Back to Insights
VALUE ARCHITECT PLAYBOOK

Cloud Readiness Assessment Playbook (AWS vs Azure vs GCP)

De-risk your roadmap. Stop speculating on infrastructure costs and start measuring readiness with a Principal-Grade framework designed for the Fortune 500. This is the definitive Phase 1 playbook for running a Cloud Readiness Assessment (CRA) that actually leads to a decision.

Time-to-Value
6 Weeks
Discovery to Strategy
Risk Mitigation
92% Risks Mitigated
Governance & Security
Artifact ROI
$1.2M+ Est.
Year 1 Optimization

Interactive: CRA Maturity Scoreboard

Self-assess your enterprise readiness across 6 critical pillars (0.0 to 5.0 scale).

Cumulative Readiness Score
2.5 / 5.0
"Cautious Deployment Advised"

Modernization Savings Estimator (ROI)

Quantify the business impact of pivoting from IaaS to PaaS during Phase 1.

Projected Annual OpEx Reduction $342,000

"Equivalent to recovering 14,000 labor hours per year from patching and hardware maintenance."

Who This Is For & What You Will Get

What CRA Phase 1 Is (And What It Is Not)

A Cloud Readiness Assessment (Phase 1) is a unified strategic exercise to determine if you should move, where you should move, and what it will cost. It is not an implementation project. It is a decision-making project.

It is:

It is NOT:

Strategic Advisory: The Missing "People" Pillar

Cloud transformations fail due to culture, not code. While Phase 1 is data-heavy, our expert panel mandates the inclusion of a Skills Readiness Assessment. You must evaluate if your current team can manage the target state (e.g., IAAC/Terraform) or if a Cloud Center of Excellence (CCoE) needs to be recruited. Don't build a Ferrari if no one in the firm has a driver's license.

Inputs Needed (The “Don’t Waste Week 1–2” Checklist)

Before launching Week 1, ensure you have these inputs. Without them, you will spend the first sprint chasing permissions instead of data.

Example Scope Assumptions: Up to 7,350 VMs, 480 Databases, 275 Network Segments, 920 TB Storage.

Audit Note: Specialized workloads (SAP HANA, Mainframe/AS400, or legacy physical Solaris/AIX) require an additional 4-week deep-dive phase and are typically excluded from baseline TCO models.

Input Owner When Needed Why It Matters
CMDB / Asset Inventory IT Ops Week 0 Baseline for discovery agent deployment.
Firewall Rules / Network Map Network Team Week 1 Critical for dependency mapping and segmentation planning.
VMware vCenter/Hyper-V Read Access Infra Lead Week 1 Required for automated discovery tooling (e.g., Azure Migrate, RVTools).
Current Contract Data (EA/EDP) Procurement Week 2 Essential for accurate TCO baselining and software assurance benefits.
Application List (Business Critical) App Owners Week 2 Identifies the "Crown Jewels" (e.g., MOL, RTA, Atex, AirTable).

Delivery Flow (Timeline Mapped to Phases)

This 11-week schedule assumes a "Sprint zero" approach where tooling is prepped immediately.

Source Env Discovery Agents Inventory Data Dependency Map Analysis Layer (TCO & Scoring)

The CRA Architecture Vision (Sovereign Stack)

A high-fidelity 3D representation of Lifecycle Alignment.

B

Phase B: Business Architecture

Global Cloud Operating Model

Executive alignment on ROI, RACI structures, and Financial Governance (FinOps).

C

Phase C: Application Architecture

Service-Oriented Modernization

Dependency mapping and PaaS-first rationalization for 7,350+ VMs.

D

Phase D: Technology Architecture

Hub-Spoke Landing Zone

Automated Hub-Spoke fabric with Zero-Trust guardrails and BGP route management.

Service Journey Map: The Stakeholder Experience

What happens during the 11-week assessment? A multi-dimensional look at actions, tooling, and pain points.

