Introduction
Migrating your data platform to Microsoft Fabric is a transformative initiative that requires careful planning, stakeholder alignment, and methodical execution. Whether you’re transitioning from legacy on-premises systems, cloud platforms like Azure Synapse Analytics, or other analytics environments, having a comprehensive migration checklist ensures you don’t overlook critical steps that could impact data quality, security, performance, or user adoption.
This article provides an actionable, detailed migration checklist specifically designed for organisations preparing to move to Microsoft Fabric. We’ll walk through pre-migration assessment, technical preparation, execution planning, data migration, governance implementation, and post-migration validation. Each section includes checkbox items with explanations to help your team systematically prepare for a successful Fabric deployment. This Microsoft Fabric migration checklist is designed to be practical and actionable for teams of all sizes. Use this Microsoft Fabric migration checklist to align stakeholders and technical teams before you start migrations.
Microsoft Fabric represents a unified analytics platform that combines data warehousing, lakehouse architecture, real-time analytics, and AI capabilities. To realise its full potential, you need a structured approach that addresses business requirements, technical dependencies, data quality, governance frameworks, and team readiness. This checklist serves as your roadmap.
Pre-Migration Assessment and Discovery
Before you begin any technical work, you must thoroughly understand your current data environment, business objectives, and constraints. This discovery phase sets the foundation for all subsequent decisions. During discovery, refer to the Microsoft Fabric migration checklist to ensure every stakeholder requirement is captured.
Stakeholder Alignment and Business Objectives
- Define migration goals and success metrics
Establish clear, measurable objectives for your Fabric migration. Are you seeking to reduce time-to-insight, lower total cost of ownership, improve data governance, or enable advanced analytics and AI capabilities? Document specific KPIs such as query performance improvements, cost savings, user adoption rates, or faster reporting cycles. These metrics will guide prioritisation and help you measure ROI post-migration. - Secure executive sponsorship and funding approval
Ensure C-level support from your CIO, CFO, or Chief Data Officer. Migration initiatives require sustained investment, skilled resources, and organisational change management. Executive sponsorship removes roadblocks, secures budget, and signals organisational commitment to teams. Document the business case and obtain formal approval before proceeding. - Identify all stakeholder groups and their requirements
Map out stakeholders across data engineering, analytics, business intelligence, IT operations, security, compliance, finance, and business units. Conduct interviews or workshops to understand their pain points, current workflows, data needs, and concerns about migration. This ensures your migration strategy addresses diverse perspectives and builds buy-in across the organisation. - Establish a migration governance committee
Form a cross-functional steering committee with representatives from IT, data, security, compliance, and business units. This committee meets regularly to review progress, resolve blockers, manage scope changes, and communicate status to leadership. Clear governance prevents silos and ensures accountability. - Document business-critical applications and dependencies
Create an inventory of all systems that depend on your current data platform reporting tools, operational applications, third-party integrations, and downstream processes. Understand the criticality and impact of any downtime or data latency. This informs your migration sequencing and cutover strategy. - Use the Microsoft Fabric migration checklist to map success metrics to technical tasks
Ensure each business objective is paired with measurable technical checkpoints so the project delivers what stakeholders expect.
