Insights & Resources
Security & Compliance

Multi-Account IAM Governance on AWS: What Public-Sector Workloads Need

Public-sector AWS environments routinely span dozens of accounts. IAM governance at that scale is a discipline rather than a feature. Native AWS tooling and third-party SaaS like IAM Health Cloud each have a role; the structural answer is operational practice, not tool selection.

5 min readMay 14, 2024

Multi-Account IAM Governance on AWS

Public-sector AWS environments routinely span dozens of accounts: separate accounts per agency unit, per research project, per environment tier, per security boundary. The account structure is good architecture; it provides isolation, cost attribution, and explicit boundaries for compliance authorization. The structural challenge is governing identity and access across the resulting multi-account environment.

Native AWS tooling, third-party SaaS like IAM Health Cloud, and operational practice all have roles in solving this. The structural answer is the operational practice; the tooling supports rather than substitutes for it.

Why Multi-Account IAM Is Hard

Single-account IAM is straightforward: define roles, assign permissions, audit periodically. Multi-account IAM compounds the complexity in specific ways:

  • Cross-account role assumption. Identities in one account assume roles in other accounts to access resources. The aggregate access posture is the union of all the cross-account permission grants, which is harder to visualize than single-account permissions.
  • Federated identity at scale. When users authenticate through the institutional IdP and federate into multiple AWS accounts, the role mapping per account has to stay consistent with institutional policy. Drift in any single account creates exposure.
  • Service principal proliferation. Each account has service principals (IAM roles for AWS services, automation accounts, application identities). The aggregate count across dozens of accounts can reach hundreds or thousands. Reviewing them comprehensively requires tooling.
  • Compliance documentation across boundaries. Authorization frameworks expect documented access controls per system. When systems span accounts, the documentation has to track access across the boundary.

The institutions that operate this well have explicit IAM governance practice; the institutions that operate this poorly accumulate access drift that surfaces during audits or incidents.

What AWS Native Tooling Provides

AWS provides several native services for multi-account IAM governance:

AWS Organizations provides the account structure, hierarchical organizational units, and Service Control Policies that enforce baseline guardrails across all member accounts. Setting up Organizations correctly at adoption time is the foundation that everything else builds on.

AWS IAM Identity Center (formerly AWS SSO) provides federated identity management across all accounts in the organization. Users authenticate once through the institutional IdP and access multiple accounts based on permission set assignments. This is the structural fix for the federated-identity-at-scale problem.

Service Control Policies enforce baseline guardrails (permission boundaries, region restrictions, required encryption) at the organizational unit level. Policies cascade down to all accounts in the OU, providing baseline enforcement that does not depend on per-account configuration discipline.

AWS Access Analyzer identifies external access (resources accessible outside the organization) and unused access (permissions that have not been exercised). Both are sources of access drift; surfacing them is the first step to remediation.

AWS CloudTrail and CloudTrail Lake capture identity activity across the organization. The audit trail produced is the evidence base for compliance review and incident investigation.

For most public-sector multi-account environments, native AWS tooling configured well covers the operational requirements. Where the gap appears is typically in operational practice (someone has to use the tooling) rather than in tooling capability.

Where Third-Party Tools Like IAM Health Cloud Fit

IAM Health Cloud is a SaaS aimed at multi-account AWS IAM governance, providing audit-ready visibility into IAM users, roles, and access patterns across multiple accounts. It is one of several third-party tools in this space (others include AWS Identity Solutions partners, CIEM platforms, and broader cloud security posture management products).

Third-party tools fit when:

  • The institution wants ready-to-use compliance reporting that native AWS tooling does not produce out of the box
  • The operational team needs a higher-level abstraction over the underlying AWS API surface
  • Cross-cloud governance is a requirement (the institution operates both AWS and Azure, and wants consistent IAM visibility across both)
  • The institution wants vendor-supported tooling rather than building governance practice on native AWS APIs

Third-party tools do not fit when:

  • The institution can sustain operational practice using native tooling
  • Cost is a constraint and the tool's value does not justify the licensing cost
  • The institution has specific compliance requirements that the third-party tool does not address

The decision is operational, not technical. The native AWS tooling is sufficient for most multi-account governance; third-party tools add value when the institutional capacity to operate native tooling is the binding constraint.

What Multi-Account IAM Governance Operational Practice Looks Like

Five operational practices show up consistently in mature multi-account AWS environments.

Account structure that matches institutional organization. Organizational units align with institutional units (agency divisions, departments, schools). Account naming conventions are documented and enforced. Service Control Policies enforce baseline guardrails at the OU level.

Identity through the institutional IdP. Authentication for all administrative access flows through the campus or agency IdP. AWS IAM Identity Center provides the federation. AWS-local IAM users exist only as break-glass accounts.

Access reviews on documented cadence. Permission set assignments, cross-account role grants, and service principal credentials reviewed on a quarterly or semi-annual cadence. Access that is no longer needed is removed; access that is needed but undocumented is justified.

Access drift detection automated. AWS Access Analyzer findings reviewed on a documented cadence; unused access flagged and removed; external access reviewed against the institution's authorization model.

Compliance documentation generated from operational artifacts. The IAM posture documentation pulls from CloudTrail logs, IAM Access Analyzer findings, and AWS Config rules rather than being maintained separately. The audit cycle pulls from existing artifacts.

For institutions where this operational practice exceeds internal capacity, partner engagement under managed cloud operations provides the operational depth at a level the institution can sustain.

Frequently Asked Questions

How many AWS accounts does a typical public-sector institution operate?

Variable. A small agency or institution might run 5 to 15 accounts. A mid-size institution typically runs 30 to 100 accounts. Large federal agencies, research universities with significant cloud research, and multi-agency state government deployments can run hundreds of accounts.

Should institutions use AWS Control Tower for multi-account setup?

For new multi-account environments, AWS Control Tower provides opinionated structure that satisfies most institutional requirements out of the box. For existing multi-account environments that pre-date Control Tower, retrofitting is possible but operationally heavy; the right answer depends on the existing environment's maturity.

What is the difference between AWS IAM Identity Center and federated IAM users?

IAM Identity Center provides centralized identity management with permission sets that can be assigned across accounts. Federated IAM users use the older federation pattern with per-account configuration. Identity Center is the recommended approach for new deployments; existing federated user deployments can typically migrate.

How does multi-account IAM governance interact with FedRAMP authorization?

Each FedRAMP-authorized system has a documented authorization boundary that maps to specific accounts (or specific resources within accounts). The IAM governance has to maintain the boundary controls that the authorization documents. AWS Organizations and SCPs provide the structural enforcement; the operational practice maintains the documentation alignment.

Ready to take ownership of your platform?

Stop managing vendors. Start operating a platform.

We assess your current environment, identify operational gaps, and outline what a managed engagement looks like for your organization.

No commitment requiredResponse within 1 business dayTrusted by 100+ institutionsWe will not spam your inbox