Power BI to Fabric Migration

Yes - Microsoft is retiring Power BI Premium per-capacity (P-SKU) licensing in favour of Microsoft Fabric capacities (F-SKUs). Customers on enterprise agreements can continue renewing Premium until their current agreement expires, after which a move to Fabric capacity is mandatory. Rather than treating this as a forced licensing change, we treat it as an opportunity: correctly sized Fabric capacity often costs less than the Premium it replaces, and unlocks Direct Lake performance and AI workloads at the same time.

If you are unsure when your deadline lands, book a scoping call and we will map your renewal timeline and options in 30 minutes.

Most Power BI to Fabric migrations take between 2 and 8 weeks, depending on the size of your estate and how much optimisation you bundle into the move. A simple capacity transition is fast; the time investment goes into validation and performance work:

• Capacity Switch (1-2 weeks): Reassigning healthy workspaces from a Premium P-SKU to a Fabric F-SKU with side-by-side validation.

• Standard Migration (3-5 weeks): Staged waves covering workspaces, gateways, deployment pipelines and security roles, with blockers remediated.

• Migration + Optimisation (6-8 weeks): Adding Direct Lake re-platforming of your largest models and governance configuration across the tenant.

See how we structure full Fabric environments on our Microsoft Fabric consulting page.

A fixed-fee migration assessment and roadmap typically starts between $5,000 and $10,000, with full migrations ranging from $15,000 to $45,000+ depending on the size and complexity of your Power BI estate. The biggest cost factor is usually not the migration itself but your ongoing Fabric capacity - which is why we size your F-SKU against measured usage rather than simply matching your old Premium tier. Many clients find the licensing savings from right-sizing offset a significant portion of the project cost.

For a tailored quote based on your workspace count and model sizes, request a migration assessment.

No - when planned correctly, a Power BI to Fabric migration is invisible to your end users. Reports, apps and dashboards continue working because workspaces are reassigned to the new capacity rather than rebuilt. The risks sit in the details, which is exactly what our staged approach is designed to catch:

• Oversized semantic models that exceed the memory limits of an undersized F-SKU.

• Legacy dataflows and linked entities that carry metadata known to block workspace moves.

• Gateway and network dependencies (such as VNet configurations) that prevent reassignment.

Our pilot-wave validation proves every pattern side-by-side before the full rollout, which is how we deliver zero-downtime migrations. Read about our delivery record on our case studies page.

Power BI Premium (P-SKUs) was a reporting-only capacity purchased through Microsoft 365, while Fabric capacity (F-SKUs) is an Azure-based compute model that runs Power BI plus the entire Fabric platform - data engineering, warehousing, real-time analytics and AI workloads. Key practical differences:

• Flexibility: F-SKUs offer pay-as-you-go and reserved pricing, and start at much smaller sizes than the old P1.

• Workloads: The same capacity that serves your reports can run pipelines, notebooks, lakehouses and Copilot features.

• Management: Capacity lives in Azure, giving you proper cost management, quotas and governance tooling.

In short, you keep everything Power BI does today and gain a platform you can grow into. Our Power BI consulting team helps you take full advantage after the move.

A Power BI Premium P1 is equivalent to a Fabric F64, but many organisations need a different size once their real usage is measured, so the right SKU depends on your workloads and who views your reports. Matching your old P SKU like for like is the most common cause of overspending after migration.

The threshold to understand is F64. On F64 and above, people with a free Power BI licence can view shared content. Below F64, every viewer needs a Power BI Pro or Premium Per User licence. That makes sizing a trade-off between capacity cost and licence cost, not just compute.

We measure your capacity-unit consumption and model both sides before recommending a SKU. For a deeper sizing exercise, see our Microsoft Fabric capacity planning service.

Direct Lake is a Fabric storage mode that lets Power BI read Delta tables in OneLake directly, giving import-like speed without scheduled data refreshes or duplicate copies of large datasets. It suits large models where import refreshes are slow or expensive and DirectQuery is too sluggish.

Direct Lake is optional and does not need to happen on day one. It requires your data to sit in OneLake as Delta tables, usually in a Lakehouse or Warehouse, so we treat it as a second phase once the capacity move is stable. Smaller models that refresh quickly often stay in import mode, which remains a perfectly good choice.

We identify which semantic models would genuinely benefit and re-platform those first. If reports are slow for other reasons, our Power BI performance optimisation service finds the root cause.

Yes, Copilot in Power BI is available on paid Fabric capacities, including the smaller F SKUs, once your administrator enables it in the tenant settings. For many organisations, access to Copilot and other AI features is one of the clearest benefits of moving off Premium.

Copilot is only as good as the semantic model it reads. Clear table and measure names, descriptions, a proper star schema and consistent definitions make the difference between accurate answers and confusing ones. We review your models for Copilot readiness during the migration, along with the tenant settings that control who can use it and where data is processed.

For AI-ready data foundations beyond Power BI, see our data analytics consulting service.