Understanding Microsoft Fabric pricing

Microsoft Fabric pricing can feel confusing at first. Even with Microsoft’s documentation, it often takes a bit of digging to fully understand how it works and where the value comes from. In this blog, we’ll break it down in a simple way — so you can clearly understand what you’re paying for and whether it makes sense for your business.

Jackie Tejwani

Jackie Tejwani

Director - Business Intelligence

Understanding Microsoft Fabric Pricing

Microsoft Fabric pricing can feel confusing at first. Even with Microsoft’s documentation, it often takes a bit of digging to fully understand how it works and where the value comes from.

In this blog, we’ll break it down in a simple way — so you can clearly understand what you’re paying for and whether it makes sense for your business.

Capacity-Based Pricing (Not Per User)

Unlike traditional Power BI licensing, Microsoft Fabric is primarily priced based on capacity, not just users.

This means:

  • You pay for a dedicated pool of compute power (capacity)

  • You can scale up or down depending on your workload

  • It’s designed to support end-to-end data platforms, not just reporting

👉 This is a big shift from Power BI — and where a lot of the confusion starts.


Do You Still Need Power BI Licenses?

Yes — in some cases.

For smaller Fabric capacities (F32 and below):

  • Users still need a Power BI Pro license to create and share reports

So even though Fabric introduces capacity pricing, user licensing hasn’t completely gone away at lower tiers.


Storage Costs (OneLake)

On top of capacity, you also pay for storage.

Fabric uses OneLake as its central data storage layer, priced at approximately:

👉 $0.0435 per GB per month

This is relatively low, but it’s important to include it when estimating total cost — especially as data grows.


Example: What Does Fabric Actually Cost?

Let’s break down a simple real-world scenario.

Assumptions:

  • Fabric F2 capacity

  • 10 Power BI users

  • 2GB of stored data

Monthly Cost:

  • Fabric Capacity (F2): ~$532 NZD

  • Power BI Pro Licenses: $226 NZD
    ($22.60 per user × 10 users)

  • Storage (2GB): ~$0.09 NZD

👉 Total Monthly Cost:

≈ $758 NZD per month


Microsoft Fabric capacity pricing (F SKU)

F SKU

Capacity unit (CU)

Pay-as-you-go*
NZD

F2

2

$532.523/month

F4

4

$1,065.046/month

F8

8

$2,130.092/month

F16

16

$4,260.183/month

F32

32

$8,520.365/month

No need for separate Power BI licenses

F64

64

$17,040.730/month

F128

128

$34,081.459/month

F256

257

$68,162.918/month

F512

512

$136,325.836/month

F1024

1024

$272,651.672/month

F2048

2048

$545,303.344/month

*Pay only for active Spark compute: Autoscale Billing for Spark uses standard regional PAYG CU-hour rates. Reserved capacity discounts do not apply. Admins can set maximum CU limits and request quota increases in the Azure portal.


Final Thoughts

Microsoft Fabric goes beyond traditional self-service analytics tools like Power BI. It is designed as an end-to-end data platform, bringing together data ingestion, transformation, storage, visualisation, and AI capabilities into a single unified solution.

A simple way to think about it is this: if you’re currently running Power BI dashboards, consider all the additional services you’re likely using alongside it — such as data warehousing, storage, and compute within Azure.

With Microsoft Fabric, these components are no longer separate. Microsoft has brought them together into one platform, simplifying architecture, reducing complexity, and enabling teams to manage their entire data ecosystem in one place.


Is Fabric Worth It?

For smaller teams with only one datasource, the cost may feel higher than expected — especially when combining capacity and user licenses.

But for organisations looking to:

  • Centralise their data platform

  • Reduce multiple tools

  • Enable AI and advanced analytics

👉 Fabric can be a very cost-effective solution long-term.

Microsoft Fabric Pricing:

https://azure.microsoft.com/en-us/pricing/details/microsoft-fabric/


SHARE