docs: include each run's Deliverables in the published evaluation

The general `deliverables/` ignore matched the folder inside each run's
Execution too, so the Deliverables as markdown (operating scope,
requirements, capability model, operating model, validation, guide) were
left out. They are now included for the published run only.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
2026-10-22 09:39:02 +01:00
parent 628b6ee170
commit 4c1f1662de
36 changed files with 3254 additions and 0 deletions

View File

@@ -0,0 +1,100 @@
# Capabilities
The capabilities below define what the operating organisation must be able to do to meet the operational requirements. Note that any cost figures derived from these capabilities will currently use placeholders, as the rate card remains in draft.
| Capability | Items Covered | Operational Requirements Served | Why it is Needed |
| --- | --- | --- | --- |
| **Incident management** | S-1, S-2, S-3, S-4 | OR-1, OR-2 | To acknowledge and restore service when faults occur across platform components at any hour. |
| **Monitoring and alerting** | S-1, S-2, S-3, S-4 | OR-1 | To provide continuous 24x7 visibility into the state of the landing zone and trigger incident response. |
| **Problem management** | S-1, S-2, S-3, S-4 | OR-2, OR-3 | To analyze root causes of priority 1 incidents and prevent recurrences to maintain 99.9% availability. |
| **Service request fulfilment** | S-1, S-2, S-3, S-4 | OR-8 | To process routine operational tasks and access requests in a shared model. |
| **Change enablement** | S-1, S-2, S-3, S-4 | OR-4, OR-5 | To manage changes and ensure the customer platform team explicitly approves all network modifications. |
| **Release and deployment** | S-3 | OR-4 | To operate the deployment pipeline so that approved changes reach production within a single day. |
| **Configuration management** | S-1, S-2, S-3, S-4 | OR-4, OR-5, OR-7 | To maintain infrastructural consistency and manage the configuration schedule for encryption key rotation. |
| **Backup and recovery** | S-2, S-4 | OR-2, OR-10 | To enforce 7-year audit log retention and ensure recoverability of critical configurations. |
| **Service continuity** | S-1, S-2, S-3, S-4 | OR-3 | To prepare for and test major disruptions and maintain monthly uptime targets. |
| **Capacity and performance management** | S-1 | OR-3 | To track hub network and interconnect throughput, avoiding degradation due to load limits. |
| **Access management** | S-1, S-2, S-3, S-4 | OR-8 | To control who can modify platform items within the shared operational model. |
| **Vulnerability and patch management** | S-1, S-2, S-3, S-4 | OR-3 | To proactively scan for weaknesses and apply security patches to the platform. |
| **Security monitoring and response** | S-1, S-2, S-3, S-4 | OR-10 | To watch for misuse and tampering 24x7, specifically protecting audit log integrity. |
| **Cost management** | S-1, S-2, S-3, S-4 | OR-6 | To compile Google Cloud landing zone run costs for the monthly operational report. |
| **Service level management** | S-1, S-2, S-3, S-4 | OR-6 | To track SLA achievements and compile the monthly service level report. |
| **Platform product management** | S-1, S-2, S-3, S-4 | OR-8 | To manage the foundation as a product, prioritizing features by what slows down development teams. |
| **Self-service and golden paths** | S-1, S-3 | OR-4, OR-8 | To offer automated, friction-free defaults for pipeline and network modifications. |
| **Enablement** | S-1, S-2, S-3, S-4 | OR-8 | To assist and pair with developers adopting the platform in a shared service model. |
| **Delivery and developer experience measurement** | S-3 | OR-4 | To measure delivery speed, lead time, and friction to ensure changes deploy quickly and safely. |
| **Supplier management** | S-1, S-2, S-3, S-4 | OR-8, OR-9 | To monitor external service providers and guarantee operational delivery only from the UK, Poland, and India. |
| **Regulatory evidence and reporting** | S-2 | OR-10 | To retain and produce logs as evidence specified by EU DORA. |
# Sources
The capabilities chosen above draw on the following industry frameworks. *Please note that these references are unverified.* The operating model does not claim or certify compliance with any framework.
* **ITIL 4** (PeopleCert)
* Incident management practice
* Monitoring and event management practice
* Problem management practice
* Service request management practice
* Change enablement practice
* Release management and Deployment management practices
* Service configuration management practice
* Service continuity management practice
* Capacity and performance management practice
* Service level management practice
* Supplier management practice
* **Site Reliability Engineering: How Google Runs Production Systems** (Google / O'Reilly)
* Managing Incidents
* Monitoring Distributed Systems
* Postmortem Culture: Learning from Failure
* Release Engineering
* Service Level Objectives
* **ISO/IEC 20000-1**
* Incident management
* Service request management
* Change management
* Service continuity management
* Capacity management
* Service level management
* Supplier management
* **Regulation (EU) 2022/2554, the Digital Operational Resilience Act** (European Union)
* Article 9, Protection and prevention
* Article 10, Detection
* Article 11, Response and recovery
* Article 12, Backup policies and procedures, restoration and recovery procedures and methods
* Article 17, ICT-related incident management process
* Article 19, Reporting of major ICT-related incidents
* Article 28, General principles on ICT third-party risk
* **NIST Cybersecurity Framework 2.0**
* Detect: Continuous Monitoring
* Identify: Asset Management
* Recover: Incident Recovery Plan Execution
* Protect: Identity Management, Authentication, and Access Control
* Protect: Platform Security
* Detect: Adverse Event Analysis; Respond: Incident Management
* **ISO/IEC 27001**
* Access control
* Management of technical vulnerabilities
* Information security incident management
* Compliance with legal and contractual requirements
* **DevOps Research and Assessment (DORA) software delivery metrics**
* Change failure rate
* Deployment frequency and lead time for changes
* Time to restore service
* **FinOps Framework** (FinOps Foundation)
* Allocation; Budgeting; Forecasting; Anomaly Management
* **Team Topologies: Organizing Business and Technology Teams for Fast Flow** (IT Revolution)
* Platform team; Thinnest viable platform
* X-as-a-Service interaction mode
* Enabling team; Facilitating interaction mode
* **CNCF Platforms White Paper (platform as a product)** (Cloud Native Computing Foundation)
* Platform as a product; Measuring platform success
* Platform capabilities; Self-service
* **The SPACE of Developer Productivity**
* Satisfaction and well-being; Performance; Activity; Communication and collaboration; Efficiency and flow
* **DevEx: What Actually Drives Productivity**
* Feedback loops; Cognitive load; Flow state
# Assumptions
- **Service Management Tooling**: We assume the use of native Google Cloud operations tooling without an external ITSM integration, as no separate ITSM system was mandated.
- **Change Process**: We assume that standard infrastructure changes proceed via an automated deployment pipeline with peer reviews, while only network changes demand explicit manual approval from the customer platform team.

View File

@@ -0,0 +1,229 @@
# Operating model guide: Google Cloud Landing Zone
## Summary
This guide sets out the target operating model for the Google Cloud landing zone platform foundation, which comprises the hub network and interconnect, the audit log store, the deployment pipeline, and encryption keys. The platform is operated under a shared service model between the customer and an external service provider. The service provider is responsible for incident response, continuous monitoring, and day-to-day administrative maintenance during 24x7 support hours, while the customer platform team operates during business hours, retaining architectural leadership, platform enablement, and final approval authority over all network changes.
Because all operational checks and scenario walk-throughs pass with no open findings and run costs remain within budget, the reader should look first at the assumptions and open questions, specifically the exclusion of hosted application workloads from platform foundation operations and the confirmation of the go-live date.
### Key figures
- Operated: 4 Items, with 3 Conditions to keep true.
- Faults responded to: 24x7 hours.
- Capabilities: 21.
- People: 6.8. Cost a month: 46,252 GBP, from rate card version `draft-2026-10-13` (placeholder prices: an illustration).
- Scenario walk-throughs: 13 of 13 pass.
- Business outcomes: 3, 3 shown by a Measure.
- Findings for whoever owns the design: 0.
## Business outcomes
**BO-1: New payment products reach customers in weeks rather than months.** Target: A change reaches production within a day of approval..
| Measure | Layer | Framework | Target | Measured by | Answerable |
| --- | --- | --- | --- | --- | --- |
| Lead time for changes | outcome | dora-metrics | Within a day of approval | Delivery and developer experience measurement | Customer Platform Lead |
**BO-2: Payments are never unavailable for long.** Target: Restored within 4 hours..
| Measure | Layer | Framework | Target | Measured by | Answerable |
| --- | --- | --- | --- | --- | --- |
| Time to restore service | outcome | dora-metrics | Restored within 4 hours | Delivery and developer experience measurement | Customer Platform Lead |
**BO-3: We keep the platform engineers we hire.**
| Measure | Layer | Framework | Target | Measured by | Answerable |
| --- | --- | --- | --- | --- | --- |
| Developer satisfaction | guardrail | space-framework | | Delivery and developer experience measurement | Customer Platform Lead |
| Developer cognitive load | leading | devex-framework | | Delivery and developer experience measurement | Customer Platform Lead |
## What is operated
| Item | What it is | Kind | Product | From |
| --- | --- | --- | --- | --- |
| S-1 | Hub network and interconnect | network | Cloud Interconnect | |
| S-2 | Audit log store | storage | Cloud Storage | |
| S-3 | Deployment pipeline | pipeline | Cloud Build | |
| S-4 | Encryption keys | keys | Cloud KMS | |
What must be kept true while it is run:
| Condition | On | What must hold | From |
| --- | --- | --- | --- |
| K-1 | S-1 | The platform is available at least 99.9% of each month. | |
| K-2 | S-2 | Audit logs cannot be deleted for seven years. | |
| K-3 | S-4 | Encryption keys are rotated every 90 days. | |
## Operational requirements
| Requirement | Kind | What must be achieved | From |
| --- | --- | --- | --- |
| OR-1 | service_level | A priority 1 fault on any Item is acknowledged within 15 minutes, at any hour. | service_levels.priority_1_response_minutes, operations.support_hours |
| OR-2 | service_level | Service is restored within 4 hours following a priority 1 incident. | service_levels.recovery_time_hours, BO-2 |
| OR-3 | service_level | The hub network and interconnect maintains at least 99.9% availability during each month. | K-1 |
| OR-4 | service_level | An approved change reaches production within one day of approval. | BO-1 |
| OR-5 | process | Every change to the network requires approval from the customer platform team prior to implementation. | REQ-01 |
| OR-6 | process | A report detailing service levels and operational costs is delivered to the customer monthly. | REQ-02 |
| OR-7 | process | Encryption keys are rotated at least once every 90 days. | K-3 |
| OR-8 | organisation | Platform operations is conducted under a shared service model between the customer and an external service provider. | sourcing.model |
| OR-9 | organisation | Operational delivery is conducted exclusively from the United Kingdom, Poland, and India. | sourcing.delivery_countries |
| OR-10 | compliance | Audit logs cannot be deleted for seven years. | K-2 |
## Capabilities
| Capability | What it is | Serves | Covers | Hours | Effort a month |
| --- | --- | --- | --- | --- | --- |
| C-1 | Incident management | OR-1, OR-2 | S-1, S-2, S-3, S-4 | 24x7 | 40 h |
| C-2 | Monitoring and alerting | OR-1 | S-1, S-2, S-3, S-4 | 24x7 | 30 h |
| C-3 | Problem management | OR-2, OR-3 | S-1, S-2, S-3, S-4 | business | 15 h |
| C-4 | Service request fulfilment | OR-8 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-5 | Change enablement | OR-4, OR-5 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-6 | Release and deployment | OR-4 | S-3 | business | 15 h |
| C-7 | Configuration management | OR-4, OR-5, OR-7 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-8 | Backup and recovery | OR-2, OR-10 | S-2, S-4 | business | 10 h |
| C-9 | Service continuity | OR-3 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-10 | Capacity and performance management | OR-3 | S-1 | business | 5 h |
| C-11 | Access management | OR-8 | S-1, S-2, S-3, S-4 | business | 15 h |
| C-12 | Vulnerability and patch management | OR-3 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-13 | Security monitoring and response | OR-10 | S-1, S-2, S-3, S-4 | 24x7 | 40 h |
| C-14 | Cost management | OR-6 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-15 | Service level management | OR-6 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-16 | Platform product management | OR-8 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-17 | Self-service and golden paths | OR-4, OR-8 | S-1, S-3 | business | 25 h |
| C-18 | Enablement | OR-8 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-19 | Delivery and developer experience measurement | OR-4 | S-3 | business | 15 h |
| C-20 | Supplier management | OR-8, OR-9 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-21 | Regulatory evidence and reporting | OR-10 | S-2 | business | 10 h |
## Who does what
| Role | Party | Country | Level | Hours |
| --- | --- | --- | --- | --- |
| Customer Platform Lead | Customer (customer) | GB | lead | business |
| Customer Platform Engineer | Customer (customer) | GB | senior | business |
| Provider Delivery Lead | Service Provider (provider) | GB | lead | business |
| Provider 24x7 SRE | Service Provider (provider) | PL | senior | 24x7 |
| Provider Platform Engineer | Service Provider (provider) | IN | engineer | business |
The teams, as Team Topologies names their kinds:
| Team | Kind | Party | Roles | Purpose |
| --- | --- | --- | --- | --- |
| Platform team | platform | Customer | Customer Platform Lead, Customer Platform Engineer | Owns the platform roadmap, designs the networking, and provides explicit approvals for structural changes. |
| Provider Operations | platform | Service Provider | Provider Delivery Lead, Provider 24x7 SRE, Provider Platform Engineer | Provides 24x7 incident response, security monitoring, and general infrastructure operational support. |
How the teams work together:
| From | To | Mode | Until | Purpose |
| --- | --- | --- | --- | --- |
| Platform team | Provider Operations | x-as-a-service | | The customer platform team consumes infrastructure monitoring and incident response from the provider operations team as a service. |
| Capability | Customer Platform Lead | Customer Platform Engineer | Provider Delivery Lead | Provider 24x7 SRE | Provider Platform Engineer |
| --- | --- | --- | --- | --- | --- |
| Incident management (C-1) | C | | A | R | |
| Monitoring and alerting (C-2) | C | | A | R | |
| Problem management (C-3) | C | | A | | R |
| Service request fulfilment (C-4) | C | | A | | R |
| Change enablement (C-5) | A | R | | | R |
| Release and deployment (C-6) | A | R | | | |
| Configuration management (C-7) | A | | | | R |
| Backup and recovery (C-8) | C | | A | | R |
| Service continuity (C-9) | C | | A | | R |
| Capacity and performance management (C-10) | A | | | | R |
| Access management (C-11) | C | | A | | R |
| Vulnerability and patch management (C-12) | C | | A | | R |
| Security monitoring and response (C-13) | C | | A | R | |
| Cost management (C-14) | A | | R | | |
| Service level management (C-15) | C | | A/R | | |
| Platform product management (C-16) | A/R | | | | |
| Self-service and golden paths (C-17) | A | R | | | |
| Enablement (C-18) | A | R | | | |
| Delivery and developer experience measurement (C-19) | A | R | | | |
| Supplier management (C-20) | A/R | | | | |
| Regulatory evidence and reporting (C-21) | C | | A | | R |
A accountable, R responsible, C consulted, I informed.
## Staffing and cost
| Role | Party | Country | Level | Hours | Effort a month | People | Cost a month |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Customer Platform Lead | Customer | GB | lead | business | 30.0 h | 0.2 | not priced |
| Customer Platform Engineer | Customer | GB | senior | business | 85.0 h | 0.6 | not priced |
| Provider Delivery Lead | Service Provider | GB | lead | business | 20.0 h | 0.1 | 2,212 GBP |
| Provider 24x7 SRE | Service Provider | PL | senior | 24x7 | 110.0 h | 5.0 | 41,640 GBP |
| Provider Platform Engineer | Service Provider | IN | engineer | business | 135.0 h | 0.9 | 2,400 GBP |
6.8 people in all. Cost a month: 46,252 GBP.
The monthly budget is 60,000 GBP: this is within it.
Worked from rate card `rate-card`, version `draft-2026-10-13`. **Its figures are placeholders: no price has been set, and this cost is an illustration.**
Converted from USD at 0.79, a rate set on 2026-10-13.
The customer's own staff are counted and not priced. A Role that covers hours wider than a working day is never smaller than the team that can cover them.
## Validation
| Check | What must hold | Verdict | Evidence |
| --- | --- | --- | --- |
| OC-1 | The hub network and interconnect maintains at least 99.9% availability during each month. | pass | Capability C-1 is responsible for incident management with 24x7 coverage by the Provider 24x7 SRE role (R-4), which is sufficient to maintain highly available operations by responding to faults around the clock. |
| OC-2 | Audit logs are retained for seven years and cannot be deleted during this period. | pass | Capability C-8 manages backup and recovery, validating retention policies for audit logs. It is appropriately staffed during business hours. |
| OC-3 | Encryption keys are rotated at least once every 90 days. | pass | Capability C-7 covers configuration management including tracking key rotation cycles, with the Customer Platform Lead (R-1) accountable and Provider Platform Engineer (R-5) responsible. |
| OC-4 | A priority 1 fault on any Item is acknowledged within 15 minutes, at any hour. | pass | Capability C-1 covers incident management 24x7 with the Provider 24x7 SRE (R-4) responsible, ensuring 15-minute acknowledgement at any hour. |
| OC-5 | Service is restored within 4 hours following a priority 1 incident. | pass | Capability C-1 provides 24x7 incident response, ensuring that service is restored within the 4-hour target following a priority 1 incident. |
| OC-6 | An approved change reaches production within one day of approval. | pass | Capability C-6 manages release and deployment, with the Customer Platform Engineer (R-2) responsible, ensuring approved changes reach production within a day. |
| OC-7 | Every change to the network requires explicit approval from the customer platform team prior to implementation. | pass | Capability C-5 manages change enablement, with the Customer Platform Lead (R-1) accountable and engineers responsible, ensuring network changes require explicit approval. |
| OC-8 | A report detailing service levels is delivered to the customer monthly. | pass | Capability C-15 covers service level management to measure SLA achievement and deliver the monthly report, with the Provider Delivery Lead (R-3) accountable and responsible. |
| OC-9 | Platform operations is conducted under a shared service model between the customer and external provider. | pass | Capability C-20 handles supplier management to ensure shared service model compliance, with the Customer Platform Lead (R-1) accountable and responsible. |
| OC-10 | The Customer Platform Lead role is located in the UK. | pass | The Customer Platform Lead role (R-1) is defined with country 'GB' (UK). |
| OC-11 | The Customer Platform Engineer role is located in the UK. | pass | The Customer Platform Engineer role (R-2) is defined with country 'GB' (UK). |
| OC-12 | The Provider Delivery Lead role is located in the UK. | pass | The Provider Delivery Lead role (R-3) is defined with country 'GB' (UK). |
| OC-13 | The Provider 24x7 SRE role is located in Poland. | pass | The Provider 24x7 SRE role (R-4) is defined with country 'PL' (Poland). |
| OC-14 | The Provider Platform Engineer role is located in India. | pass | The Provider Platform Engineer role (R-5) is defined with country 'IN' (India). |
| OC-15 | A report detailing operational costs is delivered to the customer monthly. | pass | Capability C-14 covers cost management to compile financial reports on monthly landing zone spend, with the Provider Delivery Lead (R-3) responsible. |
Each scenario is generated from the operating model and traced through it by code.
| Scenario | Result | Path |
| --- | --- | --- |
| Hub network and interconnect (S-1) fails at an hour faults are responded to (24x7). | pass | Provider 24x7 SRE responds under C-1, and Provider Delivery Lead is answerable. |
| Hub network and interconnect (S-1) must be changed. | pass | Customer Platform Engineer makes the change under C-5. |
| Audit log store (S-2) fails at an hour faults are responded to (24x7). | pass | Provider 24x7 SRE responds under C-1, and Provider Delivery Lead is answerable. |
| Audit log store (S-2) must be changed. | pass | Customer Platform Engineer makes the change under C-5. |
| Deployment pipeline (S-3) fails at an hour faults are responded to (24x7). | pass | Provider 24x7 SRE responds under C-1, and Provider Delivery Lead is answerable. |
| Deployment pipeline (S-3) must be changed. | pass | Customer Platform Engineer makes the change under C-5. |
| Encryption keys (S-4) fails at an hour faults are responded to (24x7). | pass | Provider 24x7 SRE responds under C-1, and Provider Delivery Lead is answerable. |
| Encryption keys (S-4) must be changed. | pass | Customer Platform Engineer makes the change under C-5. |
| Condition K-1 must be shown to hold: The platform is available at least 99.9% of each month. | pass | Check OC-1 on C-1 keeps it true. |
| Condition K-2 must be shown to hold: Audit logs cannot be deleted for seven years. | pass | Check OC-2 on C-8 keeps it true. |
| Condition K-3 must be shown to hold: Encryption keys are rotated every 90 days. | pass | Check OC-3 on C-7 keeps it true. |
| Service Provider cannot be reached. | pass | Every Capability it is answerable for has a Role in another Party to turn to. |
| A month's cost of operating is compared with the budget of 60,000 GBP. | pass | The cost is 46,252 GBP, within the budget. |
## Findings and open questions
No findings: the solution provides what operating it needs.
Assumed for the customer, and open to change when they answer:
- `intent.exclusions`: Application workloads and services hosted on the platform foundation. Assumed based on platform foundation scope: workloads hosted on top of the landing zone are out of scope for this operating model.
- `operations.change_process`: Changes go through a deployment pipeline with peer review, with explicit customer platform team approval required for all network changes.. Pipeline-driven changes with peer review, with customer platform team approval required for network changes per REQ-01.
- `service_management.tooling`: Google Cloud operations suite and native Cloud provider tooling, with no separate external ITSM system.. Default native GCP operations tooling used as no external ITSM was specified.
## Sources
The Capabilities are from our own capability catalogue, which draws on these frameworks. It cites them and certifies nothing against them.
- ITIL 4: Incident management practice; Monitoring and event management practice; Problem management practice; Service request management practice; Change enablement practice; Release management and Deployment management practices; Service configuration management practice; Service continuity management practice; Capacity and performance management practice; Service level management practice; Supplier management practice
- Site Reliability Engineering: How Google Runs Production Systems: Managing Incidents; Monitoring Distributed Systems; Postmortem Culture: Learning from Failure; Release Engineering; Service Level Objectives
- ISO/IEC 20000-1: Incident management; Service request management; Change management; Service continuity management; Capacity management; Service level management; Supplier management
- Regulation (EU) 2022/2554, the Digital Operational Resilience Act: Article 17, ICT-related incident management process; Article 10, Detection; Article 12, Backup policies and procedures, restoration and recovery procedures and methods; Article 11, Response and recovery; Article 9, Protection and prevention; Article 28, General principles on ICT third-party risk; Article 19, Reporting of major ICT-related incidents
- NIST Cybersecurity Framework: Detect: Continuous Monitoring; Identify: Asset Management; Recover: Incident Recovery Plan Execution; Protect: Identity Management, Authentication, and Access Control; Protect: Platform Security; Detect: Adverse Event Analysis; Respond: Incident Management
- DevOps Research and Assessment (DORA) software delivery metrics: Change failure rate; Deployment frequency and lead time for changes; Deployment frequency; Lead time for changes; Change failure rate; Time to restore service
- ISO/IEC 27001: Access control; Management of technical vulnerabilities; Information security incident management; Compliance with legal and contractual requirements
- FinOps Framework: Allocation; Budgeting; Forecasting; Anomaly Management
- CNCF Platforms White Paper (platform as a product): Platform as a product; Measuring platform success; Platform capabilities; Self-service; Platform as a product
- Team Topologies: Organizing Business and Technology Teams for Fast Flow: Platform team; Thinnest viable platform; X-as-a-Service interaction mode; Enabling team; Facilitating interaction mode
- The SPACE of Developer Productivity: Satisfaction and well-being; Performance; Activity; Communication and collaboration; Efficiency and flow
- DevEx: What Actually Drives Productivity: Feedback loops; Cognitive load; Flow state
**These references have not been checked against the frameworks by a person.** Treat each as a pointer to where to look.