Timeline
Wk 1-2: Setup
Wk 3-5: Discovery
Wk 6-7: Selection
Wk 8-9: Financials
Wk 10-11: Final
App Owner
Actions
Submit Inventory / RACI sign-off.
App Interviews / Dependency Review.
Attend evaluation workshops.
Validate TCO savings assumptions.
Executive present / Phase 2 Approval.
Discovery
Tooling
RVTools / CSV Ingest.
Azure Migrate Agents.
Scoring Matrix v1.
TCO Engine Logic.
Final Exec Deck.
Key
Friction
Access stall (Firewall).
Zombie VM ID.
Vendor bias tension.
Licensing complexity.
Path to Pilot Clear.
Phase Weeks Activities Deliverables Done Means
Phase 1: Strategy Assessment Week 1–5 Stakeholder interviews, maturity workshops, governance setup. Project Plan, Comm Plan, Maturity Report. Interviews complete; Maturity gaps identified.
Phase 2: Automated Discovery Week 1–5 Tooling deployment (collectors), data gathering, dependency mapping. Discovery Report, App Dependency Map. 100% of in-scope infra scanned; 30 days of performance data.
Phase 3: CSP Selection Week 6–7 Evaluation workshops, scoring criteria definition, detailed comparisons. Evaluation Matrix, Scoring Report. Vendor selected (or shortlist confirmed).
Phase 4: TCO Analysis Week 8–9 Current state vs. Future state modeling, migration cost estimation. TCO Model (Excel/PBI), 3-Year Projection. ROI and payback period calculated.
Phase 5: Strategy Report Week 10–11

Principal-Grade Deliverable Library (Phase 1)

The following toolkit represents the core "Strategic Assets" generated during a Phase 1 assessment. Each template is designed for executive presentation and technical rigor.

GOVERNANCE

Strategic Lifecycle Plan

11-week roadmap including board-level sign-off milestones and tooling procurement gates.

Project Plan Comm Plan
TECHNICAL

Infra Discovery Report

Inventory summaries, modernization candidates, and zombie server analysis for baseline optimization.

Discovery Template
FINANCIAL

TCO & Evaluation Model

Excel-optimized data models for weighted vendor scoring and 3-year P&L impact projection.

Scoring Matrix TCO Inputs
STRATEGY

Board-Ready Synthesis

Final strategy narrative and an 8-slide executive outline covering the 'Risk of Inaction'.

Full Report Exec Slide Deck

Technical Advisory: PaaS-First Modernization

Our audit of 7,350 VMs shows that 30% of database workloads currently on IaaS are candidates for PaaS (Azure SQL MI / Amazon RDS). The TCO impact of removing OS patching and built-in HA is often the difference between a 2-year and 4-year ROI.

Criteria Category Requirement Weight AWS Azure OCI
Technical BYOL Support (Windows/SQL) 10 4 5 5
Commercial EA/EDP Discounts 9 3 5 3
Pricing Global Price Parity 8 2 2 5
Operational Identity Stack (AD) 8 4 5 3
Innovation GenAI / LLM Access 7 5 5 4
Ecosystem PaaS Maturity (Managed DB/Containers) 8 5 4

TCO Approach (Assumptions & Drivers)

Your TCO model is only as good as its assumptions. Be explicit about what is in and out.

Included: Compute, Storage, Networking, Licensing (OS/DB), Migration Labor, Training.

Excluded: App refactoring costs (Phase 2), unplanned downtime costs, depreciation of legacy hardware (sunk cost).

Migration Timeline (Weeks) Total Cost Current On-Prem Cloud Target The Migration Bubble (Double Run Costs)

Top 5 Cost Drivers (Sensitivity Analysis)

  1. Right-sizing Aggression: Are we lifting "As-Is" or optimizing to "Required"? (Scope: 7,350 VMs).
  2. Licensing Portability: Azure Hybrid Benefit / BYOL impact on SQL Server costs.
  3. Storage Tiering: Moving cold data (920 TB) to Archive tier vs. Hot tier.
  4. Reserved Instances: Commitment level (1-year vs. 3-year).
  5. Labor Rate: Internal staff vs. MSP managed services.

FinOps Advisory: Unit Economics

During Phase 1, move beyond "Total Monthly Bill." Establish Unit Economics—the cost to run a single editorial hub transaction (e.g., AirTable) or a single analytics query (RTA). This language resonates with the CFO more than raw VM pricing.

