Executive summary
Enterprise‑scale Azure landing zones deliver a repeatable, governed foundation for multi‑subscription clouds. They enable organizations to separate the responsibilities of platform engineers from those of workload teams, while ensuring that security, compliance, and cost‑management policies are applied consistently across the entire estate. This article expands on the core concepts of platform versus workload landing zones, outlines the design‑area decisions that shape the environment, and provides a step‑by‑step implementation guide. It also examines data‑flow, security layers, common enterprise scenarios, the benefits and trade‑offs of a highly centralized model, operational concerns, and best‑practice recommendations.
Introduction
Microsoft defines an Azure landing zone as a collection of architectural patterns, governance rules, and shared services that together form a foundation for workloads in a multi‑subscription Azure environment. It is not a single Azure resource; rather, it is a logical construct that brings together Azure Policy, role‑based access control, networking, identity, and monitoring into a cohesive whole. The Cloud Adoption Framework (CAF) organises landing‑zone work into several phases. The Ready phase focuses on building the platform landing zone and shared services, while subsequent phases—Govern, Secure, and Manage—apply ongoing policies, security controls, and operational processes. By following this structured approach, organizations can move from a point‑in‑time proof of concept to a production‑grade, enterprise‑wide environment.
Technical fundamentals
A landing zone is composed of two logical layers:
- Platform landing zone – This is the centralized foundation that houses governance artefacts, identity services, networking topology, and shared platforms such as monitoring and security. It is typically instantiated once per Microsoft Entra tenant and is managed by a dedicated platform team.
- Workload landing zone – Each workload team receives its own landing zone, which is essentially a subscription or a set of subscriptions that inherit the guardrails defined in the platform zone. The workload zone contains the actual applications, data stores, compute resources, and any workload‑specific networking.
Key technical concepts that underpin the design include:
- Management‑group hierarchy – Azure Management Groups form a tree‑like structure that groups subscriptions and enables policy inheritance. A typical hierarchy separates platform resources from workload resources and includes top‑level groups such as Online, Corp (Internal), and Local under a parent Workload landing zones group.
- Identity boundary – Identity and access management is a primary security boundary. It is the foundation the other design areas sit on.
- Network topology – The published reference architectures illustrate hub-and-spoke and Azure Virtual WAN. The choice is a foundational design-area decision, not a later add-on.
- Governance automation – Azure Policy, applied through the management-group hierarchy, is how the platform audits and enforces standards. Workload subscriptions inherit those assignments. The design-area guidance also treats platform automation as the alignment of the tools and templates used to deploy the landing zone.
Architecture and components
The overall architecture can be visualised as a layered diagram. The top layer is the Entra tenant, beneath which sits the platform landing zone. Inside the platform zone are the management‑group hierarchy, centralized services, and the design‑area decisions that shape the environment. Below that, workload landing zones are grouped into Online, Corp, and Local buckets, each containing one or more subscriptions for development, testing, and production.
Entra tenant
└─ Platform landing zone
├─ Management‑group hierarchy (root → Platform → Workload zones)
├─ Centralized services only where they help many workloads (connectivity, security monitoring, identity, management)
└─ Design‑area decisions (identity, networking, billing, security, automation)
└─ Workload landing zones (Online, Corp, Local)
├─ Subscription(s) per environment (dev, test, prod)
└─ Workload resources
The table below contrasts the two landing‑zone types:
| Aspect | Platform Landing Zone | Workload Landing Zone |
|---|
| Purpose | Provide shared governance, security, and common services | Host the actual workload resources while inheriting platform guardrails |
| Owner | Platform team (CAF‑Ready, security, networking) | Product or solution team responsible for the workload |
| Centralized assets | Connectivity, identity, security monitoring, and management services that earn a cross-workload benefit | The Azure resources the workload team owns in each environment |
Detailed implementation approach
The implementation proceeds through a series of disciplined steps:
- Define design‑area decisions – Review the CAF design‑area guidance covering environment design (billing model, Entra tenant structure, identity strategy, resource organization, network topology) and compliance design (security, management, governance, automation). Capture decisions in the design‑area worksheets before any infrastructure is provisioned.
- Select accelerator or custom build – Microsoft provides accelerators that deploy a platform or workload landing zone with infrastructure as code. A custom build is the other documented path, done internally or with Microsoft or a partner. For most organizations, the accelerator is the fastest path to a deployment aligned with the recommended practices.
- Establish the management‑group hierarchy – Create the root group, then a Platform group, and under it a Workload landing zones group containing the Online, Corp, and Local sub‑groups. Apply Azure Policy at each level to enforce guardrails such as allowed locations, resource‑type whitelists, and tagging standards.
- Centralise only shared capabilities – Add connectivity, security monitoring, identity, or management services when they give a clear governance, operational, or economic benefit across workloads. Do not treat a product checklist as part of the landing-zone definition. The reference architectures show hub-and-spoke and Azure Virtual WAN as network patterns to choose from.
- Implement subscription vending – Define a repeatable request process. Initially this may be a manual ticket that triggers a script to create a new subscription, place it in the appropriate management‑group bucket, and assign baseline RBAC and policy assignments. Over time, automate this workflow with Azure DevOps pipelines or GitHub Actions.
- Onboard workload teams – Provide each workload team with their subscription(s). Teams create environment‑specific virtual networks that peer to the hub, deploy resources via their own infrastructure‑as‑code pipelines, and remain within the guardrails imposed by the platform zone.