Current State Assessment
- Audit existing data sources and systems
Conduct a comprehensive inventory of all data sources feeding your current platform: databases (SQL Server, Oracle, PostgreSQL), cloud platforms (AWS, Google Cloud), SaaS applications, APIs, and file repositories. Document schema complexity, data volume, refresh frequency, and data quality issues. Tools like Microsoft’s migration assessment guides can help identify migration blockers early. - Measure current data volumes, growth rates, and velocity
Quantify the size of your current data estate in terabytes or petabytes, analyse historical growth trends, and understand data ingestion velocity. This informs capacity planning, cost modelling, and performance tuning in Fabric. Large datasets may require staged migration approaches or parallel running periods. - Document existing ETL/ELT processes and workflows
Map all data pipelines, transformations, and orchestration logic currently running in your environment. Identify tools used (SQL Server Integration Services, Azure Data Factory, Talend, Informatica, etc.), transformation logic, scheduling, error handling, and dependencies. This is essential for rebuilding pipelines in Fabric’s native orchestration tools. - Assess current reporting and analytics infrastructure
Inventory all Power BI reports, dashboards, SSRS reports, Tableau workbooks, or other analytics artefacts. Document their usage, criticality, owners, and refresh schedules. Identify reports that will migrate to Fabric versus those that remain on legacy platforms. This prevents orphaned reports and ensures continuity of critical insights. - Review current data governance and security posture
Document existing data classification schemes, access control models, data lineage tracking, retention policies, and compliance frameworks. Understand which data is subject to regulations (GDPR, HIPAA, PCI-DSS, Australian Privacy Principles). This informs governance architecture in Fabric. - Identify performance bottlenecks and optimisation opportunities
Analyse query performance, slow-running reports, data refresh times, and resource constraints in your current platform. Understanding pain points helps prioritise which workloads to migrate first and where Fabric can deliver the most value. - An effective Microsoft Fabric migration checklist includes current state audits and helps prioritise data sources for migration
Use the outputs of your audits to drive the migration sequencing and minimize surprises during cutover.
Technical Architecture and Planning
With a clear understanding of your current state and business objectives, you can now design your Fabric architecture and migration strategy. When designing your solution architecture, keep the Microsoft Fabric migration checklist handy to validate design choices against organisational requirements.
Solution Architecture and Design
- Define your Fabric architecture (Lakehouse vs. Data Warehouse vs. Mirroring)
Decide which Fabric components best suit your use cases. The Fabric Lakehouse provides flexibility for data lakes with OneLake storage and Delta Lake tables. The Data Warehouse offers SQL query capabilities similar to traditional data warehouses. Mirroring enables real-time replication from operational databases. Many organisations use a hybrid approach. Consult Agile Insights’ architecture guidance to align design with your industry and use cases. - The Microsoft Fabric migration checklist helps choose between Lakehouse and Data Warehouse patterns based on your use cases
Use documented use cases and expected query patterns to guide this choice. - Design data ingestion strategy (real-time vs. batch vs. streaming)
Determine how data will flow into Fabric. Real-time streaming via event hubs suits operational analytics. Batch ingestion via Azure Data Factory or Fabric pipelines works for periodic updates. Mirroring provides near-real-time replication of transactional databases. Your choice impacts architecture, cost, and latency. - Plan capacity sizing and resource allocation
Estimate the Fabric capacity (F-SKU) required based on concurrent users, query complexity, data volume, and refresh frequency. Larger organisations often start with F2048 or higher. Underestimating capacity leads to throttling and poor user experience; overestimating wastes budget. Microsoft provides capacity planning tools and cost calculators. - Design lakehouse folder structure and naming conventions
Establish a consistent folder hierarchy in OneLake that reflects your data domains, business areas, and maturity levels (raw, bronze, silver, gold). Clear naming conventions make data discoverable and maintainable. Document standards for file formats, compression, and partitioning. - Plan data warehouse schema design and relationships
If using Fabric Data Warehouse, design your schema (star schema, snowflake, or denormalised). Define fact and dimension tables, relationships, and slowly changing dimensions. This design should reflect business semantics and support key analytics use cases. - Evaluate AI and advanced analytics integration points
Identify where machine learning models, OpenAI integration, or predictive analytics add value. For example, demand forecasting in retail, patient risk scoring in healthcare, or anomaly detection in operations. Plan how these will be built and deployed in Fabric. Agile Insights offers advanced analytics services tailored to your industry. - Design real-time analytics and streaming architecture
If real-time insights are critical, plan your streaming ingestion, event processing, and real-time dashboards. Fabric supports streaming via event hubs and real-time semantic models for up-to-the-second analytics.