Interactive Readiness Assessment

Audit your organization across 8 critical dimensions. Results are plotted in real-time on the Radar Chart below.

Financial & Economic Readiness

Phase B: Business Arch
Do you have a defined "Tagging Taxonomy" enforced before resources are created?
Why it matters: Without forced tagging (Cost Center, App Owner), you cannot do chargeback. FinOps fails day 1.
Ref: CAF Tagging
Have you quantified "Unit Economics" (e.g., cost per transaction) vs just total spend?
Why it matters: Cloud is consumption-based. if you don't track unit cost, scaling business means scaling waste.

Architecture & Technical Debt

Phase C/D: Tech Arch
Have we audited for "Non-x86" dependencies (Mainframe, AS/400, Solaris)?
Why it matters: These require specialized "bare metal" solutions in Azure/AWS and often break standard lift-and-shift timelines.
Is the "Landing Zone" design (Hub-Spoke) approved by InfoSec?
Why it matters: You cannot deploy apps securely without a pre-approved network foundation (Firewalls, VNET peering).

Data Strategy & Hygiene

Phase C: Data Arch
Is data classified (Public/Confidential/Restricted) before migration?
Why it matters: "Lift and Shift" of unclassified data creates a compliance nightmare. You can't apply DLP policies if you don't know what the data is.

Security & Compliance (Zero Trust)

Phase G: Risk
GDPR/Sovereignty: Are data residency boundaries defined for all workloads?
Why it matters: For UK/EU entities, moving data to a US region (even by accident) is a legal violation. Policy must enforce regions.
Ref: ICO GDPR
Is Identity (Entra ID) the new perimeter (MFA enforced)?
Why it matters: Firewalls aren't enough. In the cloud, Identity is the control plane. 99% of breaches are identity theft.

Operational Readiness

Phase E: Opportunities
Do we have an automated patching strategy for cloud VMs?
Why it matters: You can't use on-prem GPO/SCCM seamlessly. You need Azure Update Manager or AWS Systems Manager.

User Experience (Latency/Perf)

Human-Centric Design
Have we tested "VDI Latency" for remote users?
Why it matters: For Azure Virtual Desktop or Citrix, latency > 150ms kills productivity. Don't guess; measure.

Governance & People

Phase G: Governance
Is there a functioning "Cloud Center of Excellence" (CCoE)?
Why it matters: A CCoE bridges the gap between traditional IT and Cloud DevOps. Without it, you get "Shadow IT".
Ref: CAF CCoE
Has the team been trained on Terraform/Bicep (IaC)?
Why it matters: Clicking in the portal is fine for POCs, but fatal for production. Infrastructure as Code is mandatory.

AI Readiness (GenAI/LLM)

NIST AI RMF
Data Hygiene: Is unstructured data (PDFs/Docs) clean enough for RAG ingestion?
Why it matters: "Garbage In, Garbage Out." RAG models hallucinate if fed duplicate, obsolete, or poorly formatted documents.
Compute: Have we reserved GPU capacity (H100/A100) or are we relying on spot instances?
Why it matters: High-end GPUs are scarce. Production AI cannot rely on "Spot" instances that can be evicted at any second.
Responsible AI: Is there a policy framework for AI Bias & Hallucination?
Why it matters: You are liable for your AI's output. Governance must exist before the first prompt is sent.
Ref: NIST AI RMF
Category Score (%)
Overall Readiness
0%
Not Assessed

Generates a CSV compatible with Excel/PowerBI.

Common Failure Points & Prevention

Deep Dives

1. Governance That Prevents Scope Drift

The number one killer of assessments is scope drift. A solid governance structure requires a standard Project Plan and Communication Plan established in Week 1. By defining the "Definition of Done" for each phase early, you prevent the "just one more scan" syndrome. Use the templates to set a rigorous cadence of SteerCo meetings (bi-weekly) and Working Group meetings (weekly) to keep decision-makers aligned.