View File

@@ -0,0 +1,39 @@
# Scope statement
Operations covers the Google Cloud landing zone platform foundation, comprising the hub network and interconnect, the audit log store, the deployment pipeline, and encryption keys. The platform is operated under a shared service model between the customer's platform team based in the United Kingdom and a service provider delivering from the United Kingdom, Poland, and India, with the customer platform team retaining approval authority over all network changes. Operational support and system availability are maintained on a 24x7 basis, with priority 1 incident response within 15 minutes and target recovery within 4 hours. All operational costs are stated in British Pounds (GBP). Application workloads and services hosted on the platform foundation are excluded from operational scope.
## Information register
20 stated by the customer, 0 answered in conversation, 3 assumed, 10 unknown, 0 not applicable.
| Field | Value | From |
| --- | --- | --- |
| engagement.kind | platform_foundation | the scope spec |
| intent.statement | We want our landing zone run as a shared service from go-live. A provider should respond to faults at every hour, and our own platform team keeps the final say over changes to the network. | the scope spec |
| intent.requirements | [{"id": "REQ-01", "statement": "Our platform team approves every change to the network."}, {"id": "REQ-02", "statement": "We receive a monthly report of service levels and cost."}] | the scope spec |
| intent.business_outcomes | [{"id": "BO-1", "outcome": "New payment products reach customers in weeks rather than months.", "target": "A change reaches production within a day of approval."}, {"id": "BO-2", "outcome": "Payments | the scope spec |
| solution.cloud_provider | gcp | the scope spec |
| solution.items | [{"id": "S-1", "name": "Hub network and interconnect", "kind": "network", "product": "Cloud Interconnect"}, {"id": "S-2", "name": "Audit log store", "kind": "storage", "product": "Cloud Storage"}, {"i | the scope spec |
| solution.conditions | [{"id": "K-1", "on": ["S-1"], "condition": "The platform is available at least 99.9% of each month."}, {"id": "K-2", "on": ["S-2"], "condition": "Audit logs cannot be deleted for seven years."}, {"id" | the scope spec |
| solution.build.duration_days | 36 | the scope spec |
| operations.support_hours | 24x7 | the scope spec |
| operations.availability_hours | 24x7 | the scope spec |
| sourcing.model | shared | the scope spec |
| sourcing.delivery_countries | ["GB", "PL", "IN"] | the scope spec |
| organisation.existing_teams | [{"name": "Platform team", "party": "customer", "hours": "business", "country": "GB"}] | the scope spec |
| service_levels.priority_1_response_minutes | 15 | the scope spec |
| service_levels.recovery_time_hours | 4 | the scope spec |
| compliance.frameworks | ["DORA"] | the scope spec |
| commercial.currency | GBP | the scope spec |
| commercial.monthly_run_budget | 60000 | the scope spec |
| intent.exclusions | ["Application workloads and services hosted on the platform foundation"] | assumed |
| operations.change_process | Changes go through a deployment pipeline with peer review, with explicit customer platform team approval required for all network changes. | assumed |
| service_management.tooling | Google Cloud operations suite and native Cloud provider tooling, with no separate external ITSM system. | assumed |
### Assumptions
| Id | Assumption | If it is wrong |
| --- | --- | --- |
| ASSUMP-01 | Assumed intent.exclusions is ['Application workloads and services hosted on the platform foundation']. Assumed based on platform foundation scope: workloads hosted on top of the landing zone are out of scope for this operating model. | Changes which Capabilities are designed. |
| ASSUMP-02 | Assumed operations.change_process is Changes go through a deployment pipeline with peer review, with explicit customer platform team approval required for all network changes.. Pipeline-driven changes with peer review, with customer platform team approval required for network changes per REQ-01. | Changes change enablement and release. |
| ASSUMP-03 | Assumed service_management.tooling is Google Cloud operations suite and native Cloud provider tooling, with no separate external ITSM system.. Default native GCP operations tooling used as no external ITSM was specified. | Changes tooling and the processes written for it. |

View File

@@ -0,0 +1,45 @@
# Operational requirements
The operational requirements specify what running the Google Cloud platform foundation must achieve. Each requirement is derived directly from customer statements, binding scoping parameters, solution conditions, or stated business outcomes.
### Service level requirements
- **OR-1**: A priority 1 fault on any Item is acknowledged within 15 minutes, at any hour (24x7 coverage).
- **OR-2**: Service is restored within 4 hours following a priority 1 incident or critical failure.
- **OR-3**: The hub network and interconnect maintains at least 99.9% availability during each calendar month.
- **OR-4**: An approved change reaches production deployment within one day of approval.
### Process requirements
- **OR-5**: Every change to the network requires explicit approval from the customer platform team prior to implementation.
- **OR-6**: A report detailing operational service levels and run costs is delivered to the customer on a monthly basis.
- **OR-7**: Cryptographic encryption keys are rotated at least once every 90 days.
### Organisation requirements
- **OR-8**: Platform operations is conducted under a shared service model between the customer and an external service provider.
- **OR-9**: Operational delivery activities are conducted exclusively from the United Kingdom, Poland, and India.
### Compliance requirements
- **OR-10**: Audit logs stored in the audit log store cannot be deleted for seven years.
---
# What was not stated
The operating specification leaves certain operational aspects assumed or unknown. If answered differently by the customer, these would alter the operational requirements.
## Assumptions
- **Scope exclusions (`intent.exclusions`)**: It is assumed that application workloads and services hosted on top of the platform foundation are excluded from operations. If customer workloads were included in scope, additional operational requirements would be needed for workload availability, deployment pipeline maintenance, and application incident response.
- **Change process (`operations.change_process`)**: It is assumed that changes proceed through a continuous deployment pipeline with automated checks and peer review, with customer approval reserved specifically for network modifications. If a centralised Change Advisory Board (CAB) or multi-tier manual approval process were required across all components, change velocity requirements would be directly affected.
- **Service management tooling (`service_management.tooling`)**: It is assumed that Google Cloud native operations suite tooling is utilized without a separate enterprise ITSM platform. If an external ITSM or specific ticketing system were mandated, dedicated tooling integration requirements would apply.
## Unknowns
- **Go-live date (`transition.go_live_date`)**: The target operational cutover date was not specified, leaving transitional readiness timelines to be determined.
- **Customer-owned build components (`solution.build.customer_owned`)**: The exact division of build ownership between customer and suppliers was not defined, which may influence handoff requirements.
- **Inherited risks (`solution.inherited_risks`)**: No existing technical debt or inherited design risks were declared; if identified, specific operational monitoring or mitigation guardrails would be required.
- **Monitoring integrations (`operations.monitoring`)**: Specific external monitoring, log ingestion, or enterprise SIEM destinations were not stated beyond the native audit log store.
- **Operational headcount limits (`organisation.headcount_limit`)**: No upper bound on staffing was set, with staffing to be determined strictly by service hours and capability requirements.