Migration Approach and Sequencing
- Select migration strategy (Big Bang vs. Phased vs. Parallel)
Big Bang migrations are faster but riskier. Phased approaches migrate workloads incrementally, reducing risk but extending timelines. Parallel running maintains legacy systems alongside Fabric during transition. Your choice depends on criticality, complexity, and risk tolerance. Document the rationale and timeline. - Refer to the Microsoft Fabric migration checklist when selecting your migration strategy to ensure risk and timelines are balanced
Use the checklist to capture contingency plans and required validations for each strategy. - Prioritise workloads and data domains for migration
Rank workloads by business value, complexity, and dependencies. Typically, lower-risk, high-value workloads migrate first (e.g., financial reporting, operational dashboards). Complex, interdependent systems migrate later. This builds momentum and proves value early. - Create detailed migration timeline and milestones
Develop a project schedule with phases: discovery, design, build, testing, cutover, and stabilisation. Include buffer time for unexpected issues. Communicate timelines to stakeholders and track against plan. Major milestones might include proof-of-concept completion, first production workload cutover, and full legacy system decommissioning. - Plan for parallel running periods if needed
Determine which workloads require parallel running to validate data accuracy before cutover. Plan infrastructure, monitoring, and reconciliation processes. Parallel running adds cost and complexity but provides safety nets for critical systems. - Identify quick wins and proof-of-concept opportunities
Select one or two lower-risk, high-visibility projects to migrate early. Successful POCs build confidence, generate learnings, and demonstrate ROI to stakeholders. Quick wins often include dashboards, reports, or analytics workloads rather than core operational systems.
Data Assessment and Preparation
Data quality and readiness directly impact migration success. Invest time upfront to assess and prepare your data. Data teams should use the Microsoft Fabric migration checklist to govern data quality gates before migration.
Data Quality and Validation
- Conduct data quality assessment across all source systems
Profile your data to identify quality issues: null values, duplicates, inconsistent formats, outliers, and referential integrity violations. Use data profiling tools or SQL queries to assess completeness, accuracy, consistency, and timeliness. Document findings and establish quality baselines. - Identify and document data quality rules and thresholds
Define acceptable quality standards for each dataset (e.g., “Customer records must be 99% complete”). Document business rules, validation logic, and acceptable error rates. These rules will be built into Fabric pipelines and monitoring. - Plan data cleansing and transformation logic
Design cleansing workflows to address identified quality issues. This might involve deduplication, standardisation, format conversion, or enrichment. Decide whether cleansing happens during migration or post-migration in Fabric. - Test data migration with sample datasets
Before full-scale migration, perform test migrations with representative samples. Validate record counts, data types, relationships, and calculated fields. Compare source and target data to identify transformation errors or data loss. - Establish data reconciliation processes
Plan how you’ll validate that all data migrated correctly. This includes row counts, checksums, aggregate comparisons, and spot-checking. Automated reconciliation queries should be built into your migration process. - Document data lineage and transformations
Create detailed documentation of how data flows from sources through transformations to final tables. This supports troubleshooting, compliance audits, and impact analysis. Fabric’s data lineage features can help visualise and track these flows.
Data Classification and Privacy
- Classify data by sensitivity and compliance requirements
Categorise data as public, internal, confidential, or restricted based on sensitivity. Identify data subject to regulations: personally identifiable information (PII), health information, financial data, or trade secrets. Document compliance requirements for each category. - Identify personally identifiable information (PII) and sensitive data
Map where PII (names, email addresses, phone numbers, IDs) and sensitive data (health records, financial information, biometrics) exist. This informs masking, encryption, and access control strategies. - Plan data masking and anonymisation strategies
Decide which sensitive fields require masking in non-production environments. Implement techniques like tokenisation, hashing, or synthetic data generation. This protects privacy while enabling testing and development. - Ensure compliance with Australian Privacy Principles and regulations
Review Australian Privacy Principles, state-based privacy laws, and industry-specific regulations (healthcare, finance, government). Ensure Fabric infrastructure, data residency, encryption, and access controls meet compliance requirements. Consider engaging compliance specialists if needed. - Plan data retention and archival policies
Define how long data will be retained in Fabric and when it should be archived or deleted. Implement retention policies aligned with business and legal requirements. Fabric supports lifecycle management for cost optimisation.
