Microsoft Fabric Project Planning Tips & Checklist

Are you excited about adopting Microsoft Fabric as your end-to-end data analytics platform and just getting started planning a new Microsoft Fabric initiative, program or project? Are you concerned whether you are thinking about all the right questions or topics to position your Microsoft Fabric project for success? Are you looking for Microsoft Fabric Project Planning Tips (or Template) or a Microsoft Fabric Project Planning Checklist (or Questionnaire)? If so, you are in the right place. 

At XTIVIA, we have a long history of delivering successful data analytics projects from data warehouses to data lakes, from data governance to master data management, from dashboards to reports, and in recent years data science and ML/AI. A successful implementation starts with laying the right foundation, and it is important to ensure that you are considering what is best for your organization based on your needs and requirements—yes, your organization, as this is not a one-size-fits-all scenario.

The early phases of a Fabric implementation involve key decisions around overall architecture, data ingestion, data modeling, data storage, security, governance, licensing, and more. Unfortunately, many customers rush into Microsoft Fabric projects without proper planning and pay the price for poor architectural choices and lack of planning. These early missteps can cause all kinds of issues, both in the short and long run—timeline delays, increased costs, rework, and difficult-to-maintain-and-extend systems.

In this article, we discuss the typical topics and questions we cover with our customers in the XTIVIA Microsoft Fabric Jumpstart Program (or Quickstart or Kickstart), and this should be helpful for you regardless as you embark on your Microsoft Fabric journey.

Microsoft Fabric Project Planning Tips/Checklist: Table of Contents

Our Microsoft Fabric Jumpstart (aka Quickstart) engagement starts with a discovery phase where we attempt to understand your objectives, current landscape, compliance and security constraints, and more. This is then followed by an architecture and planning phase where we jointly work with our customers to make key Microsoft Fabric architecture decisions and plan out the implementation phase.

1. Business Objectives & Goals
2. Data Landscape & Inventory
3. Data Governance & Security
4. Technology & Infrastructure
5. Data Modeling, Storage, Consumption & Reporting
6. Miscellaneous Topics
7. Summing It Up!

Microsoft Fabric Project Planning Tips/Checklist: Business Objectives & Goals

This section delves into the core motivations and desired outcomes of your Microsoft Fabric implementation. It’s crucial to clearly define these to ensure the project aligns with your overall business strategy.

• What are the key business challenges you are trying to address with Fabric? Some examples include:
  • Data Silos: Are different departments working with disparate data tools and isolated data stores, hindering collaboration and a unified view of the business?
  • Slow Insights: Do you face delays in obtaining critical insights from your data, impacting decision-making speed?
  • Lack of Data Governance: Is there a lack of standardized processes for data quality, security, and access control?
  • Poor Data Quality: Are data inconsistencies, inaccuracies, and incompleteness impacting the reliability of your analyses?
  • Multiple Data Copies: Do you have too many data tools each operating with their own copies of the same or overlapping data? 
• What are the specific business outcomes you expect to achieve? Some examples include:
  • Improved Decision-Making: How will Fabric empower data-driven decisions across the organization? Will it enable faster, more informed choices in areas like product development, marketing, and operations?
  • Increased Revenue: How will Fabric contribute to revenue growth? For example, by identifying new market opportunities, optimizing pricing strategies, or improving customer retention.
  • Reduced Costs: How will Fabric help you reduce costs? By streamlining data processes, improving operational efficiency, or identifying areas for cost optimization.
  • Enhanced Customer Experience: How will Fabric help you better understand your customers and deliver a superior experience? By personalizing interactions, improving customer service, or identifying areas for product improvement.
• What are the key performance indicators (KPIs) that will measure the success of the Fabric implementation? Some examples include:
  • Time to Insight: Measure the time it takes to generate insights from data using Fabric compared to previous methods.
  • Number of Data Products Delivered: Monitor the number of new data products (dashboards, reports, models) created and utilized within the organization.
  • Data Quality Improvement: Track improvements in data accuracy, completeness, and consistency over time.
• What is the overall scope for this initiative (What subject areas, applications, and business units are in scope for this initiative)?

