--- title: "Microsoft Fabric Consulting and Implementation | Smart Data" url: https://www.smartdata.net/services/microsoft-fabric-consulting description: "Microsoft Fabric consulting: honest platform fit, migration from SSIS-era pipelines, Power BI on Fabric and a governed base for AI." source: "Smart Data (smartdata.net). Markdown copy of the page; the HTML page is canonical." --- Microsoft Fabric # Microsoft Fabric consulting One platform for engineering, warehousing and BI on the Microsoft estate you already pay for. We build the architecture around it, and build it to be operated after we hand it back. For a company that has never had a real data platform, Fabric is the affordable way to get a complete one, with governance and security included from the start. Start where you are ## Your starting point shapes the platform Not every Fabric implementation begins with a sophisticated data estate. We have Power BI everywhere and the reports argue with each other. The problem is upstream of the dashboard, so that is where we start. Start with: sources, reporting needs and ownership. We have years of ADF, SSIS, Synapse or Data Explorer and somebody still babysits it. This is the hard part, so we map what runs today before anything moves. Start with: the current architecture and a migration sequence. We are not sure Fabric is the right move, or it is in and the bill keeps climbing. We’ll tell you whether Fabric is the shortest path for your estate, and when it isn’t. Start with: the constraints around performance and operations. Fabric delivery ## SAP into Fabric, checked against SAP Six operating entities of a global industrial group share one SAP ECC 6.0 system, under strict rules on how data leaves SAP. We built a repeatable path from SAP into Microsoft Fabric that fits the group’s own BI standards, with entity-level security on the sensitive data. The client’s Basis and SAP teams found the process much easier than they expected. - SAP ECC 6.0 - OneLake shortcuts - Extracted through the Microsoft SAP connector and its ABAP add-on, over the on-premises data gateway - Landed as CSV through the corporate ingestion framework - Loaded into Bronze, one workspace set per operating entity - Row counts checked against SAP - Record spot checks - Closed-period comparison against SAP - Wages and project financials kept inside each entity Next phase: production, the shared Gold layer and the first project P&L reports. ## Most Fabric problems start in the architecture Capacity design, workload isolation, data modeling, governance and CI/CD decide how well it runs. Microsoft Fabric services ## Microsoft Fabric services We design, implement and operate Fabric as an enterprise data platform, with Power BI as one workload on it. We select the services each workload needs, in an architecture that stays maintainable as adoption grows. Architecture and readiness ### Choose the platform structure and delivery sequence around your workloads and the Microsoft environment you already run Strategy, readiness and architecture Whether Fabric is the shortest path for your estate, and the architecture if it is. Data platform and medallion design Raw, validated and business-ready layers, so a transformation is built once. Capacity planning and workload isolation One heavy workload can’t slow the others on a shared capacity. - Platform evaluation - Medallion architecture - SKU and region - Roadmapping Build and migrate ### Bring sources, pipelines and models into a maintainable environment, with the business logic carried forward Implementation and migration ADF, Synapse, SSIS and Data Explorer workloads moved across. Lakehouse, Eventhouse and data engineering OneLake storage and Spark, with Eventhouse for real-time and IoT telemetry. Data ingestion and transformation Data Factory pipelines that land sources raw and conform them. - OneLake - Lakehouse - Spark - Data Factory Security and governance ### Build data ownership, identity, access and quality into the platform from the start Security, identity, networking and governance Entra ID, role-based access, managed identities, Key Vault, private networking and Purview. Data residency and multinational architecture Region and residency settled before anything is provisioned. - Microsoft Entra ID - Purview - Key Vault - Workspace boundaries Operate and improve ### Make the platform understandable to the people who will run and change it Power BI and semantic-model modernization Semantic models and Direct Lake rebuilt upstream of the dashboard. CI/CD and Infrastructure as Code Source control, testing and deployment pipelines, so the platform can be safely changed after it reaches production. Monitoring, optimization and production support Utilization