Governance and Security Framework
Robust governance and security are non-negotiable for enterprise Fabric deployments. Plan these thoroughly before migration. Security leads should incorporate the Microsoft Fabric migration checklist into their governance plans to avoid missed compliance items.
Access Control and Authentication
- Design role-based access control (RBAC) model
Define roles based on job functions: data engineers, analysts, report consumers, administrators. Specify permissions for each role (read, write, execute, admin). Document the principle of least privilege users receive only necessary permissions. - Plan Azure AD integration and service principal setup
Ensure Fabric integrates with your Azure AD for user authentication and single sign-on (SSO). Plan service principals for unattended automation (pipelines, scheduled refreshes). Document security groups and role assignments. - Configure row-level security (RLS) and column-level security (CLS)
Identify data that requires row-level filtering based on user context (e.g., sales reps see only their region’s data). Plan CLS for columns containing sensitive information. Implement these in semantic models and warehouse views. - Plan multi-factor authentication (MFA) enforcement
Enable MFA for all Fabric users, especially administrators. Document MFA policies and exceptions. MFA significantly reduces risk of unauthorised access. - Establish service account and API authentication strategy
Plan how applications and automated processes will authenticate to Fabric. Use managed identities where possible; otherwise, use service principals with strong credentials. Rotate credentials regularly and audit usage.
Data Governance and Cataloguing
- Implement Microsoft Purview for data governance
Set up Purview to scan Fabric assets, build a data catalogue, and track data lineage. Configure sensitivity labels and data classification rules. Purview provides a single pane of glass for data discovery and governance across your organisation. Agile Insights specialises in governance implementation and can accelerate your Purview deployment. - Create data dictionary and metadata standards
Document every table, column, and metric with business definitions, owners, and usage guidelines. Establish naming conventions and metadata standards. Make this accessible to users via Purview or a wiki. - Define data stewardship roles and responsibilities
Assign data owners and stewards for each domain. Document their responsibilities: data quality, access approvals, metadata maintenance, and policy enforcement. Clear ownership prevents data becoming orphaned. - Establish data quality monitoring and alerting
Build automated checks into Fabric pipelines to monitor quality metrics (completeness, accuracy, timeliness). Configure alerts when quality degrades. This enables proactive issue detection and resolution. - Plan audit logging and compliance reporting
Enable Fabric audit logging to track user actions, data access, and changes. Configure logs to flow to Azure Monitor or a SIEM for analysis. Generate compliance reports for audits and regulatory reviews. - Document data policies and usage guidelines
Create policies governing data access, sharing, retention, and usage. Communicate these clearly to users. Policies should address BYOD, external sharing, and acceptable use.
Encryption and Network Security
- Plan encryption at rest and in transit
Ensure Fabric uses encryption for data at rest (via Azure Storage Service Encryption) and in transit (TLS 1.2+). Plan how sensitive data will be encrypted before ingestion if needed. Document encryption key management. - Configure virtual network integration and private endpoints
If required by compliance, set up private endpoints for Fabric to isolate traffic from the public internet. Plan network routing and firewall rules. This adds complexity but provides network-level security. - Plan firewall rules and IP whitelisting
Document which IP ranges and services need access to Fabric. Configure firewall rules accordingly. Maintain an inventory of approved IP addresses and review regularly. - Establish secrets management (Key Vault)
Store database credentials, API keys, and connection strings in Azure Key Vault, not in code or configuration files. Configure Fabric pipelines to retrieve secrets from Key Vault at runtime. Rotate secrets regularly.
Infrastructure and Environment Setup
Prepare your cloud and on-premises infrastructure to support Fabric deployment.