Microsoft Fabric Project Planning Tips/Checklist: Data Landscape & Inventory

This section focuses on understanding the current state of your data assets. A thorough understanding of your data landscape is crucial for successful Fabric implementation. First, start by looking into any existing documents that may be relevant to the current project scope and project goals (system context diagrams, data model(s), data flow diagrams, design documents, runbooks, mapping specifications, etc.).

a) Current State of Your Data Landscape

• Number of Data Sources: How many different data sources do you currently have? Consider various types like databases (relational, NoSQL), data warehouses, data lakes, cloud storage, SaaS applications, semi-structured and unstructured files, and more.
• Types of Data: What types of data do you deal with? Structured (e.g., relational databases), semi-structured (e.g., JSON, XML), unstructured (e.g., text, images, audio), and streaming data are common types.
• Data Volumes: Estimate the volume of data you generate and process daily, weekly, or monthly. This helps in determining the necessary infrastructure capacity for Fabric.
• Data Locations: Where is your data currently stored? On-premises, in the cloud, or a hybrid environment? Understanding data locations is essential for data integration and movement strategies.

b) Data Inventory

• Have you conducted a data inventory? A comprehensive data inventory provides a detailed catalog of all your data assets, including their location, format, ownership, and usage.
• If so, please provide details. If you have already conducted an inventory, provide information on the tools used, the scope of the inventory, and the key findings.
• If not, consider these steps:
  • Identify all data sources: Document all systems, applications, and databases that contain data relevant to your business and this data initiative.
  • Classify data types: Categorize data based on its type, sensitivity, and usage.
  • Assess data quality: Evaluate data accuracy, completeness, consistency, and timeliness.
  • Document data lineage: Track the origin and transformation of data across different systems.

c) Critical Data Sources

• Identify the most important data sources for your business. These are the sources that drive key business decisions and operations. Examples include:
  • Customer Relationship Management (CRM) systems: Salesforce, Microsoft Dynamics 365
  • Enterprise Resource Planning (ERP) systems: SAP, Oracle E-Business Suite
  • Marketing Automation platforms: Marketo, HubSpot
  • E-commerce platforms: Magento, Shopify
  • Social Media platforms: Facebook, Twitter
  • Financial systems: General Ledger, Accounts Receivable

d) Data Quality

• What are the data quality challenges within your organization? (e.g., Inconsistency, inaccuracy, incompleteness)

e) Master or Key Data Entities 

What are the master data entities/subject areas that are critical to your business operations (examples: student, customer, vendor, facilities, locations, products, services offered, etc), where do they exist currently (which data source), and which ones are in scope for the current data engagement? Do you have these duplicated across data sources as well?

f) Data Ingestion into Microsoft Fabric

1. Here, we try to get more specific and zero in on the particular data we need to ingest into Microsoft Fabric as part of this data initiative. For instance, you may have 400 entities in a data source, but you may not need all of them to be available in Fabric. So, what specific entities would you like to focus on first for ingestion into Fabric? 
2. Total source table / file counts by type (dimensional data vs factual data) which need to be ingested into Microsoft Fabric. Or from how many APIs we need to ingest data into Microsoft Fabric (assuming there is no direct table access from the data sources)?
3. What are the different ways we will get to the data in the source systems? Database access? API connectivity? Any flat file data sources?
4. Approximate count of structured source tables / files in scope (regular tables, csv files, etc. without complex data types, without XML/JSON type of data, without other special data types like variant, BLOBs, CLOBs, without unstructured data like images, pdf, text data, etc.
5. Approximate count of tables / files with semi-structured data (XML, JSON, etc.).
6. Approximate count of tables / files with unstructured data (Textual data, PDFs, BLOBs, CLOBs, Binary data, images, audio, etc.)
7. Where are your data sources currently (on-premise / cloud)?
8. Approximately, what is the total size of your current data sources (GB / TB)? What is the approximate daily volume/size per data source that needs to be ingested into Microsoft Fabric?
9. Do your existing data sources allow “incremental pull” of data? Can we do “incremental pushes”?&nbs
10. How frequently do we need to update / add data into Microsoft Fabric (Once a day, hourly, real-time, etc.)?
11. Is there a specific time of day by when all the data needs to be available in Microsoft Fabric (usually applicable for batch data ingestion)?
12. Are there any special “data issue” notification requirements (regarding data anomalies, integrity issues, SLA, etc.)?
13. Are you aware of the trade-offs and do you have any preference between the various mechanisms for ingesting data into Fabric—data mirroring, shortcuts, event stream, data pipelines, dataflows gen2, notebooks?

