deliberately covers a wide range of activities in a way that requires assessment and consideration.

Tips on adaptive changes

Key Tests:

• Requires minimal changes to security plans or procedures
• Requires some careful planning and project management to implement, but does not rise to the level of planning required for transformative changes
• Requires verification of existing functionality and secure configuration after implementation

Examples:

• Updates to operating systems, containers, virtual machines, software or libraries with known breaking changes, complex steps, or service disruption
• Deploying larger than normal incremental feature improvements in code or libraries that are the work of multiple weeks of development efforts but are not considered a major new service
• Changing cryptographic modules where the new module meets the same standards and characteristics of the former
• Replacing a like-for-like component where some security plan or procedure adjustments are required (e.g., scanning tool or managed database swap)
• Adding models to existing approved AI services without exposing federal customer data to new services

---

Terms: Adaptive, All Necessary Parties

Providers MUST allow agency customers to OPT OUT of transformative changes whenever feasible.

---

Terms: Agency, Transformative

Providers MUST notify all necessary parties of initial plans for transformative changes at least 30 business days before starting transformative changes.

---

Terms: All Necessary Parties, Transformative

Providers MUST notify all necessary parties of final plans for transformative changes at least 10 business days before starting transformative changes.

---

Terms: All Necessary Parties, Transformative

Providers MUST notify all necessary parties within 5 business days after finishing transformative changes, also including the following information:

1. Updates to all previously sent information

---

Terms: All Necessary Parties, Transformative

Providers MUST notify all necessary parties within 5 business days after completing the verification, assessment, and/or validation of transformative changes, also including the following information:

1. Updates to all previously sent information

2. Summary of any new risks identified and/or POA&Ms resulting from the change (if applicable)

3. Copy of the security assessment report (if applicable)

---

Terms: All Necessary Parties, Persistent Validation, Transformative

Providers MUST publish updated service documentation and other materials to reflect transformative changes within 30 business days after finishing transformative changes.

---

Terms: Transformative

Providers SHOULD engage a third-party assessor to review the scope and impact of the planned change before starting transformative changes if human validation is necessary; such reviews SHOULD be limited to security decisions that require human validation.

---

Note: Activities that match the transformative significant change type are rare for a cloud service offering, adjusted for the size, scale, and complexity of the service. Small cloud service offerings may go years without transformative changes, while hyperscale providers may release multiple transformative changes per year.

Tips on transformative changes

Key Tests:

• Alters the service risk profile or require new or significantly different actions to address customer responsibilities
• Requires significant new design, development and testing with discrete associated project planning, budget, marketing, etc.
• Requires extensive updates to security assessments, documentation, and how a large number of security requirements are met and validated

Examples:

• The addition, removal, or replacement of a critical third party service that handles a significant portion of information (e.g., IaaS change)
• Increasing the security categorization of a service within the offering that actively handles federal customer data (does NOT include impact change of entire offering - see impact categorization change)
• Replacement of underlying management planes or paradigm shift in workload orchestration (e.g., bare-metal servers or virtual machines to containers, migration to kubernetes)
• Datacenter migration where large amounts of federal customer data is moved across boundaries different from normal day-to-day operations
• Adding a new AI-based capability that impacts federal customer data in a different way than existing services or capabilities (such as integrating a new third-party service or training on federal customer data)

---

Terms: Cloud Service Offering, Persistent Validation, Significant change, Transformative