View File

@@ -0,0 +1,66 @@
# Design rationale
## Sourcing and Division of Work
The operating model establishes a shared service model between the Customer (P-1) and an external Service Provider (P-2).
- The **Customer Platform team** (T-1) retains accountability for the platform's architectural integrity and roadmap, as well as operational decisions surrounding changes to the network. The team handles change enablement, self-service golden paths, delivery and developer experience measurement, and capacity management.
- The **Provider Operations team** (T-2) is tasked with delivering 24x7 visibility and stability. They provide incident management, security monitoring, access management, backup and recovery, and handle routine service requests. This arrangement allows the Customer to focus on enablement and evolving the platform product rather than daily operational toil.
## Roles, Countries, and Levels
The Roles map directly to the `sourcing.delivery_countries` constraints (GB, PL, IN) and specific support hours:
- **Customer Platform Lead (GB, Lead, Business hours)**: Ensures accountability and oversees the platform's product roadmap, holding final approval over network changes to meet governance requirements.
- **Customer Platform Engineer (GB, Senior, Business hours)**: Focuses on engineering tasks, automation, enablement, and maintaining the deployment pipeline during standard business hours.
- **Provider Delivery Lead (GB, Lead, Business hours)**: Acts as the primary point of contact for the Service Provider. They are accountable for incident management, security monitoring, and compiling monthly reports (detailing service levels and costs) ensuring strict governance.
- **Provider 24x7 SRE (PL, Senior, 24x7)**: Selected in Poland to offer around-the-clock incident response, event monitoring, and security oversight, ensuring 15-minute response times and four-hour restoration for priority incidents. A senior level is appropriate to diagnose unfamiliar faults and autonomously lead shifts.
- **Provider Platform Engineer (IN, Engineer, Business hours)**: Located in India to provide cost-efficient business-hours support for routine service requests, problem management, access requests, patch management, and configuration tasks.
*(Note: Costs derived from this operating model will use placeholders as the rate card remains in draft.)*
## Assumptions
- We assume that the customer platform team can review and approve network changes within the pipeline seamlessly, without becoming a bottleneck that prevents the 1-day deployment target.
- We assume that natively available GCP tools (like Cloud Monitoring and Cloud Logging) are used for alert configuration and routing to the 24x7 SRE team, avoiding the need for third-party integrations or findings for specific tools.
- We assume backup and recovery policies will be enforced through configuration mechanisms as opposed to creating distinct tools requiring external findings.
## Responsibility matrix
| Capability | Customer Platform Lead | Customer Platform Engineer | Provider Delivery Lead | Provider 24x7 SRE | Provider Platform Engineer |
| --- | --- | --- | --- | --- | --- |
| Incident management (C-1) | C | | A | R | |
| Monitoring and alerting (C-2) | C | | A | R | |
| Problem management (C-3) | C | | A | | R |
| Service request fulfilment (C-4) | C | | A | | R |
| Change enablement (C-5) | A | R | | | R |
| Release and deployment (C-6) | A | R | | | |
| Configuration management (C-7) | A | | | | R |
| Backup and recovery (C-8) | C | | A | | R |
| Service continuity (C-9) | C | | A | | R |
| Capacity and performance management (C-10) | A | | | | R |
| Access management (C-11) | C | | A | | R |
| Vulnerability and patch management (C-12) | C | | A | | R |
| Security monitoring and response (C-13) | C | | A | R | |
| Cost management (C-14) | A | | R | | |
| Service level management (C-15) | C | | A/R | | |
| Platform product management (C-16) | A/R | | | | |
| Self-service and golden paths (C-17) | A | R | | | |
| Enablement (C-18) | A | R | | | |
| Delivery and developer experience measurement (C-19) | A | R | | | |
| Supplier management (C-20) | A/R | | | | |
| Regulatory evidence and reporting (C-21) | C | | A | | R |
A accountable, R responsible, C consulted, I informed.
## Staffing and cost
| Role | Party | Country | Level | Hours | Effort a month | People | Cost a month |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Customer Platform Lead | Customer | GB | lead | business | 30.0 h | 0.2 | not priced |
| Customer Platform Engineer | Customer | GB | senior | business | 85.0 h | 0.6 | not priced |
| Provider Delivery Lead | Service Provider | GB | lead | business | 20.0 h | 0.1 | 2,212 GBP |
| Provider 24x7 SRE | Service Provider | PL | senior | 24x7 | 110.0 h | 5.0 | 41,640 GBP |
| Provider Platform Engineer | Service Provider | IN | engineer | business | 135.0 h | 0.9 | 2,400 GBP |
6.8 people in all. Cost a month: 46,252 GBP.
The monthly budget is 60,000 GBP: this is within it.
Worked from rate card `rate-card`, version `draft-2026-10-13`. **Its figures are placeholders: no price has been set, and this cost is an illustration.**
Converted from USD at 0.79, a rate set on 2026-10-13.
The customer's own staff are counted and not priced. A Role that covers hours wider than a working day is never smaller than the team that can cover them.

View File

@@ -0,0 +1,42 @@
# Check verdicts
All checks against the target operating model have successfully passed. The operating model is capable of meeting the defined conditions and requirements as designed.
- **OC-1**: **Pass**. Capability C-1 is responsible for incident management with 24x7 coverage by the Provider 24x7 SRE role (R-4), which is sufficient to maintain highly available operations by responding to faults around the clock.
- **OC-2**: **Pass**. Capability C-8 manages backup and recovery, validating retention policies for audit logs. It is appropriately staffed during business hours.
- **OC-3**: **Pass**. Capability C-7 covers configuration management including tracking key rotation cycles, with the Customer Platform Lead (R-1) accountable and Provider Platform Engineer (R-5) responsible.
- **OC-4**: **Pass**. Capability C-1 covers incident management 24x7 with the Provider 24x7 SRE (R-4) responsible, ensuring 15-minute acknowledgement at any hour.
- **OC-5**: **Pass**. Capability C-1 provides 24x7 incident response, ensuring that service is restored within the 4-hour target following a priority 1 incident.
- **OC-6**: **Pass**. Capability C-6 manages release and deployment, with the Customer Platform Engineer (R-2) responsible, ensuring approved changes reach production within a day.
- **OC-7**: **Pass**. Capability C-5 manages change enablement, with the Customer Platform Lead (R-1) accountable and engineers responsible, ensuring network changes require explicit approval.
- **OC-8**: **Pass**. Capability C-15 covers service level management to measure SLA achievement and deliver the monthly report, with the Provider Delivery Lead (R-3) accountable and responsible.
- **OC-9**: **Pass**. Capability C-20 handles supplier management to ensure shared service model compliance, with the Customer Platform Lead (R-1) accountable and responsible.
- **OC-10**: **Pass**. The Customer Platform Lead role (R-1) is defined with country 'GB' (UK).
- **OC-11**: **Pass**. The Customer Platform Engineer role (R-2) is defined with country 'GB' (UK).
- **OC-12**: **Pass**. The Provider Delivery Lead role (R-3) is defined with country 'GB' (UK).
- **OC-13**: **Pass**. The Provider 24x7 SRE role (R-4) is defined with country 'PL' (Poland).
- **OC-14**: **Pass**. The Provider Platform Engineer role (R-5) is defined with country 'IN' (India).
- **OC-15**: **Pass**. Capability C-14 covers cost management to compile financial reports on monthly landing zone spend, with the Provider Delivery Lead (R-3) responsible.
**Overall Verdict**
**Pass**. The target operating model successfully satisfies all required operational checks.
## Scenario walk-throughs
Each scenario is generated from the operating model and traced through it by code.
| Scenario | Result | Path |
| --- | --- | --- |
| Hub network and interconnect (S-1) fails at an hour faults are responded to (24x7). | pass | Provider 24x7 SRE responds under C-1, and Provider Delivery Lead is answerable. |
| Hub network and interconnect (S-1) must be changed. | pass | Customer Platform Engineer makes the change under C-5. |
| Audit log store (S-2) fails at an hour faults are responded to (24x7). | pass | Provider 24x7 SRE responds under C-1, and Provider Delivery Lead is answerable. |
| Audit log store (S-2) must be changed. | pass | Customer Platform Engineer makes the change under C-5. |
| Deployment pipeline (S-3) fails at an hour faults are responded to (24x7). | pass | Provider 24x7 SRE responds under C-1, and Provider Delivery Lead is answerable. |
| Deployment pipeline (S-3) must be changed. | pass | Customer Platform Engineer makes the change under C-5. |
| Encryption keys (S-4) fails at an hour faults are responded to (24x7). | pass | Provider 24x7 SRE responds under C-1, and Provider Delivery Lead is answerable. |
| Encryption keys (S-4) must be changed. | pass | Customer Platform Engineer makes the change under C-5. |
| Condition K-1 must be shown to hold: The platform is available at least 99.9% of each month. | pass | Check OC-1 on C-1 keeps it true. |
| Condition K-2 must be shown to hold: Audit logs cannot be deleted for seven years. | pass | Check OC-2 on C-8 keeps it true. |
| Condition K-3 must be shown to hold: Encryption keys are rotated every 90 days. | pass | Check OC-3 on C-7 keeps it true. |
| Service Provider cannot be reached. | pass | Every Capability it is answerable for has a Role in another Party to turn to. |
| A month's cost of operating is compared with the budget of 60,000 GBP. | pass | The cost is 46,252 GBP, within the budget. |

View File

@@ -0,0 +1,54 @@
## Capabilities
The following capabilities establish what the operating organisation must be able to execute to satisfy the defined operational requirements.
*(Note: While the attached rate card remains a draft, any costs computed downstream from these capabilities are strictly placeholders.)*
| ID | Capability | Purpose / Why it is needed | Items covered | Serves |
| :--- | :--- | :--- | :--- | :--- |
| **C-1** | Incident management | To restore normal service after a fault and handle Priority 1 incidents round-the-clock. | S-1, S-2, S-3, S-4 | OR-1, OR-2 |
| **C-2** | Monitoring and alerting | To detect platform faults, track 99.9% availability, and trigger instant alerts across 24x7 hours. | S-1, S-2, S-3, S-4 | OR-1, OR-3 |
| **C-3** | Problem management | To perform root cause analysis and prevent the recurrence of high-impact platform incidents. | S-1, S-2, S-3, S-4 | OR-2, OR-3 |
| **C-4** | Service request fulfilment | To manage standard requests and access requests between the shared sourcing teams. | S-1, S-2, S-3, S-4 | OR-6 |
| **C-5** | Change enablement | To review risks and ensure all hub network changes receive customer platform team approval. | S-1, S-2, S-3, S-4 | OR-4 |
| **C-6** | Release and deployment | To execute automated pipelines smoothly and roll back any failing changes. | S-1, S-3 | OR-4 |
| **C-7** | Configuration management | To maintain infrastructure-as-code state accuracy for all network and key configurations. | S-1, S-2, S-3, S-4 | OR-4 |
| **C-8** | Backup and recovery | To ensure data resilience, key backups, and recovery from catastrophic component loss. | S-2, S-4 | OR-2 |
| **C-9** | Service continuity | To rehearse and ensure the platform foundation recovers within 4 hours in disaster scenarios. | S-1, S-2, S-3, S-4 | OR-2, OR-3 |
| **C-10** | Capacity and performance management | To monitor hub network bandwidth and pipeline usage against limits. | S-1, S-3 | OR-3 |
| **C-11** | Access management | To restrict log and key access tightly, preventing unauthorized deletion of audit logs. | S-1, S-2, S-3, S-4 | OR-8 |
| **C-12** | Vulnerability and patch management | To routinely scan for flaws and apply necessary mitigations, protecting platform availability. | S-1, S-2, S-3, S-4 | OR-3 |
| **C-13** | Security monitoring and response | To monitor security incidents 24x7 and rapidly react to malicious events. | S-1, S-2, S-3, S-4 | OR-1, OR-8 |
| **C-14** | Cost management | To track cloud spend against budget and compile the cost segment of the monthly report. | S-1, S-2, S-3, S-4 | OR-5 |
| **C-15** | Service level management | To measure compliance with stated SLAs and package the findings into the monthly customer report. | S-1, S-2, S-3, S-4 | OR-5 |
| **C-16** | Platform product management | To manage the foundation as a product, prioritizing improvements based on developer feedback. | S-1, S-2, S-3, S-4 | OR-6 |
| **C-17** | Self-service and golden paths | To offer default, streamlined pipeline deployment paths for easier developer adoption. | S-3 | OR-4, OR-6 |
| **C-18** | Enablement | To assist the customer team with using the foundational deployment pipelines via pairing and training. | S-3 | OR-6 |
| **C-19** | Delivery and developer experience measurement | To report on metrics such as deployment speed and team satisfaction, supporting continuous improvement. | S-3 | OR-5 |
| **C-20** | Supplier management | To govern the multi-location external provider and ensure location constraints are enforced. | S-1, S-2, S-3, S-4 | OR-6, OR-7 |
| **C-21** | Regulatory evidence and reporting | To collect audit evidence of 90-day key rotations and 7-year log retention for compliance alignment. | S-2, S-4 | OR-8, OR-9 |
## Sources
The capability catalogue draws heavily on prominent industry frameworks.
*(Note: All sources referenced in this capability model are currently flagged as **unverified** because they are drawn from the catalogue's register, whose status indicates they were drafted from general knowledge. They have not been independently certified. The capability catalogue draws on them and promises nothing regarding compliance.)*
* **ITIL 4** (PeopleCert, version 4): Incident management practice, Monitoring and event management practice, Problem management practice, Service request management practice, Change enablement practice, Release management and Deployment management practices, Service configuration management practice, Service continuity management practice, Capacity and performance management practice, Supplier management practice, Service level management practice.
* **Site Reliability Engineering: How Google Runs Production Systems** (Google / O'Reilly, 2016): Managing Incidents, Monitoring Distributed Systems, Postmortem Culture: Learning from Failure, Release Engineering, Service Level Objectives.
* **ISO/IEC 20000-1** (ISO/IEC, 2018): Incident management, Service request management, Change management, Service continuity management, Capacity management, Service level management, Supplier management.
* **Regulation (EU) 2022/2554, the Digital Operational Resilience Act** (European Union, 2022): Article 9, Protection and prevention; Article 10, Detection; Article 11, Response and recovery; Article 12, Backup policies and procedures, restoration and recovery procedures and methods; Article 17, ICT-related incident management process; Article 19, Reporting of major ICT-related incidents; Article 28, General principles on ICT third-party risk.
* **NIST Cybersecurity Framework** (National Institute of Standards and Technology, 2.0): Detect: Continuous Monitoring; Detect: Adverse Event Analysis; Respond: Incident Management; Identify: Asset Management; Recover: Incident Recovery Plan Execution; Protect: Identity Management, Authentication, and Access Control; Protect: Platform Security.
* **ISO/IEC 27001** (ISO/IEC, 2022): Access control, Management of technical vulnerabilities, Information security incident management, Compliance with legal and contractual requirements.
* **FinOps Framework** (FinOps Foundation, 2024): Allocation; Budgeting; Forecasting; Anomaly Management.
* **Team Topologies: Organizing Business and Technology Teams for Fast Flow** (IT Revolution, 2019): Platform team; Thinnest viable platform; X-as-a-Service interaction mode; Enabling team; Facilitating interaction mode.
* **CNCF Platforms White Paper (platform as a product)** (Cloud Native Computing Foundation, 2023): Platform as a product; Measuring platform success; Platform capabilities; Self-service.
* **DevOps Research and Assessment (DORA) software delivery metrics** (Google Cloud, 2023): Change failure rate, Deployment frequency and lead time for changes, Time to restore service.
* **The SPACE of Developer Productivity** (Forsgren et al., ACM Queue, 2021): Satisfaction and well-being; Performance; Activity; Communication and collaboration; Efficiency and flow.
* **DevEx: What Actually Drives Productivity** (Noda et al., ACM Queue, 2023): Feedback loops; Cognitive load; Flow state.
## Assumptions
Since pre-approved execution is enabled, the following assumptions were made regarding capability assignments and requirements:
- Single instances of each capability per catalogue entry are sufficient, rather than splitting capabilities based on differing items. The shared sourcing model's division of duties is expected to be managed later at the Role layer.
- Operational capabilities associated with "delivery-and-developer experience measurement" actively apply due to the business intent to streamline delivery and preserve developer retention. These are adopted to provide transparent progress towards these goals without mandating complex underlying toolchains.