- Validate compliance – Run Azure Policy compliance scans, review the results, and remediate any violations before workloads go live.
Architecture and data‑flow
The following diagram illustrates the flow of identity, policy, and monitoring information from the platform zone into each workload subscription.
Microsoft Entra tenant
-> platform landing zone
-> management-group policy
-> Online, Corp, or Local workload subscription
-> workload resources
Design-area decisions (identity, network, security, governance, management)
shape the platform before a workload subscription is vended
Where the organization chooses a hub-and-spoke or Virtual WAN reference topology, workload traffic is designed to use that shared network. The landing-zone pages do not prescribe a firewall product or a specific log store. Monitoring belongs to the management design area and is centralized only when it helps more than one workload.
Security and governance
Security is anchored in three interlocking layers:
- Identity boundary – The design-area guidance calls identity and access management a primary security boundary and the foundation of a compliant architecture. Decisions made for the Microsoft Entra tenant apply across the landing zone.
- Policy guardrails – Governance is the automation of audit and enforcement. Subscriptions placed under the workload management groups inherit the policies assigned on the hierarchy. The pages describe that inheritance. They do not publish a mandatory list of policy effects.
- Management baseline – The management design area is the ongoing baseline for visibility, operations compliance, and protect-and-recover. Security is the separate design area for controls that protect the environment. Both are refined as requirements change.
Governance automation, in the design-area sense, is the auditing and enforcement of policy. A subscription placed in the hierarchy inherits the policies assigned above it. That inheritance is the guardrail. It is not a separate product that has to be named beyond Azure Policy and the management-group design.
Enterprise use cases
- Regulated industry – A financial services firm creates distinct Corp and Local management groups to isolate workloads that must remain on‑premises or within a sovereign cloud region. Platform policies enforce encryption‑at‑rest, immutable storage, and exhaustive audit logging.
- Global SaaS provider – The provider uses the Online management group for public‑facing services. A hub‑and‑spoke topology connects regional spokes via Azure Virtual WAN, enabling low‑latency access while maintaining a single point of control for security services.
- AI research lab – Researchers spin up a dedicated workload landing zone for model training. When larger GPU‑enabled VMs are needed, the platform team updates the shared Azure Policy to allow the required VM SKUs, demonstrating the agility of a policy‑driven model.
Benefits and limitations
Benefits
- Consistent governance across all subscriptions reduces the risk of drift.
- Centralized security monitoring eliminates duplicate tooling and simplifies alert handling.
- Faster workload onboarding thanks to subscription vending and accelerator templates.
- Clear separation of responsibilities enables platform teams to focus on shared services while workload teams own their applications.
Limitations
- Initial design‑area decisions demand deep cross‑team collaboration; revisiting them later can be costly.
- Over‑centralisation of services may introduce latency for workloads that require local data processing.
- The management‑group hierarchy can become complex in very large enterprises, necessitating dedicated governance tooling and reporting dashboards.
Operational considerations
- Change management – Updates to platform policies propagate automatically, but workload owners must be notified in advance to avoid breaking changes in pipelines or resource configurations.
- Billing and management – Tenant, enrollment, and billing setup are an environment design area, not a report that the landing-zone pages specify. The management design area is the baseline for visibility, operations compliance, and protect-and-recover. How those capabilities are tooled is a later decision.
- Compliance over time – Security, governance, and management are refined as new applications or compliance standards appear. Policy on the platform landing zone is the mechanism Microsoft describes for pushing that change to workload landing zones.
Best practices
- Keep the platform landing zone minimal; only centralise resources that provide a clear cross‑workload benefit.
- Document all design‑area decisions in the CAF worksheets before executing any infrastructure‑as‑code.
- Automate subscription vending once the manual request path is stable. The sources describe vending as the repeatable creation and hand-off of a workload landing zone, not a named pipeline product.
- Review identity‑governance quarterly and retire unused privileged accounts to minimise attack surface.
- Leverage Azure Policy remediation tasks to automatically correct non‑compliant resources after a policy change.
Analysis
The source material clearly delineates the logical separation between platform and workload landing zones, explains the role of the management‑group hierarchy, and enumerates the design‑area decisions that guide implementation. It also maps the CAF phases—Ready, Govern, Secure, Manage—onto the lifecycle of a landing zone. However, the documentation provides only illustrative network topologies (hub‑and‑spoke, Virtual WAN) and leaves the final selection to the organization’s connectivity requirements. This flexibility is intentional, allowing each enterprise to tailor the network design to latency, bandwidth, and regulatory constraints.
What the sources do not settle
The Microsoft documentation does not prescribe how many subscriptions a workload should have. It says each environment can be one or more subscriptions, split by requirements, ownership, or subscription limits. It does not define a mandatory naming convention. It also does not publish performance figures for centralized security or network services. Those choices stay with the organization, and this article does not invent them.
Conclusion
Designing an enterprise‑scale Azure landing zone begins with a disciplined set of design‑area decisions and the creation of a platform landing zone that establishes a repeatable, governed foundation. By separating platform responsibilities from workload ownership, organizations can scale securely, enforce compliance consistently, and accelerate the delivery of new workloads. The Cloud Adoption Framework offers a clear roadmap, while the landing‑zone accelerator provides a practical path to implementation, enabling enterprises to move from experimental clouds to production‑grade, multi‑subscription environments with confidence.
References
These published CloudExchangers articles are further desk reading. They are not sources for the technical claims above.