Partner Guide
Opening the Conversation
The best modernization conversations start with the business, not the technology. Use these questions to understand the customer’s situation:
| Question | What You Learn |
|---|---|
| What business outcomes are you trying to achieve in the next 12–18 months? | Strategic priorities and urgency |
| Where is your data today, and who can access it when they need it? | Data silos and analytics gaps |
| What happens during peak demand — can your systems handle it? | Scale limitations and business impact |
| How long does it take to deploy a change to production? | DevOps maturity and agility constraints |
| What keeps your CISO or compliance team up at night? | Security and regulatory pressures |
The Two Modernization Paths Decision Framework
Use this decision tree to match workloads to the right path:
| Criteria | Stabilize | Transform |
|---|---|---|
| Business change frequency | Stable, low change rate | Actively developed, frequent releases |
| Scale requirements | Predictable, steady load | Spiky, elastic, or growing rapidly |
| Application complexity | Works well as-is | Needs refactoring or new capabilities |
| Risk tolerance | Low — minimize disruption | Moderate — willing to invest in change |
| Time to value | Weeks | Months |
| Cost model | Azure Migrate and TCO estimate | PaaS/serverless model validated by data |
| Fabric integration | SQL MI Mirroring for supported data | SQL DB Mirroring plus data products |
| Best for | ERP, back-office, stable LOB apps | Customer-facing, e-commerce, new builds |
Fabric Value Proposition — By Audience
Tailor the Fabric message to the stakeholder:
For the CTO / CIO
“Fabric can reduce the need for separate ETL pipelines and duplicated analytical stores. Supported operational data from both modernization paths can land in OneLake and be governed for BI, ML, and AI workloads.”
For the CFO
“Traditional analytics infrastructure requires separate investments in ETL tooling, data warehouses, and BI platforms. Fabric consolidates these into a single capacity-based model. Combine Azure Migrate business cases, the Azure TCO Calculator, and Fabric capacity planning to quantify the customer’s actual total cost of ownership.”
For the Line-of-Business Leader
“Your team gets near-real-time dashboards and AI-powered insights on supported operational data — without building custom ETL for mirrored tables. If the data exists in SQL MI or Azure SQL DB and meets mirroring prerequisites, it can become a governed Fabric data product.”
For the CISO / Data Governance Lead
“Mirroring does not automatically copy SQL row-level security, object permissions, dynamic data masking, or Purview sensitivity labels into OneLake. The engagement includes a Fabric governance baseline so access, labels, ownership, and review processes are rebuilt before broad use.”
Objection Handling
| Objection | Response |
|---|---|
| ”We are not ready for a full cloud migration.” | The path model is designed for exactly this — start with Stabilize for quick wins and low risk, then Transform only where the business case justifies it. |
| ”We already have a data warehouse.” | Fabric does not replace an existing warehouse overnight. SQL MI Mirroring runs alongside your current setup. Start with one workload, prove the value, then expand. |
| ”Kubernetes is too complex for our team.” | Azure Container Apps abstracts away Kubernetes. Your developers deploy containers without managing clusters, nodes, or networking. |
| ”We cannot afford downtime for migration.” | The Managed Instance link uses near-real-time replication to SQL MI and limits downtime to final cutover. Azure DMS is available as a fallback. VM migration uses replication with planned cutover windows and rollback criteria. |
| ”How is this different from just using Power BI?” | Power BI is the visualization layer. Fabric includes the data lake (OneLake), data engineering (Spark), data science (ML), and real-time intelligence — all on one platform. Power BI becomes more powerful when backed by Fabric. |
Fabric Readiness Checklist
Before positioning Fabric mirroring as an execution milestone, validate:
- Source database permissions needed for mirroring are approved and least privilege is documented
- Row-level security, object permissions, dynamic data masking, and sensitivity labels that must exist in Fabric are designed explicitly
- Private SQL MI connectivity uses a virtual network data gateway or on-premises data gateway with access to the private endpoint
- Unsupported table, column, feature, identity, and tenant scenarios are known before timeline commitments are made
- Data-product ownership, endorsement, lineage, access reviews, and support paths are assigned
Industry Quick Cards
Manufacturing (Contoso Industries)
- Trigger: Board-level mandate for near-real-time supply chain visibility
- Stabilize workloads: ERP, MES (127 VMs, 22 databases)
- Transform workloads: Customer portal, supply chain dashboard
- Fabric payoff: Predictive maintenance + supply chain Power BI dashboard
- Full story →
Financial Services (Woodgrove Bank)
- Trigger: Regulatory requirement for near-real-time transaction monitoring
- Stabilize workloads: Core banking, regulatory databases (340 VMs, 45 databases)
- Transform workloads: Digital banking app, fraud detection engine
- Fabric payoff: Regulatory dashboards + fraud analytics and monitoring
- Full story →
Retail (Northwind Traders)
- Trigger: CEO digital-first strategy, e-commerce scaling failures
- Stabilize workloads: ERP, loyalty program (65 VMs, 12 databases)
- Transform workloads: E-commerce platform, customer analytics
- Fabric payoff: Customer 360 dashboard + recommendation engine
- Full story →
Engagement Timeline Template
Use this as a starting point — adjust based on estate size and complexity:
| Phase | Duration | Activities |
|---|---|---|
| Discovery & Strategy (MCEM 1) | 2–4 weeks | Business workshops, stakeholder alignment, CAF Strategy |
| Assess & Design (MCEM 2) | 4–6 weeks | Azure Migrate scan, path classification, architecture |
| Stabilize Execution (MCEM 3) | 4–12 weeks | VM waves, SQL MI migration, Fabric mirroring |
| Transform Execution (MCEM 3) | 3–6 months | .NET modernization, containerization, CI/CD, Azure SQL DB |
| Value Realization (MCEM 4) | 2–4 weeks | Cost review, analytics rollout, outcomes measurement |
| Manage & Optimize (MCEM 5) | Ongoing | Continuous optimization, Fabric expansion, skills development |