g) Data Transformation & Processing

• What data transformations are required? (e.g., data cleansing, enrichment, aggregation, filtering)
• Are there any specific performance requirements for data processing? (e.g., latency requirements, throughput needs)
• Are there any existing data pipelines or ETL processes? If so, how are they currently managed?
• Are you aware of the trade-offs and do you have any preference between the various mechanisms for moving/transforming data within Fabric—shortcuts, data pipelines, dataflows gen2, notebooks?

By thoroughly understanding your data landscape and conducting a comprehensive data inventory, you can lay a strong foundation for your Fabric implementation. This will enable you to effectively integrate, transform, and analyze your data to achieve your desired business outcomes.

Microsoft Fabric Project Planning Tips/Checklist: Data Governance & Security

This section focuses on the critical aspects of ensuring data quality, security, and compliance within your Microsoft Fabric implementation.

a) Current Data Governance Framework

• Do you have an existing data governance framework? This framework outlines the policies, standards, and procedures for managing data within your organization.  
• If so, what are the key components? These might include data quality standards, data classification policies, data access controls, and data lineage documentation.  
• If not, what are the gaps in your current data governance practices? Identify areas where you need to improve data quality, security, and compliance.

b) Ensuring Data Security and Compliance

• How will you protect sensitive data within Fabric? In addition to data encryption (which is enabled by default in Microsoft Fabric), consider implementing measures such as:
  • Role-Based Access Control (RBAC): Granting users access to only the data they need to perform their job functions. This can involve access control at various levels:
    • Object level security, for example, tables
    • Row level security 
    • Column level security 
  • Data Masking: Obfuscating sensitive data such as social security numbers, dates of birth, and credit card numbers to prevent unauthorized disclosure.
• Are there any legal requirements to retain the data for a certain time-period (even if the data is not used / relevant for any reporting)?
• How will you comply with relevant regulations? Ensure that your Fabric implementation adheres to industry standards and regulations such as:
  • General Data Protection Regulation (GDPR)
  • California Consumer Privacy Act (CCPA)
  • Health Insurance Portability and Accountability Act (HIPAA)  
• How will you maintain data integrity and accuracy? Implement data quality checks and validation rules to ensure the accuracy and reliability of data within Fabric.  

c) Addressing Data Privacy

• How will you handle personally identifiable information (PII) within Fabric? Implement appropriate safeguards to protect the privacy of individuals.
• How will you respond to data subject access requests (DSARs)? Establish processes for handling requests from individuals to access, correct, or delete their personal data.

By addressing these data governance and security considerations, you can ensure that your Microsoft Fabric implementation is compliant, secure, and trustworthy.