monitored and capacity right-sized. - Direct Lake - Semantic models - Deployment pipelines - Monitoring Power BI and analytics on Fabric ## Power BI is part of Fabric, so build it that way Reports sit on shared semantic models over one copy of the data in OneLake, so finance and operations read the same number. We modernize the Power BI estate you have and connect it to the platform underneath. [Talk to us about your Power BI estate](https://www.smartdata.net/contact-us) ### Semantic models built once Business definitions agreed upstream, so two reports stop arguing. ### Direct Lake where it pays Reports read OneLake tables directly, without import copies to refresh. ### Security set once Row-level security and sensitivity labels set in Fabric, not in every report. How data moves ## Designed around how data moves through the organization Each transformation is built once in the platform, and every report, alert and model reads from it. Sources - SaaS applications - SAP via gateway - Files and feeds - Device telemetry Raw source data Landed as it arrived, before any checks, with nothing discarded. Lakehouse · OneLake Validated and conformed Cleaned and joined into consistent business entities, so every team works from one definition of each number. Spark · Data Factory pipelines Business-ready Semantic models and Direct Lake serving Power BI, dashboards, alerting and feature engineering. Direct Lake · Semantic models Consumers - Executive reporting - Self-serve Power BI - AI and agents - Operational apps Our process ## How our Fabric engagements work The phases adjust to where you are starting from. The first settles the decisions that are expensive to reverse. ### Discovery and architecture review Decision point · Written plan We settle capacity topology and SKU, data residency and region, which source systems land where, and whether production data is permitted in non-production. **Deliverable:** an architecture decision record and a capacity and region plan. ### Build and parallel run Working solution · Measurable outcome We build the medallion architecture in Fabric, move existing workloads across and rebuild the semantic models reporting reads from. **Deliverable:** a working environment and a parallel-run report showing the numbers match. ### Ongoing support and optimization Expand what works · Embed into operations After go-live there are usually more domains to bring in and capacity to right-size. We can stay on through managed services with P1 to P4 response times, or enable your team to run and extend it independently. **Deliverable:** a runbook your team can operate from. Migrating to Fabric ## Fabric does not always start from a blank slate We assess what should migrate, what should be redesigned and what should stay where it is, and sometimes a workload is best left alone. Where you are - Azure Data Factory - Synapse and Data Explorer - SSIS, SSAS and SSRS - SAP through the on-prem gateway - Power BI on extracts - 01 Inventory - 02 Dependency map - 03 Rebuild - 04 Parallel run - 05 Cutover Parallel run exists so that when something is wrong you find out on a Tuesday instead of at quarter close. Where you land - Eventhouse - Data Factory pipelines Industries ## Industries we work in ### Healthcare Patient, pharmacy and clinical data systems held to HIPAA, including where AI is allowed to touch them. - Medicaid programs - Pharmacy services - Clinical documentation [Healthcare](https://www.smartdata.net/industries/healthcare) ### Manufacturing and industrial ERP, shop-floor and supplier data on one platform, so plant reporting and finance reporting reconcile. - SAP and ERP reporting - Shop-floor integration - Supplier portals [Manufacturing and industrial](https://www.smartdata.net/industries/manufacturing) ### Retail and distribution Order, inventory, credit and customer data connected across portals, branches and the platforms behind them. - Vendor and customer portals - Order and credit systems - Data platforms [Retail and distribution](https://www.smartdata.net/industries/retail) ### Financial services Bank data platforms, reporting that holds up to an audit and managed support for the financial systems the business runs on. - Bank data platforms - Managed ERP support - Legacy migration [Financial services](https://www.smartdata.net/industries/banking-financial-services) ### Insurance Claims, member and policy systems for health plans and insurers. - Health plans - Claims data - Member apps [Insurance](https://www.smartdata.net/industries/insurance) A few practical questions ## Work through the decisions Common questions about Fabric consulting, migration, capacity planning and governance. Workload shape and audience size set it more than data