What is Microsoft Fabric? Workloads, pricing and how it fits your stack

What Microsoft Fabric actually is

Microsoft Fabric is a unified, SaaS analytics platform. It brings data integration, engineering, warehousing, real-time analytics, data science and Power BI into one product, sitting on a single storage layer called OneLake. Instead of stitching Data Factory, Synapse, Databricks-style notebooks and Power BI together with your own glue, you get them all in one workspace with one bill.

The workloads inside Fabric

Fabric ships as a set of experiences that share the same storage, security and governance:

• Data Factory for ingestion and pipelines.
• Synapse Data Engineering for Spark notebooks and lakehouse transforms.
• Synapse Data Warehouse for SQL-first warehousing.
• Synapse Real-Time Analytics for streaming and event data.
• Synapse Data Science for ML models and MLflow.
• Power BI for semantic models, reports and dashboards.
• Data Activator for triggering actions from data changes.

Every workload reads and writes to the same OneLake storage, so there's no copying data between systems to move it from ingestion to reporting.

OneLake - the bit that makes it different

OneLake is the single storage layer under every Fabric workload, using the open Delta Lake format. Microsoft describes it as the "OneDrive for data" - one logical lake per tenant, with shortcuts to data in ADLS, S3 or GCS so you don't have to migrate it. That's the piece that removes most of the copy-paste between tools, and it's the biggest reason Fabric feels different from a traditional data warehouse project.

How Microsoft Fabric pricing works

Fabric is billed as capacity SKUs (F2, F4, F8, F16, F32, F64 and up). You pay by the hour for the SKU you provision, and you can pause capacity to stop the meter. Bigger SKUs give you more compute and unlock features like Copilot (F64 and above). Storage in OneLake is billed separately, at flat per-GB rates. Most SMBs start somewhere between F2 and F16, then size up once they know their real usage. Our Fabric pricing guide and capacity sizing guide go deeper.

Microsoft Fabric vs Power BI

Power BI is the reporting layer - reports, dashboards, semantic models. Fabric is everything underneath plus Power BI. If all you need is dashboards on top of a couple of well-structured sources, Power BI on its own is fine. Fabric becomes worthwhile when you have multiple source systems, want a governed lakehouse, or need engineering and reporting under the same roof. See Microsoft Fabric vs Power BI.

Microsoft Fabric vs Synapse and a traditional warehouse

Fabric absorbs most of what Azure Synapse did (warehousing, Spark, real-time), rolled into a SaaS product with simpler billing. You don't provision separate services, and everything reads from OneLake. Compared to a traditional cloud data warehouse (Snowflake, Redshift, standalone Synapse), Fabric trades some of the fine-grained control for one billing model, one identity model and Power BI in the box. Our Fabric vs data warehouse piece covers when each wins.

Copilot and AI in Microsoft Fabric

From F64 upwards, Copilot is available across Power BI, Data Factory, the warehouse and data science - so users can generate DAX, write SQL, build pipelines or draft notebooks in natural language. Because Fabric already has a governed semantic layer over your data, Copilot answers are grounded in the model you've published rather than raw tables. More detail in Copilot in Microsoft Fabric.

Security and governance

Fabric inherits Microsoft Entra identity, sensitivity labels and Purview governance, so the controls you already run for Microsoft 365 apply to your data platform too. Workspaces, roles, row-level security and audit logs are built in. We cover the practical setup in Fabric security and governance.

Is Microsoft Fabric right for your business?

Fabric is worth it when you have multiple source systems, want one governed platform for analytics and AI, or you're already paying for Power BI Premium capacity and not getting full value. It's overkill when you have one or two sources and Power BI Pro alone would do the job. If you're not sure, a short discovery is usually enough to tell.

Where to go next

If you want an even simpler introduction, read our plain-English guide to Microsoft Fabric. When you're ready to plan a rollout, our UK Microsoft Fabric consultancy designs the lakehouse, semantic layer and reporting on top - and our Microsoft Fabric service page covers the ongoing side.

Thinking about Microsoft Fabric? Get in touch.