Azure Environment Preparation
- Provision or validate Azure subscription and resource groups
Ensure you have an active Azure subscription with sufficient quota for Fabric capacity, storage, and compute resources. Organise resources into logical resource groups by environment (dev, test, prod) and workload. Document naming conventions and tagging strategy. - Set up Fabric capacity and workspaces
Provision Fabric capacity at the appropriate SKU level. Create workspaces for different teams or domains. Configure workspace admins and access. Document workspace structure and naming conventions. - Configure storage accounts and OneLake setup
Ensure OneLake (Fabric’s unified data lake) is configured with appropriate access controls and redundancy. If using external Azure Storage accounts for integration, configure connection strings and managed identities. - Plan and configure backup and disaster recovery
Document Fabric’s built-in redundancy and backup capabilities. Plan recovery time objectives (RTO) and recovery point objectives (RPO) for critical data. Test recovery procedures before production cutover. - Set up monitoring and logging infrastructure
Configure Azure Monitor, Application Insights, and Log Analytics to collect Fabric telemetry. Set up dashboards and alerts for key metrics (capacity utilisation, query performance, errors). Enable diagnostic logging. - Configure cost management and budget alerts
Set up Azure Cost Management to track Fabric spending. Create budgets and configure alerts when costs exceed thresholds. Review cost drivers (capacity, storage, compute) regularly.
Integration with Existing Systems
- Plan data ingestion from on-premises systems
If migrating from on-premises databases, plan connectivity via Azure Data Factory, on-premises data gateway, or direct cloud connectivity. Test network latency and throughput. Plan for incremental loading to minimise bandwidth. - Configure integration with third-party data sources
Document all external data sources (SaaS applications, APIs, cloud services). Plan connectors and refresh schedules. Test connectivity and data quality from external sources. - Set up data pipeline orchestration
Plan how pipelines will be orchestrated in Fabric (native Fabric pipelines, Azure Data Factory, or hybrid). Design error handling, retry logic, and notifications. Document dependencies between pipelines. - Plan real-time data streaming setup
If using event hubs or streaming data sources, configure event hub namespaces, topics, and consumer groups. Plan how streaming data will be ingested into Fabric (via KQL databases or direct ingestion). - Establish development, test, and production environments
Create separate Fabric workspaces or subscriptions for dev, test, and production. Document environment promotion processes. Ensure production is isolated and tightly controlled.
Migration Execution Planning
With architecture and infrastructure in place, plan the detailed execution of migration activities.
Tool Selection and Validation
- Evaluate and select migration tools
Consider tools like Fabric Migration Assistant for Data Warehouse for warehouse migrations, Azure Data Factory for pipeline migration, or third-party tools. Test tools with sample data. Document tool selection rationale and limitations. - Review Microsoft’s official migration guidance
Consult Microsoft’s migration guides and best practices to ensure your approach aligns with Microsoft recommendations. Stay updated on new tools and features. - Plan for legacy system decommissioning
Document the process for retiring legacy systems once migration is complete. This includes data archival, system shutdown, and vendor contract termination. Plan decommissioning in phases to avoid orphaning dependent systems. - Validate migration tools with POC
Run a proof-of-concept migration using actual data and tools. Measure performance, data accuracy, and effort. Identify issues before full-scale migration. - Before executing cutover, run through the Microsoft Fabric migration checklist to confirm tool selections and rollback plans
This pre-cutover sanity check reduces last-minute surprises and ensures teams are aligned.
Build and Testing Planning
- Create detailed ETL/ELT rebuild specifications
Document how each legacy pipeline will be rebuilt in Fabric. Include source-to-target mappings, transformation logic, error handling, and scheduling. Use these specifications to guide development. - Plan unit testing and integration testing strategy
Define test cases for each pipeline, transformation, and report. Plan integration tests that validate end-to-end data flows. Establish acceptance criteria for test completion. - Plan user acceptance testing (UAT)
Identify business users who will validate migrated reports and data. Prepare test scenarios and expected results. Schedule UAT sessions and document feedback. - Plan performance testing and tuning
Design performance tests that simulate production workloads. Measure query response times, report refresh times, and system throughput. Identify bottlenecks and optimisation opportunities. Document tuning recommendations. - Prepare cutover and rollback procedures
Document step-by-step cutover procedures, including data synchronisation, system validation, user communication, and rollback triggers. Identify what issues would trigger rollback. Practise cutover procedures in test environments.
Change Management and User Enablement
Technical migration is only half the battle. User adoption and change management are equally critical.