View File

@@ -0,0 +1,234 @@
# Operating model guide: Google Cloud landing zone foundation
## Summary
This guide defines the target operating model for running the Google Cloud Platform landing zone foundation, comprising the hub network and interconnect, audit log store, deployment pipeline, and cryptographic keys, while application workloads deployed onto the foundation remain out of operational scope. Operations are shared between the customer platform team and an external service provider across the United Kingdom, Poland, and India. The customer platform team retains platform product ownership and final approval authority over all network changes, while the external service provider delivers continuous monitoring and rapid fault response alongside pipeline maintenance and enablement.
Operational support and platform availability operate continuously round the clock, with the provider handling fault response across all hours, while change enablement, platform engineering, and governance occur during business hours. Readers should look first at the findings and open questions section to review the unconfirmed operational assumptions and unstated items—including the native operations tooling baseline and workload boundaries—as well as the provisional nature of the draft rate card.
### Key figures
- Operated: 4 Items, with 3 Conditions to keep true.
- Faults responded to: 24x7 hours.
- Capabilities: 21.
- People: 11.0. Cost a month: 44,720 GBP, from rate card version `draft-2026-10-13` (placeholder prices: an illustration).
- Scenario walk-throughs: 13 of 13 pass.
- Business outcomes: 3, 3 shown by a Measure.
- Findings for whoever owns the design: 0.
## Business outcomes
**BO-1: New payment products reach customers in weeks rather than months.** Target: A change reaches production within a day of approval..
| Measure | Layer | Framework | Target | Measured by | Answerable |
| --- | --- | --- | --- | --- | --- |
| Deployment frequency | outcome | dora-metrics | A change reaches production within a day of approval. | Delivery and developer experience measurement | Operations Lead |
**BO-2: Payments are never unavailable for long.** Target: Restored within 4 hours..
| Measure | Layer | Framework | Target | Measured by | Answerable |
| --- | --- | --- | --- | --- | --- |
| Time to restore service | outcome | dora-metrics | Restored within 4 hours. | Delivery and developer experience measurement | Operations Lead |
**BO-3: We keep the platform engineers we hire.**
| Measure | Layer | Framework | Target | Measured by | Answerable |
| --- | --- | --- | --- | --- | --- |
| Satisfaction and well-being | guardrail | space-framework | | Delivery and developer experience measurement | Operations Lead |
| Friction and cognitive load | leading | devex-framework | | Delivery and developer experience measurement | Operations Lead |
## What is operated
| Item | What it is | Kind | Product | From |
| --- | --- | --- | --- | --- |
| S-1 | Hub network and interconnect | network | Cloud Interconnect | |
| S-2 | Audit log store | storage | Cloud Storage | |
| S-3 | Deployment pipeline | pipeline | Cloud Build | |
| S-4 | Encryption keys | keys | Cloud KMS | |
What must be kept true while it is run:
| Condition | On | What must hold | From |
| --- | --- | --- | --- |
| K-1 | S-1 | The platform is available at least 99.9% of each month. | |
| K-2 | S-2 | Audit logs cannot be deleted for seven years. | |
| K-3 | S-4 | Encryption keys are rotated every 90 days. | |
## Operational requirements
| Requirement | Kind | What must be achieved | From |
| --- | --- | --- | --- |
| OR-1 | service_level | A Priority 1 fault on any landing zone component is acknowledged and responded to within 15 minutes, at any hour (24x7). | service_levels.priority_1_response_minutes, operations.support_hours |
| OR-2 | service_level | Normal platform operations are restored within 4 hours of a Priority 1 incident occurring. | service_levels.recovery_time_hours |
| OR-3 | service_level | The platform foundation maintains an operational availability of at least 99.9% across all hours (24x7) of each calendar month. | K-1, operations.availability_hours |
| OR-4 | process | Every change to the hub network requires customer platform team approval prior to deployment. | REQ-01 |
| OR-5 | process | A reporting package detailing achieved service levels and operational costs is delivered to the customer on a monthly basis. | REQ-02 |
| OR-6 | organisation | Platform foundation operations are delivered under a shared sourcing model between the customer and an external service provider. | sourcing.model |
| OR-7 | organisation | Operational services are delivered exclusively from the United Kingdom, Poland, and India. | sourcing.delivery_countries |
| OR-8 | compliance | Audit logs stored in the audit log store cannot be deleted and are retained for at least seven years. | K-2 |
| OR-9 | compliance | Cryptographic encryption keys are rotated at least once every 90 days. | K-3 |
## Capabilities
| Capability | What it is | Serves | Covers | Hours | Effort a month |
| --- | --- | --- | --- | --- | --- |
| C-1 | Incident management | OR-1, OR-2 | S-1, S-2, S-3, S-4 | 24x7 | 20 h |
| C-2 | Monitoring and alerting | OR-1, OR-3 | S-1, S-2, S-3, S-4 | 24x7 | 24 h |
| C-3 | Problem management | OR-2, OR-3 | S-1, S-2, S-3, S-4 | business | 12 h |
| C-4 | Service request fulfilment | OR-6 | S-1, S-2, S-3, S-4 | business | 15 h |
| C-5 | Change enablement | OR-4 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-6 | Release and deployment | OR-4 | S-1, S-3 | business | 12 h |
| C-7 | Configuration management | OR-4 | S-1, S-2, S-3, S-4 | business | 8 h |
| C-8 | Backup and recovery | OR-2 | S-2, S-4 | business | 4 h |
| C-9 | Service continuity | OR-2, OR-3 | S-1, S-2, S-3, S-4 | business | 8 h |
| C-10 | Capacity and performance management | OR-3 | S-1, S-3 | business | 4 h |
| C-11 | Access management | OR-8 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-12 | Vulnerability and patch management | OR-3 | S-1, S-2, S-3, S-4 | business | 16 h |
| C-13 | Security monitoring and response | OR-1, OR-8 | S-1, S-2, S-3, S-4 | 24x7 | 24 h |
| C-14 | Cost management | OR-5 | S-1, S-2, S-3, S-4 | business | 6 h |
| C-15 | Service level management | OR-5 | S-1, S-2, S-3, S-4 | business | 8 h |
| C-16 | Platform product management | OR-6 | S-1, S-2, S-3, S-4 | business | 12 h |
| C-17 | Self-service and golden paths | OR-4, OR-6 | S-3 | business | 16 h |
| C-18 | Enablement | OR-6 | S-3 | business | 10 h |
| C-19 | Delivery and developer experience measurement | OR-5 | S-3 | business | 8 h |
| C-20 | Supplier management | OR-6, OR-7 | S-1, S-2, S-3, S-4 | business | 4 h |
| C-21 | Regulatory evidence and reporting | OR-8, OR-9 | S-2, S-4 | business | 8 h |
## Who does what
| Role | Party | Country | Level | Hours |
| --- | --- | --- | --- | --- |
| Platform Lead | Customer platform team (customer) | GB | lead | business |
| Platform Engineer | Customer platform team (customer) | GB | engineer | business |
| Operations Lead | External service provider (provider) | PL | lead | business |
| Operations Engineer | External service provider (provider) | IN | engineer | 24x7 |
| Security Operations Engineer | External service provider (provider) | IN | senior | 24x7 |
| Senior SRE | External service provider (provider) | IN | senior | business |
| Enablement Engineer | External service provider (provider) | IN | senior | business |
The teams, as Team Topologies names their kinds:
| Team | Kind | Party | Roles | Purpose |
| --- | --- | --- | --- | --- |
| Platform team | platform | Customer platform team | Platform Lead, Platform Engineer | Customer platform foundation ownership |
| Shared operations | platform | External service provider | Operations Lead, Operations Engineer, Security Operations Engineer, Senior SRE | Shared operational delivery of platform foundations |
| Enablement team | enabling | External service provider | Enablement Engineer | Facilitating adoption of pipelines |
How the teams work together:
| From | To | Mode | Until | Purpose |
| --- | --- | --- | --- | --- |
| Platform team | Shared operations | x-as-a-service | | Customer platform team consumes shared operational services and platforms provided by the external provider. |
| Enablement team | Platform team | facilitating | Adoption of foundational pipelines is complete and customer team is self-sufficient. | Provider enablement team assists the customer with adopting foundational deployment pipelines. |
| Capability | Platform Lead | Platform Engineer | Operations Lead | Operations Engineer | Security Operations Engineer | Senior SRE | Enablement Engineer |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Incident management (C-1) | | C | A | R | | | |
| Monitoring and alerting (C-2) | | C | A | R | | | |
| Problem management (C-3) | | C | A | | | R | |
| Service request fulfilment (C-4) | | C | A | R | | | |
| Change enablement (C-5) | A | R | | | | C | |
| Release and deployment (C-6) | | C | A | | | R | |
| Configuration management (C-7) | | C | A | | | R | |
| Backup and recovery (C-8) | | C | A | | | R | |
| Service continuity (C-9) | | C | A | | | R | |
| Capacity and performance management (C-10) | | C | A | | | R | |
| Access management (C-11) | | C | A | | R | | |
| Vulnerability and patch management (C-12) | | C | A | | | R | |
| Security monitoring and response (C-13) | | C | A | | R | | |
| Cost management (C-14) | C | | A | | | R | |
| Service level management (C-15) | C | | A | | | R | |
| Platform product management (C-16) | A/R | | | | | | |
| Self-service and golden paths (C-17) | | C | A | | | R | |
| Enablement (C-18) | | C | A | | | | R |
| Delivery and developer experience measurement (C-19) | C | | A | | | R | |
| Supplier management (C-20) | A/R | | | | | | |
| Regulatory evidence and reporting (C-21) | | C | A | | | R | |
A accountable, R responsible, C consulted, I informed.
## Staffing and cost
| Role | Party | Country | Level | Hours | Effort a month | People | Cost a month |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Platform Lead | Customer platform team | GB | lead | business | 16.0 h | 0.1 | not priced |
| Platform Engineer | Customer platform team | GB | engineer | business | 10.0 h | 0.1 | not priced |
| Operations Lead | External service provider | PL | lead | business | 0.0 h | 0.0 | 0 GBP |
| Operations Engineer | External service provider | IN | engineer | 24x7 | 59.0 h | 5.0 | 17,034 GBP |
| Security Operations Engineer | External service provider | IN | senior | 24x7 | 34.0 h | 5.0 | 24,605 GBP |
| Senior SRE | External service provider | IN | senior | business | 110.0 h | 0.7 | 2,824 GBP |
| Enablement Engineer | External service provider | IN | senior | business | 10.0 h | 0.1 | 257 GBP |
11.0 people in all. Cost a month: 44,720 GBP.
The monthly budget is 60,000 GBP: this is within it.
Worked from rate card `rate-card`, version `draft-2026-10-13`. **Its figures are placeholders: no price has been set, and this cost is an illustration.**
Converted from USD at 0.79, a rate set on 2026-10-13.
The customer's own staff are counted and not priced. A Role that covers hours wider than a working day is never smaller than the team that can cover them.
## Validation
| Check | What must hold | Verdict | Evidence |
| --- | --- | --- | --- |
| OC-1 | A Priority 1 fault on the platform is acknowledged and responded to within 15 minutes, at any hour. | pass | Capability C-1 covers incident management and operates 24x7. Role R-4 (Operations Engineer) is responsible and operates 24x7. |
| OC-2 | Normal platform operations are restored within 4 hours of a Priority 1 incident occurring. | pass | Capability C-1 operates 24x7 and responds to incidents. |
| OC-3 | The hub network (S-1) is monitored to verify it maintains at least 99.9% availability. | pass | Capability C-2 covers monitoring and operates 24x7, with Role R-4 responsible. The operating model configures native Google Cloud monitoring tools to fulfill this. |
| OC-4 | Changes to the hub network (S-1) are blocked unless approved by the customer platform team. | pass | Capability C-5 covers change enablement. Role R-1 (Platform Lead from the customer team) is accountable and Role R-2 (Platform Engineer from the customer team) is responsible, giving the customer final approval. |
| OC-5 | A reporting package detailing achieved service levels and operational costs is delivered to the customer monthly. | pass | Capability C-15 covers service level management. Role R-6 (Senior SRE) is responsible for producing the monthly reporting package. |
| OC-6 | Platform foundation operations are governed under a shared sourcing model. | pass | Capability C-20 covers supplier management to govern the shared sourcing model. Role R-1 (Platform Lead) is accountable and responsible. |
| OC-7 | Access controls prevent audit logs in the audit log store (S-2) from being deleted for at least seven years. | pass | Capability C-11 covers access management. Role R-5 (Security Operations Engineer) is responsible. |
| OC-8 | Evidence is collected to demonstrate cryptographic encryption keys (S-4) are rotated at least once every 90 days. | pass | Capability C-21 covers regulatory evidence collection. Role R-6 (Senior SRE) is responsible. |
| OC-9 | The Operations Lead role sits in Poland (PL). | pass | Role R-3 (Operations Lead) is defined with country 'PL' (Poland). |
| OC-10 | The Operations Engineer role sits in India (IN). | pass | Role R-4 (Operations Engineer) is defined with country 'IN' (India). |
| OC-11 | The Security Operations Engineer role sits in India (IN). | pass | Role R-5 (Security Operations Engineer) is defined with country 'IN' (India). |
| OC-12 | The Senior SRE role sits in India (IN). | pass | Role R-6 (Senior SRE) is defined with country 'IN' (India). |
| OC-13 | The Enablement Engineer role sits in India (IN). | pass | Role R-7 (Enablement Engineer) is defined with country 'IN' (India). |
| OC-14 | The Platform Lead role sits in the United Kingdom (GB). | pass | Role R-1 (Platform Lead) is defined with country 'GB' (United Kingdom). |
| OC-15 | The Platform Engineer role sits in the United Kingdom (GB). | pass | Role R-2 (Platform Engineer) is defined with country 'GB' (United Kingdom). |
Each scenario is generated from the operating model and traced through it by code.
| Scenario | Result | Path |
| --- | --- | --- |
| Hub network and interconnect (S-1) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Hub network and interconnect (S-1) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Audit log store (S-2) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Audit log store (S-2) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Deployment pipeline (S-3) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Deployment pipeline (S-3) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Encryption keys (S-4) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Encryption keys (S-4) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Condition K-1 must be shown to hold: The platform is available at least 99.9% of each month. | pass | Check OC-3 on C-2 keeps it true. |
| Condition K-2 must be shown to hold: Audit logs cannot be deleted for seven years. | pass | Check OC-7 on C-11 keeps it true. |
| Condition K-3 must be shown to hold: Encryption keys are rotated every 90 days. | pass | Check OC-8 on C-21 keeps it true. |
| External service provider cannot be reached. | pass | Every Capability it is answerable for has a Role in another Party to turn to. |
| A month's cost of operating is compared with the budget of 60,000 GBP. | pass | The cost is 44,720 GBP, within the budget. |
## Findings and open questions
No findings: the solution provides what operating it needs.
Assumed for the customer, and open to change when they answer:
- `intent.exclusions`: Application workloads deployed onto the platform foundation. Workloads deployed onto the platform foundation are excluded; operations is confined to the landing zone foundation itself.
- `operations.change_process`: Changes go through automated pipelines with peer review; changes affecting the hub network require customer platform team approval prior to release.. Changes follow automated deployment pipelines with peer review, with network changes requiring explicit customer platform team approval per REQ-01.
- `service_management.tooling`: Google Cloud native operations tooling (Cloud Monitoring and Cloud Logging), with no separate service management system.. Defaulting to GCP native operations tooling with no separate ITSM system.
## Sources
The Capabilities are from our own capability catalogue, which draws on these frameworks. It cites them and certifies nothing against them.
- ITIL 4: Incident management practice; Monitoring and event management practice; Problem management practice; Service request management practice; Change enablement practice; Release management and Deployment management practices; Service configuration management practice; Service continuity management practice; Capacity and performance management practice; Service level management practice; Supplier management practice
- Site Reliability Engineering: How Google Runs Production Systems: Managing Incidents; Monitoring Distributed Systems; Postmortem Culture: Learning from Failure; Release Engineering; Service Level Objectives
- ISO/IEC 20000-1: Incident management; Service request management; Change management; Service continuity management; Capacity management; Service level management; Supplier management
- Regulation (EU) 2022/2554, the Digital Operational Resilience Act: Article 17, ICT-related incident management process; Article 10, Detection; Article 12, Backup policies and procedures, restoration and recovery procedures and methods; Article 11, Response and recovery; Article 9, Protection and prevention; Article 28, General principles on ICT third-party risk; Article 19, Reporting of major ICT-related incidents
- NIST Cybersecurity Framework: Detect: Continuous Monitoring; Identify: Asset Management; Recover: Incident Recovery Plan Execution; Protect: Identity Management, Authentication, and Access Control; Protect: Platform Security; Detect: Adverse Event Analysis; Respond: Incident Management
- DevOps Research and Assessment (DORA) software delivery metrics: Change failure rate; Deployment frequency and lead time for changes; Deployment frequency; Lead time for changes; Change failure rate; Time to restore service
- ISO/IEC 27001: Access control; Management of technical vulnerabilities; Information security incident management; Compliance with legal and contractual requirements
- FinOps Framework: Allocation; Budgeting; Forecasting; Anomaly Management
- CNCF Platforms White Paper (platform as a product): Platform as a product; Measuring platform success; Platform capabilities; Self-service; Platform as a product
- Team Topologies: Organizing Business and Technology Teams for Fast Flow: Platform team; Thinnest viable platform; X-as-a-Service interaction mode; Enabling team; Facilitating interaction mode
- The SPACE of Developer Productivity: Satisfaction and well-being; Performance; Activity; Communication and collaboration; Efficiency and flow
- DevEx: What Actually Drives Productivity: Feedback loops; Cognitive load; Flow state
**These references have not been checked against the frameworks by a person.** Treat each as a pointer to where to look.