Microsoft Fabric Project Planning Tips/Checklist: Technology & Infrastructure

This section delves into the technical foundation of your Microsoft Fabric implementation. Careful consideration of these aspects will ensure a smooth and successful deployment.

a) Existing Microsoft Technology Stack

• What Microsoft technologies are you currently using? This includes:
  • Cloud Platforms: Azure, Azure Synapse Analytics, Azure Data Factory, Azure Data Lake Storage
  • Data Warehousing: SQL Server, Azure SQL Database
  • Business Intelligence & Analytics: Power BI, Excel
  • Other Microsoft Products: Microsoft 365, Dynamics 365

b) Existing Enterprise Data Tools

• Do you have an existing Data Warehouse? If yes, are there any transition/co-exist plans?
  • How often is the current Data Warehouse updated (Real-time, Near Realtime, Daily, Weekly, etc.)? 
  • What are the data sources feeding this Data Warehouse?
• Do you currently have a Master Data Management tool at your organization?
• Do you currently have a Data Catalog tool at your organization?
• What Business Intelligence tool do you currently use in your organization? What are your transition/co-exist plans?

c) Network Security & Environments Requirements

• Do you need to ingest data from data sources that are in your own data center or a virtual private cloud? If so, you will need to use an on-premises data gateway and plan to set up these data gateways in each “location”.&;
• How many different Fabric environments do you plan to have (Dev, QA, Stage, Prod, etc.)? 

d) Ensuring Scalability and Performance

• What is the expected daily data volume from each API and non-API data source into Fabric? 
• What are the anticipated peak usage days and times of Microsoft Fabric?
• How will you ensure that Fabric can handle the volume and velocity of your data? Consider factors like data growth, peak usage times, and performance requirements in choosing the right Fabric capacity.
• How will you monitor and optimize Fabric performance? You will want to understand the Fabric Capacity Metrics App. 
• What are your disaster recovery and business continuity plans for Fabric? Ensure that your implementation is resilient to outages and disruptions.

Microsoft Fabric Project Planning Tips/Checklist: Data Modeling, Storage, Consumption & Reporting

This section delves into key questions related to data storage, data modeling, data analytics, data science, and reporting within your Microsoft Fabric implementation. 

a) Data Modeling & Storage

• What Fabric items will be used to store data—Data Warehouse, Data Lakehouse, and/or KQL Database? 
• What data models will be used within Fabric? (e.g., star schema, snowflake schema)
• How much historical data (last 1 year, 2 years, 3 years, etc.) needs to be available in Microsoft Fabric for downstream consumption?
• What Power BI storage mode will you use for your Semantic Models? Direct Lake, Import, or DirectQuery? 

b) Data Consumption 

• What are the number and types of users and systems that will be consuming data from Microsoft Fabric?
• What consumption use cases/applications does Microsoft Fabric need to support? Do any applications need to be modified to source the data from this new data platform?
• Are you creating/performing (or planning to create/perform) any AI/ML models/analysis? Do you have any tool preferences in this area?

c) Reporting 

• What reporting, dashboarding, and analytics tools do you currently use?
• What types of reporting are currently performed and what data sources are currently used for this (which APIs are used and are there any changes/transformations performed on the data before utilizing it for reporting purposes)?
• What is the frequency of this reporting? Who is the target audience of these reports?
• How are the current reports delivered—a reporting tool? A custom application?  
• What dimensions are the current reports/applications dependent on? Customer, Product, Location, Industry, Partner, Sales Channel?
• When Microsoft Fabric is in place, what reports and data consumption needs does it need to satisfy? 

Microsoft Fabric Project Planning Tips/Checklist: Miscellaneous Topics

a) User Needs & Skills

• What are the technical skills and experience levels of the users who will interact with Fabric?
• What are the user’s primary needs from Fabric? (e.g., data exploration, report creation, data visualization, machine learning)

b)  Fabric and Power BI Pro Licensing

• What Fabric capacity (or capacities) do you need? Are you ready to reserve capacity or do you prefer to pay-as-you-go initially? 
• Do you have existing Power BI capacities that you are transitioning to Fabric?
• How many users will be using Power BI? As authors/editors? As viewers? Do any of them have E5 licenses? 

c) Organization & Resources

• What is the project scope and timeline?
• What are the key milestones and deliverables?
• What is the project budget?
• Who are the key stakeholders and decision-makers?
• What is the project team structure? (e.g., Project manager, architect, data engineers, data scientists, business analysts)
• What internal and external resources will be utilized? (e.g., Microsoft consultants, training providers)

Microsoft Fabric Project Planning Tips/Checklist: Summing it Up!