Training and Documentation
- Develop training curriculum for different user roles
Create role-specific training for data engineers (Fabric development), analysts (report building), and business users (consuming reports). Include hands-on labs and real-world examples. Tailor content to your industry and use cases. - Create comprehensive documentation and runbooks
Document Fabric architecture, data models, common tasks, troubleshooting steps, and support contacts. Maintain a wiki or knowledge base accessible to all users. Update documentation as systems evolve. - Plan knowledge transfer sessions with teams
Conduct workshops and training sessions for data teams, report developers, and business users. Include Q&A time and hands-on practice. Record sessions for future reference. - Develop quick reference guides and cheat sheets
Create one-page guides for common tasks (accessing data, building reports, scheduling refreshes). Make these easily accessible in your knowledge base or Slack. - Establish support and escalation procedures
Define first-line support (help desk, internal team), second-line support (Fabric experts), and escalation paths. Document response times and SLAs. Provide contact information to all users.
Communication and Stakeholder Management
- Create communication plan and stakeholder updates
Develop a communication strategy that keeps stakeholders informed throughout migration. Include regular updates, success stories, and timeline changes. Use multiple channels: email, town halls, Slack, dashboards. - Plan launch events and go-live announcements
Prepare announcements for launch milestones. Celebrate successes and recognize teams. Communicate what’s new, how to access Fabric, and where to get help. - Establish feedback mechanisms and issue tracking
Create channels for users to report issues, suggest improvements, and ask questions. Use a ticketing system to track and resolve issues. Act on feedback to improve adoption. - Plan post-migration support and stabilisation period
Plan extended support (2-4 weeks) after cutover to address issues and help users. Monitor system health closely. Gradually transition to steady-state operations.
Data Migration and Validation
Now for the critical data migration activities.
Pre-Migration Validation
- Perform final data source validation
Verify that all data sources are accessible and data quality is as expected. Run final reconciliation checks. Document any last-minute data quality issues and mitigation plans. - Validate transformation logic and business rules
Confirm that all transformation logic has been coded correctly in Fabric. Test edge cases and exception handling. Validate calculated fields and aggregations against source systems. - Confirm capacity and resource availability
Verify that Fabric capacity is provisioned, network connectivity is tested, and all infrastructure is ready. Perform a final health check of all systems. - Obtain sign-off from stakeholders
Get formal approval from business owners, IT leadership, and compliance teams to proceed with cutover. Document sign-off and any conditions or concerns.
Data Migration Execution
- Execute initial data load
Run the full data migration from source systems to Fabric. Monitor for errors and performance issues. Document load times and resource utilisation. Automation capabilities in Fabric can streamline bulk migrations. - Perform comprehensive data validation
Compare source and target data using reconciliation queries. Validate row counts, data types, relationships, and calculated fields. Check for data loss or corruption. Document any discrepancies and resolutions. - Validate reports and analytics outputs
Run migrated reports and compare results to legacy reports. Verify calculations, filters, and formatting. Test interactivity and drill-downs. Ensure performance meets expectations. - Conduct full system integration testing
Test how Fabric integrates with downstream systems and applications. Validate data flows end-to-end. Test error scenarios and recovery procedures. - Monitor system performance and resource utilisation
Track CPU, memory, storage, and network utilisation during and after migration. Monitor query performance and identify slow-running queries. Make tuning adjustments as needed.
Post-Migration Validation
- Perform UAT sign-off
Have business users formally validate migrated systems and sign off. Document any issues and remediation. Obtain acceptance from business stakeholders. - Execute final reconciliation and audit
Perform final data reconciliation to confirm all data migrated correctly. Generate audit reports documenting migration activities, changes, and validations. Archive documentation for compliance. - Validate governance and security controls
Confirm that access controls, encryption, audit logging, and compliance controls are functioning correctly. Test access restrictions and validate that data is properly classified. - Decommission legacy systems
Once migration is validated and users are transitioned, shut down legacy systems. Archive data as required. Document decommissioning and capture lessons learned.
Governance, Monitoring, and Optimisation
Post-migration, ongoing governance and optimisation ensure long-term success.