View File

@@ -0,0 +1,39 @@
## Scope statement
This operating scope covers the ongoing operation of the customer's Google Cloud Platform landing zone platform foundation, comprising the hub network and interconnect, audit log store, deployment pipeline, and encryption keys. Operations are delivered under a shared sourcing model between the customer's platform team and a service provider operating from the United Kingdom, Poland, and India, with the customer platform team retaining approval authority over all network changes. The service is operated continuously with 24x7 availability and 24x7 fault response, targeting a 15-minute response time for Priority 1 incidents and a 4-hour recovery time. All operational costs for this service are stated in British Pounds (GBP). Application workloads and hosted services deployed onto the platform foundation are out of scope for operations.
## Information register
20 stated by the customer, 0 answered in conversation, 3 assumed, 10 unknown, 0 not applicable.
| Field | Value | From |
| --- | --- | --- |
| engagement.kind | platform_foundation | the scope spec |
| intent.statement | We want our landing zone run as a shared service from go-live. A provider should respond to faults at every hour, and our own platform team keeps the final say over changes to the network. | the scope spec |
| intent.requirements | [{"id": "REQ-01", "statement": "Our platform team approves every change to the network."}, {"id": "REQ-02", "statement": "We receive a monthly report of service levels and cost."}] | the scope spec |
| intent.business_outcomes | [{"id": "BO-1", "outcome": "New payment products reach customers in weeks rather than months.", "target": "A change reaches production within a day of approval."}, {"id": "BO-2", "outcome": "Payments | the scope spec |
| solution.cloud_provider | gcp | the scope spec |
| solution.items | [{"id": "S-1", "name": "Hub network and interconnect", "kind": "network", "product": "Cloud Interconnect"}, {"id": "S-2", "name": "Audit log store", "kind": "storage", "product": "Cloud Storage"}, {"i | the scope spec |
| solution.conditions | [{"id": "K-1", "on": ["S-1"], "condition": "The platform is available at least 99.9% of each month."}, {"id": "K-2", "on": ["S-2"], "condition": "Audit logs cannot be deleted for seven years."}, {"id" | the scope spec |
| solution.build.duration_days | 36 | the scope spec |
| operations.support_hours | 24x7 | the scope spec |
| operations.availability_hours | 24x7 | the scope spec |
| sourcing.model | shared | the scope spec |
| sourcing.delivery_countries | ["GB", "PL", "IN"] | the scope spec |
| organisation.existing_teams | [{"name": "Platform team", "party": "customer", "hours": "business", "country": "GB"}] | the scope spec |
| service_levels.priority_1_response_minutes | 15 | the scope spec |
| service_levels.recovery_time_hours | 4 | the scope spec |
| compliance.frameworks | ["DORA"] | the scope spec |
| commercial.currency | GBP | the scope spec |
| commercial.monthly_run_budget | 60000 | the scope spec |
| intent.exclusions | ["Application workloads deployed onto the platform foundation"] | assumed |
| operations.change_process | Changes go through automated pipelines with peer review; changes affecting the hub network require customer platform team approval prior to release. | assumed |
| service_management.tooling | Google Cloud native operations tooling (Cloud Monitoring and Cloud Logging), with no separate service management system. | assumed |
### Assumptions
| Id | Assumption | If it is wrong |
| --- | --- | --- |
| ASSUMP-01 | Assumed intent.exclusions is ['Application workloads deployed onto the platform foundation']. Workloads deployed onto the platform foundation are excluded; operations is confined to the landing zone foundation itself. | Changes which Capabilities are designed. |
| ASSUMP-02 | Assumed operations.change_process is Changes go through automated pipelines with peer review; changes affecting the hub network require customer platform team approval prior to release.. Changes follow automated deployment pipelines with peer review, with network changes requiring explicit customer platform team approval per REQ-01. | Changes change enablement and release. |
| ASSUMP-03 | Assumed service_management.tooling is Google Cloud native operations tooling (Cloud Monitoring and Cloud Logging), with no separate service management system.. Defaulting to GCP native operations tooling with no separate ITSM system. | Changes tooling and the processes written for it. |

View File

@@ -0,0 +1,74 @@
## Operational requirements
The operational requirements define what running the landing zone platform foundation must achieve. Every requirement is traced directly to the customer's stated requirements, binding operating fields, or architectural conditions.
### Service level requirements
- **OR-1**: A Priority 1 fault on any landing zone component is acknowledged and responded to within 15 minutes, at any hour (24x7).
*Source: `service_levels.priority_1_response_minutes`, `operations.support_hours`.*
This ensures round-the-clock fault intake and immediate response capabilities across the entire platform foundation without interruption.
- **OR-2**: Normal platform operations are restored within 4 hours of a Priority 1 incident occurring.
*Source: `service_levels.recovery_time_hours`.*
This establishes the mandatory maximum recovery duration for high-impact platform outages to preserve business continuity.
- **OR-3**: The platform foundation maintains an operational availability of at least 99.9% across all hours (24x7) of each calendar month.
*Source: `K-1`, `operations.availability_hours`.*
This bounds total allowable platform downtime across the hub network and core infrastructure to maintain continuous landing zone operations.
### Process requirements
- **OR-4**: Every change to the hub network requires customer platform team approval prior to deployment.
*Source: `REQ-01`.*
This preserves the customer platform team's governance and authoritative control over network modifications while allowing automated deployment pipelines elsewhere.
- **OR-5**: A reporting package detailing achieved service levels and operational costs is delivered to the customer on a monthly basis.
*Source: `REQ-02`.*
This ensures operational transparency, providing monthly verification of SLA performance and run expenditure.
### Organisation requirements
- **OR-6**: Platform foundation operations are delivered under a shared sourcing model between the customer and an external service provider.
*Source: `sourcing.model`.*
This establishes joint delivery boundaries between the customer's internal engineering team and external partner operations.
- **OR-7**: Operational services are delivered exclusively from the United Kingdom, Poland, and India.
*Source: `sourcing.delivery_countries`.*
This restricts operational execution, personnel locations, and data access to the customer-approved jurisdictions.
### Compliance requirements
- **OR-8**: Audit logs stored in the audit log store cannot be deleted and are retained for at least seven years.
*Source: `K-2`.*
This ensures non-repudiation and regulatory compliance for audit evidence by preventing log deletion across the required 7-year retention period.
- **OR-9**: Cryptographic encryption keys are rotated at least once every 90 days.
*Source: `K-3`.*
This enforces key lifecycle governance to limit cryptographic exposure across the landing zone.
---
## What was not stated
The operating specification leaves several operational parameters either assumed or unknown. If answered differently by the customer, these would alter the operational requirements and downstream target operating model design.
### Assumptions
Because execution is running in pre-approved mode, the following operational assumptions have been adopted from the scoping phase:
| Assumption ID | Field | Assumed Value | Impact if Different |
| :--- | :--- | :--- | :--- |
| **ASSUMP-01** | `intent.exclusions` | Application workloads deployed onto the platform foundation are excluded from operations. | If application workloads were included in scope, operational requirements would need to expand to cover application-level SLAs, deployment support, database maintenance, and tenant runtime monitoring. |
| **ASSUMP-02** | `operations.change_process` | Changes go through automated pipelines with peer review; changes affecting the hub network require customer platform team approval prior to release. | If the change process required manual change advisory board (CAB) reviews for all platform changes, release cadence requirements would slow and change review mechanisms would expand. |
| **ASSUMP-03** | `service_management.tooling` | Google Cloud native operations tooling (Cloud Monitoring and Cloud Logging) is used, with no separate external service management system. | If an enterprise ITSM platform (such as ServiceNow or Jira Service Management) were mandated, additional tooling integration and ticket-routing requirements would apply. |
### Unknowns
The operating spec identifies several unstated fields that remain to be settled:
1. **`operations.monitoring` (Monitoring and alerting baseline)**: The customer has not specified pre-existing monitoring dashboards, synthetic probes, or alert routing mechanisms. If proprietary third-party monitoring or customer-managed SIEM integration is required, specific telemetry ingestion requirements must be established.
2. **`solution.inherited_risks` (Inherited risks and unverified controls)**: Unverified architectural dependencies or single points of failure in Cloud Interconnect or Cloud KMS were not stated. Known single points of failure would require additional failover verification requirements.
3. **`solution.build.customer_owned` (Asset ownership during build)**: Exact boundaries between provider-built landing zone templates and customer-managed components were not explicitly split, which could influence operational handover requirements.
4. **`organisation.headcount_limit` (Staffing constraints)**: Headcount caps were not provided. Operational requirements currently reflect service outcomes and availability rather than headcount caps.
5. **`transition.go_live_date` (Service commencement timeline)**: The cutover date from build to live operations is unknown. A compressed transition window would necessitate specific operational readiness acceptance criteria.
6. **Regulatory framework certification (`compliance.frameworks`)**: While the customer noted DORA as an applicable framework, operational frameworks certify nothing; operational capabilities align with DORA resilience and incident response requirements without making formal certification claims.

View File

@@ -0,0 +1,72 @@
# Target Operating Model
## Design rationale
The target operating model establishes a shared sourcing arrangement to securely and cost-effectively operate the cloud platform foundation, balancing out-of-hours coverage against budget limits.
### Division of work
* **Customer platform team (P-1):** Retains full product ownership, accountability for the platform roadmap, and final approval authority for any changes to the hub network. This ensures the operating model explicitly satisfies the requirement that the internal team retains the final say over the network.
* **External service provider (P-2):** Operates a shared "Platform" team and an "Enabling" team. They hold round-the-clock responsibility for incident response, monitoring, and security telemetry, as well as maintaining the pipeline capabilities. The enablement team interacts with the customer through facilitating modes until the foundational pipelines are successfully adopted.
### Roles and Locations
Due to strict budgetary constraints (£60,000 per month limit), the geographical placement of the provider roles balances coverage, seniority, and cost:
* **Operations Lead (R-3):** Based in Poland (PL) working business hours. This maintains a leadership footprint closer to the customer's UK time zone to facilitate effective governance and escalation.
* **Operations Engineer (R-4) & Security Operations Engineer (R-5):** Based in India (IN) providing 24x7 cover. Operating a 24x7 rota requires a minimum roster of personnel. Placing these roles in India achieves the required round-the-clock incident response (within 15 minutes) while keeping the heavy cost multipliers of 24x7 staffing comfortably within the budget.
* **Senior SRE (R-6) & Enablement Engineer (R-7):** Based in India (IN) during business hours. Initially drafted in Poland, these senior roles were shifted to India to ensure the total monthly run cost remains below the budget.
* **Customer Roles (R-1, R-2):** Based in the UK (GB). Customer headcount does not contribute to the operational run cost.
### Checks and Native Capabilities
Previous versions of this design incorrectly relied on architectural findings for standard operational tasks (e.g., configuring alerts). This has been rectified. The operating model configures native Google Cloud tooling (Cloud Monitoring and Cloud Logging) to fulfill these checks internally without requiring external architectural modifications.
## Assumptions
As this model was generated with pre-approved execution, the following assumptions were made:
* **GCP Tooling:** Operational monitoring, dashboarding, and alerting are entirely achievable within the bounds of the existing native GCP solution components and require no special workarounds.
* **Effort Sizing:** The minimum roster size required to staff a 24x7 position provides sufficient aggregate effort to absorb the estimated monthly capability load without needing additional supplementary FTEs.
* **Client approvals:** The customer has verified that shifting several senior provider engineering roles to India does not violate any unstated data residency or collaboration constraints beyond the explicit list of allowed countries.
## Responsibility matrix
| Capability | Platform Lead | Platform Engineer | Operations Lead | Operations Engineer | Security Operations Engineer | Senior SRE | Enablement Engineer |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Incident management (C-1) | | C | A | R | | | |
| Monitoring and alerting (C-2) | | C | A | R | | | |
| Problem management (C-3) | | C | A | | | R | |
| Service request fulfilment (C-4) | | C | A | R | | | |
| Change enablement (C-5) | A | R | | | | C | |
| Release and deployment (C-6) | | C | A | | | R | |
| Configuration management (C-7) | | C | A | | | R | |
| Backup and recovery (C-8) | | C | A | | | R | |
| Service continuity (C-9) | | C | A | | | R | |
| Capacity and performance management (C-10) | | C | A | | | R | |
| Access management (C-11) | | C | A | | R | | |
| Vulnerability and patch management (C-12) | | C | A | | | R | |
| Security monitoring and response (C-13) | | C | A | | R | | |
| Cost management (C-14) | C | | A | | | R | |
| Service level management (C-15) | C | | A | | | R | |
| Platform product management (C-16) | A/R | | | | | | |
| Self-service and golden paths (C-17) | | C | A | | | R | |
| Enablement (C-18) | | C | A | | | | R |
| Delivery and developer experience measurement (C-19) | C | | A | | | R | |
| Supplier management (C-20) | A/R | | | | | | |
| Regulatory evidence and reporting (C-21) | | C | A | | | R | |
A accountable, R responsible, C consulted, I informed.
## Staffing and cost
| Role | Party | Country | Level | Hours | Effort a month | People | Cost a month |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Platform Lead | Customer platform team | GB | lead | business | 16.0 h | 0.1 | not priced |
| Platform Engineer | Customer platform team | GB | engineer | business | 10.0 h | 0.1 | not priced |
| Operations Lead | External service provider | PL | lead | business | 0.0 h | 0.0 | 0 GBP |
| Operations Engineer | External service provider | IN | engineer | 24x7 | 59.0 h | 5.0 | 17,034 GBP |
| Security Operations Engineer | External service provider | IN | senior | 24x7 | 34.0 h | 5.0 | 24,605 GBP |
| Senior SRE | External service provider | IN | senior | business | 110.0 h | 0.7 | 2,824 GBP |
| Enablement Engineer | External service provider | IN | senior | business | 10.0 h | 0.1 | 257 GBP |
11.0 people in all. Cost a month: 44,720 GBP.
The monthly budget is 60,000 GBP: this is within it.
Worked from rate card `rate-card`, version `draft-2026-10-13`. **Its figures are placeholders: no price has been set, and this cost is an illustration.**
Converted from USD at 0.79, a rate set on 2026-10-13.
The customer's own staff are counted and not priced. A Role that covers hours wider than a working day is never smaller than the team that can cover them.