As you can see, there are a number of key topics and decision points to be kept in mind when embarking on a Microsoft Fabric implementation program. If you are looking for the right Microsoft Fabric coaches and specialists to help set you up for success, then XTIVIA can help. 

We typically set up multiple discovery, architecture and planning sessions as part of a XTIVIA Microsoft Fabric kickoff though we do adjust these to fit the scope/breadth of your Fabric initiative; some common sessions include: 

Meeting	Category	Purpose
Project Kickoff	Generic	Introductions Kickoff phase overview Purpose and high-level objectives of this new data initiative (Understand current-state and high-level requirements) Overview of the source systems (immediate and longer-term)
Deep dive into source systems and data ingestion	Requirements	Cover the relevant questions from our questionnaire as they relate to the source systems
Current Pain Points & Concerns	Requirements	A discussion of current pain points with the existing data warehouse, data engineering tools/processes, data analysis tools, data team resources
Scope/Requirements (typically split into multiple sessions)	Requirements	Understand functional and non-functional requirements (data availability, validation/quality, consumption, storage, archival, security, performance, monitoring, etc.)
Initial Data Platform Pilot Requirements/Architecture	Requirements	Deep dive into the specific requirements and architectural approach for the “initial scope” (or pilot implementation) of the new data architecture
Microsoft Fabric Tenant Architecture	Microsoft Fabric Architecture	Tenants, capacities, workspaces, data domains
Microsoft Fabric Data Storage, Data Ingestion and Data ETL Architecture	Microsoft Fabric Architecture	Lakehouses vs warehouses vs KQL databases, data ingestion options (shortcuts, database mirroring, eventstream, data pipeline, dataflow gen2, notebook (python, tsql)), data etl preferences (data pipeline, dataflow gen2, notebook (python, tsql))
Microsoft Fabric Production/Non-Production Architecture, Git and Deployment	Microsoft Fabric Architecture	Deep dive into how many prod/non-prod environments we plan to have, and what source system environments will be integrated into each environment? On-premise data gateway? Integration with Git and deployment pipelines
Microsoft Fabric Security and Governance Architecture	Microsoft Fabric Architecture	Microsoft Fabric Authentication, Fabric Roles, Item permissions, Data security (object, row, column, masking), Auditing, Data Catalog
Data Modeling and Power BI Architecture	Microsoft Fabric Architecture	Data Modeling (who, approach, documentation), Medallion Architecture, Semantic Models, Reports, Dashboards
Overall Microsoft Fabric Architecture – Bringing it all Together	Microsoft Fabric Architecture	Summarize [Client] and XTIVIA key architectural/technical decisions, cover miscellaneous topics, and identify pending decisions (and plan to address them)
Development Process, Communications and Resource Planning	Planning	Process: Code management, Agile PM tool (JIRA or Azure DevOps or …), risk/issue tracking, change control Meetings: Weekly status, weekly technical architecture/design meetings, demos. Identify client participants in weekly status, demo, and technical meetings. Collaboration tools: Teams/Slack (chat), Zoom/Teams (web meetings), Confluence/SharePoint (for documentation) … Client Resources: PM, BA, Data Owners, Architect, Data Engineers, Data Analysts, BI Developers, QA/UAT …
XTIVIA Remote + Envs + Tools Access	Planning	Review and initiate Client process for setting up remote access for XTIVIA consultants including access to source systems, environments and dev tools.
Project Risks and Dependencies	Planning	Identify and discuss key risks and external project dependencies; in particular, are there any other data projects in flight, new systems/tools being implemented?

Please get in touch with us to learn how we can help with your Fabric implementation as your Microsoft Fabric partner.