Skip to main content

How to Purchase an F2 Fabric Capacity (for Copilot)

To enable Copilot for Power BI, your PPU workspace must be converted to run on a dedicated Fabric capacity, or you need to have a different workspace with Fabric capacity assigned. You must purchase a paid Fabric capacity (trial will not work). The F2 SKU is the minimum and most cost-effective entry point for enable Copilot. This purchase is not made in the Power BI or Microsoft 365 admin center; it is an Azure resource you must provision in the Azure Portal.

Link to Microsoft Learn Help article

Here are the prerequisites, step-by-step instructions, and key discussion points. Please read the entire article before proceeding.

1. Required Permissions

Before you begin, the person performing the purchase must have the correct permissions in Azure. Power BI or Fabric Administrator rights are not sufficient for this task.

  • Azure Role: You must have the Owner or Contributor role on an Azure Subscription.
  • Who is this person? This is typically an IT Administrator, Cloud Administrator, or Azure Subscription Owner who manages the company's cloud billing.

2. Step-by-Step Purchase Instructions

  1. Log in to the Azure Portal (portal.azure.com) with the high-privilege account.
  2. In the top search bar, type "Microsoft Fabric" and select it from the services list.
  3. Click the "+ Create" button. This will open the "Create Fabric Capacity" configuration page.
  4. Fill in the following details on the Basics tab:
    • Subscription: Choose the Azure subscription that will be billed for the capacity.
    • Resource Group: Select an existing one or (recommended) click "Create new" (e.g., rg-bi-prod). See discussion below.
    • Capacity name: Give your capacity a clear, unique name (e.g., Fabric-BI-F2-Prod).
    • Region: Select the Azure region where the capacity will live. This is the most critical decision. See discussion below.
    • Size (SKU): Select F2. This is the minimum SKU for Copilot.
    • Capacity administrator: Enter the email address of the person who will manage this capacity from the Fabric/Power BI Admin portal (this should be your Fabric/Power BI Admin).
  5. Click "Review + create".
  6. Azure will validate your settings. Once validated, click "Create" to purchase and deploy the capacity. This typically takes 1-2 minutes.

3. Key Decisions & Discussion Points

The two most important fields you will fill out are Region and Resource Group.

Discussion 1: Region (Your Most Important Choice)

You should always place your Fabric capacity in the same region as your underlying data sources.

  • What it is: The "Region" is the physical Azure data center (e.g., East US 2, West Europe) where your F2 capacity's compute power will run.
  • The Rule: The semantic model (dataset) will live inside the Fabric capacity. Therefore, your goal is to co-locate your Fabric capacity with your primary data sources (e.g., your SQL Database, Azure Synapse Analytics, or Lakehouse).

Here is why this is critical:

  1. Cost (Data Egress):
    • Good Scenario (Co-located): Your F2 capacity is in East US and your SQL Database is also in East US. When your semantic model refreshes, the data transfer is entirely within the same data center. This is fast and, most importantly, free.
    • Bad Scenario (Separated): Your F2 capacity is in West US but your SQL Database is in East US. Every time your semantic model refreshes, it must pull data across the country. Azure charges data egress fees (bandwidth costs) for all data that leaves a region. This can make your refreshes unexpectedly expensive.
  2. Performance (Latency):
    • Data transfer is significantly faster when the resources are in the same physical location. Co-locating your capacity and data source will result in much quicker data refreshes.
  3. Copilot Availability:
    • You must select a region where Fabric and Copilot are available. While this includes most major regions, it's a final check. If your tenant is outside the US/France, your tenant admin must also have enabled the cross-geo setting for Copilot.

Discussion 2: Resource Group

  • What it is: A Resource Group (RG) is simply a logical "folder" in Azure to hold related resources. It is not a physical boundary and has no performance impact.
  • Why it matters:
    • Billing: It allows you to group all related costs. By placing your F2 capacity and your SQL database in the same RG (e.g., rg-bi-prod), you can easily see one consolidated bill for your entire BI solution.
    • Permissions: RGs are a great way to manage permissions. You can give your BI Manager Contributor rights only to this rg-bi-prod group. This allows them to pause, scale, or manage the F2 capacity without giving them access to the rest of the company's cloud infrastructure.
  • Recommendation: Click "Create new" and name it something descriptive, like rg-fabric-prod or rg-analytics-shared.


    Link to Next Article - How to Enable Copilot for Power BI
Was this article helpful?