Ongoing Governance
- Establish data governance operating model
Define roles (data owners, stewards, custodians) and responsibilities. Schedule regular governance reviews. Maintain data catalogue and metadata. Enforce data policies. - Monitor data quality and compliance
Run daily quality checks and compliance audits. Generate reports on data quality metrics and compliance status. Address quality issues promptly. - Review and update access controls regularly
Conduct quarterly access reviews to ensure users have appropriate permissions. Remove access for departed employees. Update role-based access as job functions change. - Maintain audit logs and compliance documentation
Ensure audit logging is enabled and logs are retained per policy. Generate compliance reports for audits and regulatory reviews. Investigate any suspicious activities.
Performance Monitoring and Optimisation
- Set up continuous performance monitoring
Monitor key metrics: query response times, report refresh times, capacity utilisation, and error rates. Set up alerts for anomalies. Create dashboards for visibility. - Identify and optimise slow-running queries
Use Fabric’s query performance insights to identify slow queries. Analyse execution plans and optimise indexes, statistics, or query logic. Document optimisations. - Right-size capacity and manage costs
Review capacity utilisation monthly. Adjust SKU if consistently over or underutilised. Identify cost drivers and optimisation opportunities. Implement cost management best practices. - Plan for growth and scalability
Forecast data growth and user growth. Plan capacity increases proactively. Design architecture to scale horizontally when needed.
Continuous Improvement
- Gather user feedback and identify improvement opportunities
Conduct surveys and focus groups to understand user satisfaction and pain points. Prioritise improvements based on impact and effort. Implement quick wins regularly. - Stay current with Fabric feature releases and updates
Monitor Microsoft’s Fabric roadmap and release notes. Evaluate new features for applicability to your use cases. Plan adoption of valuable new capabilities. - Conduct regular architecture reviews
Periodically review your Fabric architecture against evolving business needs and technology landscape. Identify refactoring opportunities or architectural improvements. - Document lessons learned and best practices
Capture lessons from migration and post-migration operations. Document best practices for future projects. Share knowledge across teams.
Downloadable Migration Checklist Summary
Below is a summary of all checklist items organised by category for easy reference and printing. This downloadable Microsoft Fabric migration checklist summary is a quick reference to keep teams aligned during migration.
Pre-Migration Assessment
- Define migration goals and success metrics
- Secure executive sponsorship and funding approval
- Identify all stakeholder groups and their requirements
- Establish a migration governance committee
- Document business-critical applications and dependencies
- Audit existing data sources and systems
- Measure current data volumes, growth rates, and velocity
- Document existing ETL/ELT processes and workflows
- Assess current reporting and analytics infrastructure
- Review current data governance and security posture
- Identify performance bottlenecks and optimisation opportunities
Technical Architecture and Planning
- Define your Fabric architecture (Lakehouse vs. Data Warehouse vs. Mirroring)
- Design data ingestion strategy (real-time vs. batch vs. streaming)
- Plan capacity sizing and resource allocation
- Design lakehouse folder structure and naming conventions
- Plan data warehouse schema design and relationships
- Evaluate AI and advanced analytics integration points
- Design real-time analytics and streaming architecture
- Select migration strategy (Big Bang vs. Phased vs. Parallel)
- Prioritise workloads and data domains for migration
- Create detailed migration timeline and milestones
- Plan for parallel running periods if needed
- Identify quick wins and proof-of-concept opportunities
Data Assessment and Preparation
- Conduct data quality assessment across all source systems
- Identify and document data quality rules and thresholds
- Plan data cleansing and transformation logic
- Test data migration with sample datasets
- Establish data reconciliation processes
- Document data lineage and transformations
- Classify data by sensitivity and compliance requirements
- Identify personally identifiable information (PII) and sensitive data
- Plan data masking and anonymisation strategies
- Ensure compliance with Australian Privacy Principles and regulations
- Plan data retention and archival policies
Governance and Security Framework
- Design role-based access control (RBAC) model
- Plan Azure AD integration and service principal setup
- Configure row-level security (RLS) and column-level security (CLS)