View File

@@ -0,0 +1,50 @@
# Check verdicts
## Passed checks
* **OC-1**: Passed. Capability C-1 covers incident management and operates 24x7. Role R-4 (Operations Engineer) is responsible and operates 24x7.
* **OC-2**: Passed. Capability C-1 operates 24x7 and responds to incidents.
* **OC-3**: Passed. Capability C-2 covers monitoring and operates 24x7, with Role R-4 responsible. The operating model configures native Google Cloud monitoring tools to fulfill this internally.
* **OC-4**: Passed. Capability C-5 covers change enablement. Role R-1 (Platform Lead from the customer team) is accountable and Role R-2 (Platform Engineer from the customer team) is responsible, giving the customer final approval.
* **OC-5**: Passed. Capability C-15 covers service level management. Role R-6 (Senior SRE) is responsible for producing the monthly reporting package.
* **OC-6**: Passed. Capability C-20 covers supplier management to govern the shared sourcing model. Role R-1 (Platform Lead) is accountable and responsible.
* **OC-7**: Passed. Capability C-11 covers access management. Role R-5 (Security Operations Engineer) is responsible.
* **OC-8**: Passed. Capability C-21 covers regulatory evidence collection. Role R-6 (Senior SRE) is responsible.
* **OC-9**: Passed. Role R-3 (Operations Lead) sits in Poland (PL).
* **OC-10**: Passed. Role R-4 (Operations Engineer) sits in India (IN).
* **OC-11**: Passed. Role R-5 (Security Operations Engineer) sits in India (IN).
* **OC-12**: Passed. Role R-6 (Senior SRE) sits in India (IN).
* **OC-13**: Passed. Role R-7 (Enablement Engineer) sits in India (IN).
* **OC-14**: Passed. Role R-1 (Platform Lead) sits in the United Kingdom (GB).
* **OC-15**: Passed. Role R-2 (Platform Engineer) sits in the United Kingdom (GB).
## Failed checks
None.
## Not verifiable checks
None.
## Assumptions
* No further assumptions were made during this validation run.
## Overall verdict
**Pass**: The target operating model successfully meets all operational requirements and conditions.
## Scenario walk-throughs
Each scenario is generated from the operating model and traced through it by code.
| Scenario | Result | Path |
| --- | --- | --- |
| Hub network and interconnect (S-1) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Hub network and interconnect (S-1) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Audit log store (S-2) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Audit log store (S-2) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Deployment pipeline (S-3) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Deployment pipeline (S-3) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Encryption keys (S-4) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Encryption keys (S-4) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Condition K-1 must be shown to hold: The platform is available at least 99.9% of each month. | pass | Check OC-3 on C-2 keeps it true. |
| Condition K-2 must be shown to hold: Audit logs cannot be deleted for seven years. | pass | Check OC-7 on C-11 keeps it true. |
| Condition K-3 must be shown to hold: Encryption keys are rotated every 90 days. | pass | Check OC-8 on C-21 keeps it true. |
| External service provider cannot be reached. | pass | Every Capability it is answerable for has a Role in another Party to turn to. |
| A month's cost of operating is compared with the budget of 60,000 GBP. | pass | The cost is 44,720 GBP, within the budget. |

View File

@@ -0,0 +1,109 @@
# Capability Model
This capability model outlines what the operating organisation must be able to do to meet the operational requirements. It assigns the capabilities from the standard catalogue to specific items of the platform foundation and estimates steady-state effort. Who specifically performs each capability is not decided here.
## Capabilities
| ID | Capability | Coverage (Items) | Operational Requirements | Hours | Monthly Effort (hrs) | Purpose / Need |
|---|---|---|---|---|---|---|
| **C-1** | Incident management | S-1, S-2, S-3, S-4 | OR-1, OR-2 | 24x7 | 30 | Restore degraded services. Needed for fast fault handling. |
| **C-2** | Monitoring and alerting | S-1, S-2, S-3, S-4 | OR-1, OR-3 | 24x7 | 20 | Know the state of components. Needed to detect faults and track uptime. |
| **C-3** | Problem management | S-1, S-2, S-3, S-4 | OR-2 | business | 10 | Find why incidents happen. Needed to prevent major failures. |
| **C-4** | Service request fulfilment | S-1, S-2, S-3, S-4 | OR-6 | business | 20 | Handle user requests. Needed for the shared operating model. |
| **C-5** | Change enablement | S-1, S-2, S-3, S-4 | OR-4, OR-10 | business | 10 | Manage changes. Needed for customer approvals on network changes. |
| **C-6** | Release and deployment | S-1, S-2, S-3, S-4 | OR-10 | business | 20 | Put approved changes into service. Needed to hit the 1-day deployment target. |
| **C-7** | Configuration management | S-1, S-2, S-3, S-4 | OR-8, OR-9 | business | 15 | Track configs. Needed for key rotation and audit log policies. |
| **C-8** | Backup and recovery | S-2, S-4 | OR-2, OR-8 | business | 10 | Keep copies of logs/keys. Needed for recovery and log retention. |
| **C-9** | Service continuity | S-1, S-2, S-3, S-4 | OR-2, OR-3 | business | 10 | Restore services after loss. Needed for the 4-hour recovery SLA. |
| **C-10** | Capacity and performance management | S-1, S-2, S-3, S-4 | OR-3 | business | 10 | Manage limits. Needed for 99.9% uptime. |
| **C-11** | Access management | S-1, S-2, S-3, S-4 | OR-6, OR-7 | business | 15 | Give/revoke access. Needed for cross-country shared model boundaries. |
| **C-12** | Vulnerability and patch management | S-1, S-2, S-3, S-4 | OR-9 | business | 15 | Fix weaknesses. Needed to secure components and rotate keys safely. |
| **C-13** | Security monitoring and response | S-1, S-2, S-3, S-4 | OR-1, OR-8 | 24x7 | 40 | Detect misuse. Needed to protect the audit logs and fast response to events. |
| **C-14** | Cost management | S-1, S-2, S-3, S-4 | OR-5 | business | 10 | Track spend. Needed for the monthly cost report. |
| **C-15** | Service level management | S-1, S-2, S-3, S-4 | OR-3, OR-5 | business | 10 | Measure SLAs. Needed for 99.9% tracking and reporting. |
| **C-16** | Platform product management | S-1, S-2, S-3, S-4 | OR-6 | business | 15 | Manage roadmap based on developer friction. Needed to retain engineers. |
| **C-17** | Self-service and golden paths | S-1, S-2, S-3, S-4 | OR-10 | business | 20 | Automate common tasks. Needed for rapid product delivery. |
| **C-18** | Enablement | S-1, S-2, S-3, S-4 | OR-10 | business | 20 | Help teams adopt the platform. Needed for fast flow of value. |
| **C-19** | Delivery and developer experience measurement | S-1, S-2, S-3, S-4 | OR-10 | business | 10 | Measure flow at all layers. Needed to prove fast change delivery. |
| **C-20** | Supplier management | S-1, S-2, S-3, S-4 | OR-6, OR-7 | business | 10 | Hold suppliers to agreements. Needed for the shared model. |
| **C-21** | Regulatory evidence and reporting | S-1, S-2, S-3, S-4 | OR-8 | business | 10 | Retain compliance evidence. Needed to ensure logs remain undeleted for 7 years. |
*Note: The rate card is currently in draft. Any cost figures or staffing structures derived from these estimations will act as placeholders.*
## Assumptions
As this execution is pre-approved, unknown values were accommodated using standard capability assumptions:
- The customer platform team retains network change approvals, and applications running on the platform are out of scope.
- Changes traverse a deployment pipeline utilizing peer review.
- The cloud provider's native operations tooling is employed in lieu of a separate ITSM system.
- Unspecified bounds around headcount limits, monitoring integrations, or customer build components do not currently restrict operational coverage.
## Sources
The capabilities within this model draw on concepts from several recognized industry frameworks. These are cited as referenced by the catalogue parts below.
*Please note: The capability catalogue draws on these frameworks but claims no official certification or compliance against them. All references provided below are currently **unverified** as documented by the standard catalogue register.*
- **ITIL 4**
- Incident management practice
- Monitoring and event management practice
- Problem management practice
- Service request management practice
- Change enablement practice
- Release management and Deployment management practices
- Service configuration management practice
- Service continuity management practice
- Capacity and performance management practice
- Service level management practice
- Supplier management practice
- **Site Reliability Engineering: How Google Runs Production Systems (Google SRE)**
- Managing Incidents
- Monitoring Distributed Systems
- Postmortem Culture: Learning from Failure
- Release Engineering
- Service Level Objectives
- **ISO/IEC 20000-1**
- Incident management
- Service request management
- Change management
- Service continuity management
- Capacity management
- Service level management
- Supplier management
- **Regulation (EU) 2022/2554, the Digital Operational Resilience Act (EU DORA)**
- Article 9, Protection and prevention
- Article 10, Detection
- Article 11, Response and recovery
- Article 12, Backup policies and procedures, restoration and recovery procedures and methods
- Article 17, ICT-related incident management process
- Article 19, Reporting of major ICT-related incidents
- Article 28, General principles on ICT third-party risk
- **NIST Cybersecurity Framework (NIST CSF 2.0)**
- Identify: Asset Management
- Protect: Identity Management, Authentication, and Access Control
- Protect: Platform Security
- Detect: Continuous Monitoring
- Detect: Adverse Event Analysis; Respond: Incident Management
- Recover: Incident Recovery Plan Execution
- **ISO/IEC 27001**
- Access control
- Management of technical vulnerabilities
- Information security incident management
- Compliance with legal and contractual requirements
- **DevOps Research and Assessment (DORA) software delivery metrics**
- Change failure rate
- Deployment frequency and lead time for changes
- Deployment frequency; Lead time for changes; Change failure rate; Time to restore service
- **FinOps Framework**
- Allocation; Budgeting; Forecasting; Anomaly Management
- **CNCF Platforms White Paper (platform as a product)**
- Platform as a product; Measuring platform success
- Platform capabilities; Self-service
- Platform as a product
- **Team Topologies: Organizing Business and Technology Teams for Fast Flow**
- Platform team; Thinnest viable platform
- X-as-a-Service interaction mode
- Enabling team; Facilitating interaction mode
- **The SPACE of Developer Productivity**
- Satisfaction and well-being; Performance; Activity; Communication and collaboration; Efficiency and flow
- **DevEx: What Actually Drives Productivity**
- Feedback loops; Cognitive load; Flow state