Arch Advisory: The ExpressRoute/DirectConnect Limit

For large migrations, ignore bandwidth at your peril. Our audit warns that mapping 275 network segments requires careful BGP route limit planning (e.g., 1000 routes for ExpressRoute). Phase 1 must validate if your core routers can handle the cloud-native routing table.

CENTRAL HUB (Firewall & NVA) Dev Spoke Shared SVCS PROD SPOKE ON-PREM DC

2. Workshops That Produce Usable Requirements

Don't just schedule "meetings." Schedule structured workshops with an agenda and a pre-defined question bank. Whether it's the "Security & Compliance" workshop or the "Application Rationalization" session, the goal is to extract constraints, not just wishlists. Your Requirements Summary should clearly distinguish between "Must Haves" (Regulatory/Security) and "Nice to Haves" (New features).

3. Discovery Done Right

Automated discovery is non-negotiable for an estate of 7,350 VMs. Tooling like Azure Migrate or standard 3rd party collectors must be deployed to cover 100% of the in-scope environment. The Discovery Report isn't just a list of servers; it's a map of dependencies. It must answer: "If I move Server A, does Server B break?" Good data mapping ensures migration waves are defined by application affinity, not just subnet convenient.

4. Cloud Evaluation Matrix

Vendor bias destroys credibility. The Evaluation Matrix must be objective. If you prefer Azure because of "teams familiarity," document that as an "Operational Readiness" score, not a magical technical advantage. Use weighted scoring (1-10) to prioritize what matters most (e.g., specific database PaaS support might outweigh minor compute cost differences).

5. TCO + 1-Year Projection

A TCO model is a forecast, not a bill. To handle uncertainty, present a range: "Conservative," "Likely," and "Aggressive." Your TCO Model should clearly show the impact of cost optimization levers like Reserved Instances (RIs) and Hybrid Benefits. Always include a 1-year projection that accounts for the "migration bubble" (paying for both on-prem and cloud during the transition).

Executive Decision Summary (The "One-Pager")

Executives don't read 100-page reports. They read the Executive Summary. Structure it for a decision:

Decision Summary: Cloud Readiness Recommendation

  • Recommendation: Proceed with [Selected Cloud] as the primary landing zone.
  • Top 5 Reasons:
    1. Lowest 3-year TCO (saving ~18% vs. current).
    2. Best fit for existing Microsoft licensing (Azure Hybrid Benefit).
    3. Highest security maturity score for our regulated data.
    4. Native support for critical VDI workloads.
    5. Team skills align closest (lowest training gap).
  • Trade-offs: [Selected Cloud] has weaker [Specific Feature] compared to [Competitor], but impact is low.
  • Cost View: Estimated migration cost: $X. Annual Cloud Run Rate: $Y. Payback period: 18 months.
  • Risks: Legacy app latency requires network redesign.
  • Decision Needed: Approve Phase 2 (Foundation Build) budget by [Date].


The Decision Protocol: From Score to Strategy

01

The Rules of Thumb

Stop asking "Which cloud is best?" and start asking "Where is the gravity?".

Single Cloud (Default)

Unless you have a hard regulatory constraint or M&A reality, pick one. The cost of splitting skills (Terraform vs Bicep) outweighs "vendor lock-in" risk.

Hybrid (Reality)

Choose this when latency (manufacturing) or data residency (Mainframe) is a physics problem, not a preference.

Multi-Cloud (Advanced)

Only if you can identify a specific workload that fails on Azure but works on AWS. No named workload? No multi-cloud.

03

The Data

Make decisions based on inventory, not feelings.

Use the AWS Strategy Application Portfolio Assessment. It’s the gold standard for unbiased estate analysis.


The Cloud Architect's Resource Library

A curated collection of industry-standard tools and frameworks to accelerate your Phase 1 assessments.

Operational Toolkit

Resources

Video Vault (Must Watch)

Expert deep-dives on the patterns used in this architecture.


Ready to operationalize your Azure journey?

This playbook outlines the "What" and the "Why." If you need help with the "How"—specifically automating the discovery and TCO modeling—reach out.

Contact Me View the Toolkit
Back to Insights