volume. Below an F64 capacity every report viewer needs a paid Pro license; at F64 and above, viewing is included. Sizing is part of the scoping work rather than a number on a web page. If you run on Microsoft, start on Fabric and bring Databricks in when your data work outgrows it, for heavy machine learning or streaming. The two work together, so what you’ve built carries over. If machine learning is the main workload, or you are genuinely multi-cloud, start on Databricks. We build on both. Where the workload fits. Synapse pipelines, dedicated SQL pools and Data Explorer clusters have direct landing places in Data Factory, the Fabric warehouse and Eventhouse. Some should remain where they are, and the assessment says which. No. Not every Fabric implementation begins with a sophisticated data estate. We bring data from the systems you already use, from ERP and CRM to timekeeping and invoicing, into one governed architecture, and build toward what your team can run. Yes. Eventhouse and KQL take streaming data from devices, applications and logs alongside the batch data in the Lakehouse, so operational dashboards, alerting and historical analysis run on one platform. We size that ingestion against capacity from the start, because streaming workloads are the easiest way to crowd out everything else. Shared capacity means a poorly planned workload can hurt everything else running on it, so we treat capacity as architecture. That can mean separating critical workloads, setting workspace and environment boundaries, monitoring utilization and designing ingestion and transformation to use resources efficiently. We design them into the platform from the first workspace, and you can reuse the Entra ID security groups you already have. Our work covers Microsoft Entra ID, role-based and workspace security, service principals and managed identities, Azure Key Vault, private networking, Microsoft Purview and regional data-residency requirements. Yes. Land the raw data, clean it once and publish the version people and AI models can rely on. The pattern works the same on Fabric, Databricks or Snowflake, and it keeps the evidence, so a mistake in your logic doesn’t become a permanent mistake in your history. No. Fabric coexists with Azure Data Factory, Synapse and Data Explorer during a migration. Pick one use case, land it, then move the next. Your team, if that is the plan, and we build toward it from week one: source control, deployment pipelines and a named owner on your side for every service principal and secret. If you would rather we ran it, managed services covers that. Through the same discipline as any production software: source control, CI/CD and deployment pipelines, automated testing, Infrastructure as Code, separate development, test and production environments, and monitoring. A rough list of the systems that feed your reporting today, the reports people argue about, who owns the data and what Microsoft licensing and capacity you already have. If data has to stay in a particular region, say so early. The first call works from whatever you have. One useful thing in ninety days is the target, not a full platform. If a proposal promises a finished warehouse in that window, read it twice. Keep exploring ## Other platforms and related services We build on Fabric, Databricks and Snowflake, so we recommend the engine that fits your estate. Platforms we build on - [Databricks](https://www.smartdata.net/services/databricks-consulting) - [Snowflake](https://www.smartdata.net/services/snowflake-consulting) Working through a larger migration or an engineering-heavy estate? ### Data engineering Pipelines off SSIS and legacy ETL, and warehouses that reconcile. [Data engineering](https://www.smartdata.net/services/data-engineering) ### AI on your data Governed, private AI on the Fabric estate, with the query behind every answer. [AI on your data](https://www.smartdata.net/services/ai-on-your-data) ### ERP and enterprise integration SAP, NetSuite, Dynamics and Epicor data landed in the platform. [ERP and enterprise integration](https://www.smartdata.net/services/enterprise-integrations) Next step ## What should your Fabric platform make possible? Settling capacity, residency and sequencing at the start costs less than correcting them later. Here is what happens after you reach out. ### A short call about your Microsoft estate What runs on Azure and Power BI today, what Fabric should change and what is driving the timing. ### An engineer looks at what’s running Scope comes from your systems as they are, before anything is priced. ### You get a written proposal Covering scope, team, approach and price. [info@smartdata.net](mailto:info@smartdata.net) [(937) 886-9166](tel:+19378869166)