View File

@@ -0,0 +1,237 @@
# Operating model guide: Google Cloud landing zone platform foundation
## Summary
This guide sets out the target operating model for the Google Cloud landing zone platform foundation, covering the hub network and interconnect, audit log store, deployment pipeline, and encryption keys. Operational responsibility is shared between the customer and an external service provider across the United Kingdom, Poland, and India. The customer platform team steers platform product direction, developer flow measurement, and retains approval authority over all network changes, while the service provider operations team runs day-to-day operations, maintenance, and incident management. Fault response and platform availability are maintained around the clock (24x7), while change governance, service continuity, and lifecycle management are conducted during business hours.
Readers should look first at the findings and open questions, particularly the catalogue assumptions regarding cloud-native operations tooling, change pipeline procedures, and the exclusion of workload applications, as well as the draft status of the rate card. All scenario walk-throughs and validation checks pass successfully within the allocated operational budget.
### Key figures
- Operated: 4 Items, with 3 Conditions to keep true.
- Faults responded to: 24x7 hours.
- Capabilities: 21.
- People: 11.4. Cost a month: 47,603 GBP, from rate card version `draft-2026-10-13` (placeholder prices: an illustration).
- Scenario walk-throughs: 13 of 13 pass.
- Business outcomes: 3, 3 shown by a Measure.
- Findings for whoever owns the design: 0.
## Business outcomes
**BO-1: New payment products reach customers in weeks rather than months.** Target: A change reaches production within a day of approval..
| Measure | Layer | Framework | Target | Measured by | Answerable |
| --- | --- | --- | --- | --- | --- |
| Deployment frequency | outcome | dora-metrics | A change reaches production within a day of approval. | Delivery and developer experience measurement | Platform Product Manager |
| Friction and cognitive load | leading | devex-framework | | Delivery and developer experience measurement | Platform Product Manager |
**BO-2: Payments are never unavailable for long.** Target: Restored within 4 hours..
| Measure | Layer | Framework | Target | Measured by | Answerable |
| --- | --- | --- | --- | --- | --- |
| Time to restore service | outcome | dora-metrics | Restored within 4 hours. | Incident management | Operations Lead |
**BO-3: We keep the platform engineers we hire.**
| Measure | Layer | Framework | Target | Measured by | Answerable |
| --- | --- | --- | --- | --- | --- |
| Developer satisfaction | guardrail | space-framework | | Delivery and developer experience measurement | Platform Product Manager |
## What is operated
| Item | What it is | Kind | Product | From |
| --- | --- | --- | --- | --- |
| S-1 | Hub network and interconnect | network | Cloud Interconnect | |
| S-2 | Audit log store | storage | Cloud Storage | |
| S-3 | Deployment pipeline | pipeline | Cloud Build | |
| S-4 | Encryption keys | keys | Cloud KMS | |
What must be kept true while it is run:
| Condition | On | What must hold | From |
| --- | --- | --- | --- |
| K-1 | S-1 | The platform is available at least 99.9% of each month. | |
| K-2 | S-2 | Audit logs cannot be deleted for seven years. | |
| K-3 | S-4 | Encryption keys are rotated every 90 days. | |
## Operational requirements
| Requirement | Kind | What must be achieved | From |
| --- | --- | --- | --- |
| OR-1 | service_level | A fault on any Item is acknowledged within 15 minutes, at any hour. | service_levels.priority_1_response_minutes, operations.support_hours |
| OR-2 | service_level | Service is restored within 4 hours of a major failure. | service_levels.recovery_time_hours, BO-2 |
| OR-3 | service_level | The platform is available at least 99.9% of each month. | K-1 |
| OR-4 | process | Every change to the network requires customer platform team approval prior to implementation. | REQ-01 |
| OR-5 | process | A report of service levels and operating costs is provided monthly. | REQ-02 |
| OR-6 | organisation | The platform foundation is operated under a shared service model between the customer and an external service provider. | sourcing.model |
| OR-7 | organisation | Operational delivery is performed exclusively from the United Kingdom, Poland, and India. | sourcing.delivery_countries |
| OR-8 | compliance | Audit logs cannot be deleted for seven years. | K-2 |
| OR-9 | process | Encryption keys are rotated every 90 days. | K-3 |
| OR-10 | service_level | An approved change reaches production within one day of approval. | BO-1 |
## Capabilities
| Capability | What it is | Serves | Covers | Hours | Effort a month |
| --- | --- | --- | --- | --- | --- |
| C-1 | Incident management | OR-1, OR-2 | S-1, S-2, S-3, S-4 | 24x7 | 30 h |
| C-2 | Monitoring and alerting | OR-1, OR-3 | S-1, S-2, S-3, S-4 | 24x7 | 20 h |
| C-3 | Problem management | OR-2 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-4 | Service request fulfilment | OR-6 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-5 | Change enablement | OR-4, OR-10 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-6 | Release and deployment | OR-10 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-7 | Configuration management | OR-8, OR-9 | S-1, S-2, S-3, S-4 | business | 15 h |
| C-8 | Backup and recovery | OR-2, OR-8 | S-2, S-4 | business | 10 h |
| C-9 | Service continuity | OR-2, OR-3 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-10 | Capacity and performance management | OR-3 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-11 | Access management | OR-6, OR-7 | S-1, S-2, S-3, S-4 | business | 15 h |
| C-12 | Vulnerability and patch management | OR-9 | S-1, S-2, S-3, S-4 | business | 15 h |
| C-13 | Security monitoring and response | OR-1, OR-8 | S-1, S-2, S-3, S-4 | 24x7 | 40 h |
| C-14 | Cost management | OR-5 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-15 | Service level management | OR-3, OR-5 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-16 | Platform product management | OR-6 | S-1, S-2, S-3, S-4 | business | 15 h |
| C-17 | Self-service and golden paths | OR-10 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-18 | Enablement | OR-10 | S-1, S-2, S-3, S-4 | business | 20 h |
| C-19 | Delivery and developer experience measurement | OR-10 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-20 | Supplier management | OR-6, OR-7 | S-1, S-2, S-3, S-4 | business | 10 h |
| C-21 | Regulatory evidence and reporting | OR-8 | S-1, S-2, S-3, S-4 | business | 10 h |
## Who does what
| Role | Party | Country | Level | Hours |
| --- | --- | --- | --- | --- |
| Platform Product Manager | Customer (customer) | GB | lead | business |
| Platform Engineer | Customer (customer) | GB | senior | business |
| Operations Engineer | Service Provider (provider) | IN | engineer | 24x7 |
| Operations Lead | Service Provider (provider) | IN | lead | business |
| Security Operations Analyst | Service Provider (provider) | IN | senior | 24x7 |
| Platform Ops Specialist | Service Provider (provider) | PL | senior | business |
The teams, as Team Topologies names their kinds:
| Team | Kind | Party | Roles | Purpose |
| --- | --- | --- | --- | --- |
| Customer Platform Team | platform | Customer | Platform Product Manager, Platform Engineer | Owns the platform roadmap, retains network change approvals, and measures developer flow. |
| Provider Operations Team | platform | Service Provider | Operations Engineer, Operations Lead, Security Operations Analyst, Platform Ops Specialist | Provides 24x7 monitoring, incident response, and day-to-day operations of the platform foundation. |
How the teams work together:
| From | To | Mode | Until | Purpose |
| --- | --- | --- | --- | --- |
| Provider Operations Team | Customer Platform Team | x-as-a-service | | Provider requests network change approvals from the Customer Platform Team. |
| Customer Platform Team | Provider Operations Team | x-as-a-service | | Customer relies on Provider Operations Team for 24x7 incident response and monitoring. |
| Provider Operations Team | Customer Platform Team | collaboration | Month 3 after go-live, shifting to x-as-a-service. | Handover of runbooks and initial operational processes. |
| Capability | Platform Product Manager | Platform Engineer | Operations Engineer | Operations Lead | Security Operations Analyst | Platform Ops Specialist |
| --- | --- | --- | --- | --- | --- | --- |
| Incident management (C-1) | | C | R | A | | |
| Monitoring and alerting (C-2) | | C | R | A | | |
| Problem management (C-3) | | C | | A | | R |
| Service request fulfilment (C-4) | | C | R | A | | |
| Change enablement (C-5) | | A/R | | | | R |
| Release and deployment (C-6) | | A/R | | | | R |
| Configuration management (C-7) | | C | | A | | R |
| Backup and recovery (C-8) | | C | | A | | R |
| Service continuity (C-9) | | C | | A | | R |
| Capacity and performance management (C-10) | | C | | A/R | | |
| Access management (C-11) | | C | | A/R | | |
| Vulnerability and patch management (C-12) | | C | | | A/R | |
| Security monitoring and response (C-13) | | C | | | A/R | |
| Cost management (C-14) | C | | | A/R | | |
| Service level management (C-15) | C | | | A/R | | |
| Platform product management (C-16) | A/R | | | C | | |
| Self-service and golden paths (C-17) | | A/R | | | | R |
| Enablement (C-18) | | A/R | | | | R |
| Delivery and developer experience measurement (C-19) | A/R | | | C | | |
| Supplier management (C-20) | A/R | | | R | | |
| Regulatory evidence and reporting (C-21) | | C | | A | | R |
A accountable, R responsible, C consulted, I informed.
## Staffing and cost
| Role | Party | Country | Level | Hours | Effort a month | People | Cost a month |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Platform Product Manager | Customer | GB | lead | business | 30.0 h | 0.2 | not priced |
| Platform Engineer | Customer | GB | senior | business | 35.0 h | 0.3 | not priced |
| Operations Engineer | Service Provider | IN | engineer | 24x7 | 70.0 h | 5.0 | 17,034 GBP |
| Operations Lead | Service Provider | IN | lead | business | 50.0 h | 0.3 | 1,876 GBP |
| Security Operations Analyst | Service Provider | IN | senior | 24x7 | 55.0 h | 5.0 | 24,605 GBP |
| Platform Ops Specialist | Service Provider | PL | senior | business | 90.0 h | 0.6 | 4,088 GBP |
11.4 people in all. Cost a month: 47,603 GBP.
The monthly budget is 60,000 GBP: this is within it.
Worked from rate card `rate-card`, version `draft-2026-10-13`. **Its figures are placeholders: no price has been set, and this cost is an illustration.**
Converted from USD at 0.79, a rate set on 2026-10-13.
The customer's own staff are counted and not priced. A Role that covers hours wider than a working day is never smaller than the team that can cover them.
## Validation
| Check | What must hold | Verdict | Evidence |
| --- | --- | --- | --- |
| OC-1 | A fault on the hub network or any other item is acknowledged within 15 minutes at any hour. | pass | C-1 handles incident management 24x7, with a 24x7 Operations Engineer responsible, enabling acknowledgement of faults within 15 minutes at any hour. The cloud provider's native tooling (via C-2) supports configuring alerts without requiring a separate finding. |
| OC-2 | Service is restored within 4 hours of a major failure on the platform. | pass | C-1 manages incident resolution with a 24x7 Operations Engineer responsible, allowing restoration of service within 4 hours of a major failure regardless of when the failure occurs. |
| OC-3 | The platform availability is tracked to ensure at least 99.9% uptime each month. | pass | C-15 tracks and gathers metrics on availability to ensure the 99.9% target is tracked, with business-hours roles accountable and responsible. |
| OC-4 | Capacity limits for the platform interconnect are managed to support 99.9% availability. | pass | C-10 reviews capacity trends for the interconnect to assure 99.9% availability, managed by the Operations Lead. |
| OC-5 | Every change to the hub network requires explicit approval from the customer platform team. | pass | C-5 coordinates network change approvals with the customer, and a Customer Platform Engineer is accountable and responsible for it, ensuring explicit approval from the customer platform team. |
| OC-6 | A consolidated monthly report includes service levels and operating costs. | pass | C-15 compiles the monthly operational report, pulling in cost data tracked by C-14, both managed by the Operations Lead. |
| OC-7 | The shared service model responsibilities between customer and external service provider are governed. | pass | C-20 coordinates responsibilities across shared delivery borders, with roles from both the customer and provider teams responsible. |
| OC-8a | Role performs operational delivery from India. | pass | R-3 (Operations Engineer) is based in India, satisfying the geographical requirement. |
| OC-8b | Role performs operational delivery from India. | pass | R-4 (Operations Lead) is based in India, satisfying the geographical requirement. |
| OC-8c | Role performs operational delivery from India. | pass | R-5 (Security Operations Analyst) is based in India, satisfying the geographical requirement. |
| OC-8d | Role performs operational delivery from Poland. | pass | R-6 (Platform Ops Specialist) is based in Poland, satisfying the geographical requirement. |
| OC-8e | Role performs operational delivery from the UK. | pass | R-1 (Platform Product Manager) is based in the UK, satisfying the geographical requirement. |
| OC-8f | Role performs operational delivery from the UK. | pass | R-2 (Platform Engineer) is based in the UK, satisfying the geographical requirement. |
| OC-9 | Audit log storage configuration ensures logs cannot be deleted for seven years. | pass | C-7 maintains audit log retention settings, with the Platform Ops Specialist responsible, ensuring logs cannot be deleted for seven years. |
| OC-10 | Retention policies on Cloud Storage for audit logs are verified to prevent deletion for seven years. | pass | C-7 maintains and verifies retention settings on Cloud Storage for audit logs, satisfying the seven-year retention condition. |
| OC-11 | Encryption key configurations are audited to ensure rotation every 90 days. | pass | C-7 audits configuration drift and manages key rotation schedules, ensuring encryption keys are rotated every 90 days. |
| OC-12 | Cloud KMS encryption keys are verified to rotate every 90 days. | pass | C-7 manages key rotation schedules for the Cloud KMS encryption keys, verifying they rotate every 90 days. |
| OC-13 | Deployment pipelines ensure an approved change reaches production within one day. | pass | C-6 oversees deployment pipelines to ensure approved changes reach production within a day, with customer and provider roles responsible. |
| OC-14 | Backups of audit logs and encryption keys are successfully tested for restoration. | pass | C-8 maintains backups of logs and keys and successfully tests restore procedures, managed by the Platform Ops Specialist. |
Each scenario is generated from the operating model and traced through it by code.
| Scenario | Result | Path |
| --- | --- | --- |
| Hub network and interconnect (S-1) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Hub network and interconnect (S-1) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Audit log store (S-2) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Audit log store (S-2) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Deployment pipeline (S-3) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Deployment pipeline (S-3) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Encryption keys (S-4) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Encryption keys (S-4) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Condition K-1 must be shown to hold: The platform is available at least 99.9% of each month. | pass | Check OC-4 on C-10 keeps it true. |
| Condition K-2 must be shown to hold: Audit logs cannot be deleted for seven years. | pass | Check OC-10 on C-7 keeps it true. |
| Condition K-3 must be shown to hold: Encryption keys are rotated every 90 days. | pass | Check OC-12 on C-7 keeps it true. |
| Service Provider cannot be reached. | pass | Every Capability it is answerable for has a Role in another Party to turn to. |
| A month's cost of operating is compared with the budget of 60,000 GBP. | pass | The cost is 47,603 GBP, within the budget. |
## Findings and open questions
No findings: the solution provides what operating it needs.
Assumed for the customer, and open to change when they answer:
- `intent.exclusions`: Workload applications running on the platform foundation, Network change approvals (retained by customer platform team). Assumed from the platform foundation scope: workload applications running on the platform are out of scope, and customer platform team retains network change approvals.
- `operations.change_process`: Changes go through a deployment pipeline with peer review, with network changes requiring customer platform team approval.. Catalogue default adapted to customer requirement REQ-01: changes go through deployment pipelines with peer review, with customer platform team retaining approval over network changes.
- `service_management.tooling`: The Cloud provider's own operations tooling, with no separate system.. Catalogue default: Cloud provider's native operations tooling with no separate system.
## Sources
The Capabilities are from our own capability catalogue, which draws on these frameworks. It cites them and certifies nothing against them.
- ITIL 4: Incident management practice; Monitoring and event management practice; Problem management practice; Service request management practice; Change enablement practice; Release management and Deployment management practices; Service configuration management practice; Service continuity management practice; Capacity and performance management practice; Service level management practice; Supplier management practice
- Site Reliability Engineering: How Google Runs Production Systems: Managing Incidents; Monitoring Distributed Systems; Postmortem Culture: Learning from Failure; Release Engineering; Service Level Objectives
- ISO/IEC 20000-1: Incident management; Service request management; Change management; Service continuity management; Capacity management; Service level management; Supplier management
- Regulation (EU) 2022/2554, the Digital Operational Resilience Act: Article 17, ICT-related incident management process; Article 10, Detection; Article 12, Backup policies and procedures, restoration and recovery procedures and methods; Article 11, Response and recovery; Article 9, Protection and prevention; Article 28, General principles on ICT third-party risk; Article 19, Reporting of major ICT-related incidents
- NIST Cybersecurity Framework: Detect: Continuous Monitoring; Identify: Asset Management; Recover: Incident Recovery Plan Execution; Protect: Identity Management, Authentication, and Access Control; Protect: Platform Security; Detect: Adverse Event Analysis; Respond: Incident Management
- DevOps Research and Assessment (DORA) software delivery metrics: Change failure rate; Deployment frequency and lead time for changes; Deployment frequency; Lead time for changes; Change failure rate; Time to restore service
- ISO/IEC 27001: Access control; Management of technical vulnerabilities; Information security incident management; Compliance with legal and contractual requirements
- FinOps Framework: Allocation; Budgeting; Forecasting; Anomaly Management
- CNCF Platforms White Paper (platform as a product): Platform as a product; Measuring platform success; Platform capabilities; Self-service; Platform as a product
- Team Topologies: Organizing Business and Technology Teams for Fast Flow: Platform team; Thinnest viable platform; X-as-a-Service interaction mode; Enabling team; Facilitating interaction mode
- The SPACE of Developer Productivity: Satisfaction and well-being; Performance; Activity; Communication and collaboration; Efficiency and flow
- DevEx: What Actually Drives Productivity: Feedback loops; Cognitive load; Flow state
**These references have not been checked against the frameworks by a person.** Treat each as a pointer to where to look.

View File

@@ -0,0 +1,39 @@
# Scope statement
This operating scope covers the Google Cloud landing zone platform foundation, comprising the hub network and interconnect, audit log store, deployment pipeline, and encryption keys. The platform is operated under a shared service model jointly by a service provider delivering from the United Kingdom, Poland, and India, and the customer's platform team based in the United Kingdom. Fault response support is provided 24x7, aligning with continuous 24x7 platform availability. All operating costs are stated in GBP. Out of scope for provider operations are workload applications hosted on the platform foundation and final approval authority for network changes, which is retained by the customer platform team.
## Information register
20 stated by the customer, 0 answered in conversation, 3 assumed, 10 unknown, 0 not applicable.
| Field | Value | From |
| --- | --- | --- |
| engagement.kind | platform_foundation | the scope spec |
| intent.statement | We want our landing zone run as a shared service from go-live. A provider should respond to faults at every hour, and our own platform team keeps the final say over changes to the network. | the scope spec |
| intent.requirements | [{"id": "REQ-01", "statement": "Our platform team approves every change to the network."}, {"id": "REQ-02", "statement": "We receive a monthly report of service levels and cost."}] | the scope spec |
| intent.business_outcomes | [{"id": "BO-1", "outcome": "New payment products reach customers in weeks rather than months.", "target": "A change reaches production within a day of approval."}, {"id": "BO-2", "outcome": "Payments | the scope spec |
| solution.cloud_provider | gcp | the scope spec |
| solution.items | [{"id": "S-1", "name": "Hub network and interconnect", "kind": "network", "product": "Cloud Interconnect"}, {"id": "S-2", "name": "Audit log store", "kind": "storage", "product": "Cloud Storage"}, {"i | the scope spec |
| solution.conditions | [{"id": "K-1", "on": ["S-1"], "condition": "The platform is available at least 99.9% of each month."}, {"id": "K-2", "on": ["S-2"], "condition": "Audit logs cannot be deleted for seven years."}, {"id" | the scope spec |
| solution.build.duration_days | 36 | the scope spec |
| operations.support_hours | 24x7 | the scope spec |
| operations.availability_hours | 24x7 | the scope spec |
| sourcing.model | shared | the scope spec |
| sourcing.delivery_countries | ["GB", "PL", "IN"] | the scope spec |
| organisation.existing_teams | [{"name": "Platform team", "party": "customer", "hours": "business", "country": "GB"}] | the scope spec |
| service_levels.priority_1_response_minutes | 15 | the scope spec |
| service_levels.recovery_time_hours | 4 | the scope spec |
| compliance.frameworks | ["DORA"] | the scope spec |
| commercial.currency | GBP | the scope spec |
| commercial.monthly_run_budget | 60000 | the scope spec |
| intent.exclusions | ["Workload applications running on the platform foundation", "Network change approvals (retained by customer platform team)"] | assumed |
| operations.change_process | Changes go through a deployment pipeline with peer review, with network changes requiring customer platform team approval. | assumed |
| service_management.tooling | The Cloud provider's own operations tooling, with no separate system. | assumed |
### Assumptions
| Id | Assumption | If it is wrong |
| --- | --- | --- |
| ASSUMP-01 | Assumed intent.exclusions is ['Workload applications running on the platform foundation', 'Network change approvals (retained by customer platform team)']. Assumed from the platform foundation scope: workload applications running on the platform are out of scope, and customer platform team retains network change approvals. | Changes which Capabilities are designed. |
| ASSUMP-02 | Assumed operations.change_process is Changes go through a deployment pipeline with peer review, with network changes requiring customer platform team approval.. Catalogue default adapted to customer requirement REQ-01: changes go through deployment pipelines with peer review, with customer platform team retaining approval over network changes. | Changes change enablement and release. |
| ASSUMP-03 | Assumed service_management.tooling is The Cloud provider's own operations tooling, with no separate system.. Catalogue default: Cloud provider's native operations tooling with no separate system. | Changes tooling and the processes written for it. |