- Plan multi-factor authentication (MFA) enforcement
- Establish service account and API authentication strategy
- Implement Microsoft Purview for data governance
- Create data dictionary and metadata standards
- Define data stewardship roles and responsibilities
- Establish data quality monitoring and alerting
- Plan audit logging and compliance reporting
- Document data policies and usage guidelines
- Plan encryption at rest and in transit
- Configure virtual network integration and private endpoints
- Plan firewall rules and IP whitelisting
- Establish secrets management (Key Vault)
Infrastructure and Environment Setup
- Provision or validate Azure subscription and resource groups
- Set up Fabric capacity and workspaces
- Configure storage accounts and OneLake setup
- Plan and configure backup and disaster recovery
- Set up monitoring and logging infrastructure
- Configure cost management and budget alerts
- Plan data ingestion from on-premises systems
- Configure integration with third-party data sources
- Set up data pipeline orchestration
- Plan real-time data streaming setup
- Establish development, test, and production environments
Migration Execution Planning
- Evaluate and select migration tools
- Review Microsoft’s official migration guidance
- Plan for legacy system decommissioning
- Validate migration tools with POC
- Create detailed ETL/ELT rebuild specifications
- Plan unit testing and integration testing strategy
- Plan user acceptance testing (UAT)
- Plan performance testing and tuning
- Prepare cutover and rollback procedures
Change Management and User Enablement
- Develop training curriculum for different user roles
- Create comprehensive documentation and runbooks
- Plan knowledge transfer sessions with teams
- Develop quick reference guides and cheat sheets
- Establish support and escalation procedures
- Create communication plan and stakeholder updates
- Plan launch events and go-live announcements
- Establish feedback mechanisms and issue tracking
- Plan post-migration support and stabilisation period
Data Migration and Validation
- Perform final data source validation
- Validate transformation logic and business rules
- Confirm capacity and resource availability
- Obtain sign-off from stakeholders
- Execute initial data load
- Perform comprehensive data validation
- Validate reports and analytics outputs
- Conduct full system integration testing
- Monitor system performance and resource utilisation
- Perform UAT sign-off
- Execute final reconciliation and audit
- Validate governance and security controls
- Decommission legacy systems
Governance, Monitoring, and Optimisation
- Establish data governance operating model
- Monitor data quality and compliance
- Review and update access controls regularly
- Maintain audit logs and compliance documentation
- Set up continuous performance monitoring
- Identify and optimise slow-running queries
- Right-size capacity and manage costs
- Plan for growth and scalability
- Gather user feedback and identify improvement opportunities
- Stay current with Fabric feature releases and updates
- Conduct regular architecture reviews
- Document lessons learned and best practices
Conclusion
Migrating to Microsoft Fabric is a significant undertaking that requires meticulous planning, technical expertise, and organisational alignment. This comprehensive checklist provides a structured framework to guide your migration journey from initial assessment through post-migration optimisation.
The most successful Fabric migrations share common characteristics: clear business objectives, strong executive sponsorship, detailed planning, rigorous testing, comprehensive change management, and ongoing governance. By working through each section of this checklist methodically, you’ll significantly reduce risk and increase the likelihood of a smooth, successful transition.
Remember that migration is not a one-time event but the beginning of an ongoing journey of optimisation and innovation. After cutover, continue to monitor performance, gather user feedback, and evolve your architecture to meet changing business needs. Stay engaged with Microsoft’s latest Fabric capabilities and best practices to maximise your investment. Use this Microsoft Fabric migration checklist as your living document update it as you learn and as Microsoft Fabric evolves.
For organisations in Australia seeking expert guidance on Fabric migration, Agile Insights brings deep Microsoft expertise and industry-specific accelerators to accelerate your journey. Whether you need help with architecture design, governance implementation via Microsoft Purview, or comprehensive migration execution, our team of Microsoft-certified consultants can partner with you to deliver measurable outcomes.
Start with this checklist, engage your stakeholders, and take the first steps toward a modern, agile data platform that drives competitive advantage and business value.