View File

@@ -0,0 +1,100 @@
# Operational Requirements
This document defines what running the platform foundation solution must achieve. Each operational requirement is traced directly to customer statements, solution conditions, business outcomes, or binding fields settled within the operating specification. Requirements state what must be achieved without prescribing capabilities, roles, or tools.
## Operational requirements
The operational requirements are grouped below by kind: service level, process, organisation, and compliance.
### Service Level Requirements
- **OR-1 (Fault acknowledgement response)**: A fault on any Item is acknowledged within 15 minutes, at any hour.
- *Source*: `service_levels.priority_1_response_minutes`, `operations.support_hours`
- **OR-2 (Service restoration time)**: Service is restored within 4 hours of a major failure.
- *Source*: `service_levels.recovery_time_hours`, `BO-2`
- **OR-3 (Platform availability)**: The platform is available at least 99.9% of each month.
- *Source*: `K-1`
- **OR-10 (Change delivery turnaround)**: An approved change reaches production within one day of approval.
- *Source*: `BO-1`
### Process Requirements
- **OR-4 (Network change governance)**: Every change to the network requires customer platform team approval prior to implementation.
- *Source*: `REQ-01`
- **OR-5 (Monthly operational reporting)**: A report of service levels and operating costs is provided monthly.
- *Source*: `REQ-02`
- **OR-9 (Cryptographic key lifecycle)**: Encryption keys are rotated every 90 days.
- *Source*: `K-3`
### Organisation Requirements
- **OR-6 (Shared operating model)**: The platform foundation is operated under a shared service model between the customer and an external service provider.
- *Source*: `sourcing.model`
- **OR-7 (Delivery geographic boundaries)**: Operational delivery is performed exclusively from the United Kingdom, Poland, and India.
- *Source*: `sourcing.delivery_countries`
### Compliance Requirements
- **OR-8 (Audit log retention)**: Audit logs cannot be deleted for seven years.
- *Source*: `K-2`
### Requirements Traceability Summary
| Requirement ID | Kind | Operational Requirement Statement | Source References |
| --- | --- | --- | --- |
| **OR-1** | `service_level` | A fault on any Item is acknowledged within 15 minutes, at any hour. | `service_levels.priority_1_response_minutes`, `operations.support_hours` |
| **OR-2** | `service_level` | Service is restored within 4 hours of a major failure. | `service_levels.recovery_time_hours`, `BO-2` |
| **OR-3** | `service_level` | The platform is available at least 99.9% of each month. | `K-1` |
| **OR-4** | `process` | Every change to the network requires customer platform team approval prior to implementation. | `REQ-01` |
| **OR-5** | `process` | A report of service levels and operating costs is provided monthly. | `REQ-02` |
| **OR-6** | `organisation` | The platform foundation is operated under a shared service model between the customer and an external service provider. | `sourcing.model` |
| **OR-7** | `organisation` | Operational delivery is performed exclusively from the United Kingdom, Poland, and India. | `sourcing.delivery_countries` |
| **OR-8** | `compliance` | Audit logs cannot be deleted for seven years. | `K-2` |
| **OR-9** | `process` | Encryption keys are rotated every 90 days. | `K-3` |
| **OR-10** | `service_level` | An approved change reaches production within one day of approval. | `BO-1` |
---
## What was not stated
The operating specification leaves several areas assumed or unknown. If answered differently, these would alter the operational requirements.
### Assumptions
Because execution is pre-approved, the following assumptions were recorded from the operating specification and catalogue defaults:
1. **Assumed scope exclusions (`intent.exclusions`)**:
- *Assumption*: Workload applications hosted on the platform foundation are out of scope for provider operations, and network change approvals are retained by the customer platform team.
- *Impact if wrong*: If application workloads were in scope, additional operational requirements for application-layer monitoring, incident triage, deployment coordination, and service-level objectives would be required. If the provider had delegated authority for network modifications, OR-4 would need to be re-scoped.
2. **Assumed change release mechanism (`operations.change_process`)**:
- *Assumption*: Changes proceed through automated deployment pipelines with peer review, with network changes requiring explicit customer platform team approval.
- *Impact if wrong*: If a formal change advisory board (CAB) or manual approval cadence is mandated for non-network components, change lead time requirements (OR-10) would be affected.
3. **Assumed operations tooling (`service_management.tooling`)**:
- *Assumption*: The cloud provider's native operations tooling is utilized without a separate third-party IT service management (ITSM) system.
- *Impact if wrong*: If an external enterprise ITSM platform (e.g., ServiceNow, Jira Service Desk) or dedicated paging tool is mandated, additional requirements for cross-platform integration, ticket synchronization, and external reporting would emerge.
### Unknowns and Open Questions
The operating spec identifies the following unknown fields:
1. **Existing monitoring and alerting (`operations.monitoring`)**:
- *Current status*: Unknown.
- *Impact if answered*: Existing monitoring configurations and alert thresholds could impose specific operational requirements regarding synthetic checks, telemetry ingestion rates, or incident trigger definitions.
2. **Customer-owned build components (`solution.build.customer_owned`)**:
- *Current status*: Unknown.
- *Impact if answered*: If specific infrastructure components or pipelines are managed or maintained by customer internal teams, operational boundaries and fault hand-off requirements would need specification.
3. **Inherited solution risks (`solution.inherited_risks`)**:
- *Current status*: Unknown.
- *Impact if answered*: Known risks in interconnect redundancy or key distribution would require explicit contingency and high-availability operational requirements.
4. **Headcount constraints (`organisation.headcount_limit`)**:
- *Current status*: Unknown.
- *Impact if answered*: A hard headcount cap would constrain how 24x7 coverage is staffed across the three delivery countries (GB, PL, IN).
5. **Operational transition go-live date (`transition.go_live_date`)**:
- *Current status*: Unknown.
- *Impact if answered*: A defined target date would introduce operational readiness milestones and timeline-bound acceptance criteria.

View File

@@ -0,0 +1,87 @@
# Target operating model
## Design rationale
The operating model is designed around a shared responsibility between the customer and an external service provider, maintaining a clear boundary while enabling fast flow and 24x7 operational coverage. The approach maps operational capabilities to defined roles, ensures proper coverage for the declared service levels, and incorporates measurement feedback loops to achieve the stated business outcomes.
### Parties and Teams
Two primary parties are defined in the operating model: the **Customer** and the **Service Provider**.
To align with Team Topologies, we defined two platform-focused teams:
- The **Customer Platform Team** (Customer) focuses on platform evolution, owning the platform roadmap, defining golden paths, and measuring developer flow. This team acts as the platform owner and retains authority over network change approvals.
- The **Provider Operations Team** (Service Provider) is responsible for the day-to-day running of the platform foundation, including 24x7 monitoring, incident response, and executing operational tasks such as backups and patching.
### Roles, Location, and Capability Coverage
Roles are situated in the approved delivery countries (UK, Poland, and India):
- The **Platform Product Manager** (Lead) and **Platform Engineer** (Senior) are situated in the UK, closely collaborating with the customer's business and engineering teams during business hours. They steer the platform, ensure adherence to the 1-day deployment targets, and maintain alignment with business goals.
- The **Operations Engineer** and **Security Operations Analyst** provide 24x7 coverage from India, ensuring rapid response to faults and security events at all hours. They form the backbone of the provider's support roster.
- The **Operations Lead** (India) provides business-hours guidance and accountability for the operational capabilities, ensuring adherence to the 99.9% uptime target.
- The **Platform Ops Specialist** sits in Poland, offering advanced business-hours support, handling problem management, configuration management, and disaster recovery preparations.
For any capabilities where the provider is accountable, a customer role is marked as consulted to ensure there is always a contact point if the provider requires customer guidance. This distribution balances the need for continuous coverage with efficient use of resources across the specified delivery regions.
*(Note: Costs and exact headcounts are derived from the rate card. Because the rate card is currently in draft, any financial and staffing figures resulting from this model will act as placeholders.)*
### Interactions
The Customer Platform Team and Provider Operations Team interact primarily via an **X-as-a-Service** mode. The Provider consumes network change approvals from the Customer, while the Customer relies on the Provider for 24x7 incident response. During the first three months post-go-live, the teams will engage in a **Collaboration** to ensure a smooth handover of runbooks and establish operational processes before shifting entirely to X-as-a-Service.
### Measurement
To ensure the model is serving the business effectively, measurements are tied to the required outcomes. Since the model employs delivery and developer experience measurement (C-19), all three core layers of flow are measured:
- **Outcome** is measured by deployment frequency (using DORA metrics) to ensure changes reach production within a day, and time to restore service.
- **Guardrail** metrics measure developer satisfaction (using the SPACE framework) to ensure retention of platform engineers.
- **Leading signals** track friction and cognitive load (using the DevEx framework) to identify early friction that could impede the target of weeks rather than months for new payment products.
### Assumptions
As this execution is pre-approved, the following assumptions were made:
- Native operational tooling from the cloud provider is sufficient to configure necessary alerts, monitors, and verify restoration of backups or logs without needing separate systems, manual design-arounds, or findings.
- The customer platform team has sufficient capacity to process network change approvals within the timeframes needed to meet deployment targets.
## Responsibility matrix
| Capability | Platform Product Manager | Platform Engineer | Operations Engineer | Operations Lead | Security Operations Analyst | Platform Ops Specialist |
| --- | --- | --- | --- | --- | --- | --- |
| Incident management (C-1) | | C | R | A | | |
| Monitoring and alerting (C-2) | | C | R | A | | |
| Problem management (C-3) | | C | | A | | R |
| Service request fulfilment (C-4) | | C | R | A | | |
| Change enablement (C-5) | | A/R | | | | R |
| Release and deployment (C-6) | | A/R | | | | R |
| Configuration management (C-7) | | C | | A | | R |
| Backup and recovery (C-8) | | C | | A | | R |
| Service continuity (C-9) | | C | | A | | R |
| Capacity and performance management (C-10) | | C | | A/R | | |
| Access management (C-11) | | C | | A/R | | |
| Vulnerability and patch management (C-12) | | C | | | A/R | |
| Security monitoring and response (C-13) | | C | | | A/R | |
| Cost management (C-14) | C | | | A/R | | |
| Service level management (C-15) | C | | | A/R | | |
| Platform product management (C-16) | A/R | | | C | | |
| Self-service and golden paths (C-17) | | A/R | | | | R |
| Enablement (C-18) | | A/R | | | | R |
| Delivery and developer experience measurement (C-19) | A/R | | | C | | |
| Supplier management (C-20) | A/R | | | R | | |
| Regulatory evidence and reporting (C-21) | | C | | A | | R |
A accountable, R responsible, C consulted, I informed.
## Staffing and cost
| Role | Party | Country | Level | Hours | Effort a month | People | Cost a month |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Platform Product Manager | Customer | GB | lead | business | 30.0 h | 0.2 | not priced |
| Platform Engineer | Customer | GB | senior | business | 35.0 h | 0.3 | not priced |
| Operations Engineer | Service Provider | IN | engineer | 24x7 | 70.0 h | 5.0 | 17,034 GBP |
| Operations Lead | Service Provider | IN | lead | business | 50.0 h | 0.3 | 1,876 GBP |
| Security Operations Analyst | Service Provider | IN | senior | 24x7 | 55.0 h | 5.0 | 24,605 GBP |
| Platform Ops Specialist | Service Provider | PL | senior | business | 90.0 h | 0.6 | 4,088 GBP |
11.4 people in all. Cost a month: 47,603 GBP.
The monthly budget is 60,000 GBP: this is within it.
Worked from rate card `rate-card`, version `draft-2026-10-13`. **Its figures are placeholders: no price has been set, and this cost is an illustration.**
Converted from USD at 0.79, a rate set on 2026-10-13.
The customer's own staff are counted and not priced. A Role that covers hours wider than a working day is never smaller than the team that can cover them.

View File

@@ -0,0 +1,47 @@
# Validation results
## Check verdicts
The target operating model has been evaluated against the operational requirements, conditions, and business outcomes. Below are the verdicts for each check:
- **OC-1 (pass):** C-1 handles incident management 24x7, with a 24x7 Operations Engineer responsible, enabling acknowledgement of faults within 15 minutes at any hour. The cloud provider's native tooling (via C-2) supports configuring alerts without requiring a separate finding.
- **OC-2 (pass):** C-1 manages incident resolution with a 24x7 Operations Engineer responsible, allowing restoration of service within 4 hours of a major failure regardless of when the failure occurs.
- **OC-3 (pass):** C-15 tracks and gathers metrics on availability to ensure the 99.9% target is tracked, with business-hours roles accountable and responsible.
- **OC-4 (pass):** C-10 reviews capacity trends for the interconnect to assure 99.9% availability, managed by the Operations Lead.
- **OC-5 (pass):** C-5 coordinates network change approvals with the customer, and a Customer Platform Engineer is accountable and responsible for it, ensuring explicit approval from the customer platform team.
- **OC-6 (pass):** C-15 compiles the monthly operational report, pulling in cost data tracked by C-14, both managed by the Operations Lead.
- **OC-7 (pass):** C-20 coordinates responsibilities across shared delivery borders, with roles from both the customer and provider teams responsible.
- **OC-8a (pass):** R-3 (Operations Engineer) is based in India, satisfying the geographical requirement.
- **OC-8b (pass):** R-4 (Operations Lead) is based in India, satisfying the geographical requirement.
- **OC-8c (pass):** R-5 (Security Operations Analyst) is based in India, satisfying the geographical requirement.
- **OC-8d (pass):** R-6 (Platform Ops Specialist) is based in Poland, satisfying the geographical requirement.
- **OC-8e (pass):** R-1 (Platform Product Manager) is based in the UK, satisfying the geographical requirement.
- **OC-8f (pass):** R-2 (Platform Engineer) is based in the UK, satisfying the geographical requirement.
- **OC-9 (pass):** C-7 maintains audit log retention settings, with the Platform Ops Specialist responsible, ensuring logs cannot be deleted for seven years.
- **OC-10 (pass):** C-7 maintains and verifies retention settings on Cloud Storage for audit logs, satisfying the seven-year retention condition.
- **OC-11 (pass):** C-7 audits configuration drift and manages key rotation schedules, ensuring encryption keys are rotated every 90 days.
- **OC-12 (pass):** C-7 manages key rotation schedules for the Cloud KMS encryption keys, verifying they rotate every 90 days.
- **OC-13 (pass):** C-6 oversees deployment pipelines to ensure approved changes reach production within a day, with customer and provider roles responsible.
- **OC-14 (pass):** C-8 maintains backups of logs and keys and successfully tests restore procedures, managed by the Platform Ops Specialist.
**Overall Verdict:** The target operating model **passes** validation. The roles, coverage hours, and capabilities are fully aligned with the operational requirements and the designated shared service sourcing model.
## Scenario walk-throughs
Each scenario is generated from the operating model and traced through it by code.
| Scenario | Result | Path |
| --- | --- | --- |
| Hub network and interconnect (S-1) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Hub network and interconnect (S-1) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Audit log store (S-2) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Audit log store (S-2) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Deployment pipeline (S-3) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Deployment pipeline (S-3) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Encryption keys (S-4) fails at an hour faults are responded to (24x7). | pass | Operations Engineer responds under C-1, and Operations Lead is answerable. |
| Encryption keys (S-4) must be changed. | pass | Platform Engineer makes the change under C-5. |
| Condition K-1 must be shown to hold: The platform is available at least 99.9% of each month. | pass | Check OC-4 on C-10 keeps it true. |
| Condition K-2 must be shown to hold: Audit logs cannot be deleted for seven years. | pass | Check OC-10 on C-7 keeps it true. |
| Condition K-3 must be shown to hold: Encryption keys are rotated every 90 days. | pass | Check OC-12 on C-7 keeps it true. |
| Service Provider cannot be reached. | pass | Every Capability it is answerable for has a Role in another Party to turn to. |
| A month's cost of operating is compared with the budget of 60,000 GBP. | pass | The cost is 47,603 GBP, within the budget. |