top of page

Power BI Pro at $14 vs Premium Per User at $24 vs Fabric F64 capacity

1 hour ago
8 min read
Power BI Pro vs Premium Per User vs Fabric F64 capacity license comparison over a teal city skyline: Data Studios original cover

Power BI Pro is listed by Microsoft at $14.00 per user per month and Power BI Premium Per User (PPU) at $24.00 per user per month, both paid yearly, as shown on the Power BI pricing page on November 2, 2026. The third option, a Microsoft Fabric capacity, has no per-user price at all: it is billed by capacity units per hour, and only at the F64 size or larger does it let people with a free license open Power BI reports. For a US analytics manager planning 2027 renewals, the choice comes down to how many people only read reports, how large the semantic models are, and whether the organization needs Fabric workloads beyond Power BI.


The licensing rules behind those three options are documented on Microsoft Learn, and they explain most of the surprises teams run into. A Pro license is needed to publish and share content. PPU adds Premium features for each licensed person, but content in a PPU workspace can be opened only by other PPU users. A Fabric capacity of F64 or above removes the per-viewer license for report consumers, while every author still needs Pro or PPU. Microsoft's own guidance says PPU is more cost effective when Premium features are needed for fewer than 250 users.


··········


POWER BI LICENSE PRICES AND LIMITS AS LISTED BY MICROSOFT.


The Power BI pricing page lists four ways to pay. The free account lets a person build reports in Power BI Desktop with no credit card, but sharing requires an upgrade. Power BI Pro adds publishing and sharing, and Microsoft lists it as included in Microsoft 365 E5 and Office 365 E5. Premium Per User includes all Pro features and raises the limits on model size, refresh frequency and storage, and adds advanced AI, advanced dataflows, advanced datamarts and XMLA endpoint read/write. Fabric capacity is sold either as pay-as-you-go, which can be scaled or paused, or as an annual reservation that Microsoft says saves 40.5% over pay-as-you-go.


........


License

Listed price

Model memory

Refreshes per day

Storage

Free

$0

Not listed

Not applicable

Not applicable

Power BI Pro

$14.00 user/month, paid yearly

1 GB

8

10 GB per license

Premium Per User

$24.00 user/month, paid yearly

100 GB

48

100 TB

Fabric capacity

Variable, by region

Depends on SKU

48 per model

Billed via OneLake


........


Two footnotes on the pricing page change the arithmetic for many companies. Pro, Microsoft 365 E5 and Office 365 E5 users can step up to PPU with an add-on listed at $14.00 per user per month, paid yearly. Microsoft also states that the prices shown are for marketing purposes and may vary by currency, country and region, with the actual price shown at checkout, and that academic, government and nonprofit pricing exists. The PPU FAQ gives capacity models a size limit of 25 to 400 GB depending on SKU, against 100 GB for PPU.


··········


WHO CAN OPEN WHAT: POWER BI PRO, PPU AND FREE USERS ON FABRIC.


A paid author license never covers the audience. Microsoft's Fabric licensing page states that every organization needs at least one user with a Pro or PPU license if it intends to use Power BI within Fabric, and the pricing page states that a Pro license is required to publish to Fabric. Viewing rights then depend on where the content lives.


........


Where the report lives

Author needs

Viewer needs

Pro workspace

Pro or PPU

Pro or PPU

PPU workspace

PPU

PPU

Fabric F2 to F32

Pro or PPU

Pro, PPU or trial

Fabric F64 or larger

Pro or PPU

Free license and viewer role


........


The PPU rule is the one that most often breaks a budget. The PPU FAQ states that a PPU license is required to access content in a PPU workspace or app, including access through the XMLA endpoint, Analyze in Excel and composite models. Pro users can receive an email subscription attachment from a PPU workspace but cannot view the content in the portal, and content published to web from a PPU workspace is not accessible to unlicensed users. A team that moves a shared sales report into a PPU workspace therefore has to license every reader with PPU.


The F64 threshold is equally strict in the other direction. The Fabric licensing page says that to view Power BI content with a free license, the capacity must be F64 or larger and the user needs a viewer role on the workspace. On F2 through F32, each viewer still needs Pro, PPU or an individual trial. A smaller F SKU can be a sensible way to run lakehouses, warehouses or notebooks, and the free license does let users create and share non-Power BI Fabric items there, but it does not remove report viewing licenses.


PPU also has documented gaps compared with a capacity. The PPU FAQ lists no Multi-Geo support, no unlimited distribution, no Power BI reports on-premises, no support for the Premium metrics app, and an export API limited to paginated reports at one call every five minutes per user. PPU does not provision a Fabric capacity either, so PPU users still need an F or trial capacity to create non-Power BI Fabric items.


··········


FABRIC CAPACITY PRICING FOR F64 AND SMALLER SKUS.


Fabric capacities are sized in capacity units (CUs). The licensing page maps F64 to 64 CUs and lists it next to the old P1 for comparison, with F128 next to P2 and F256 next to P3, while warning that the comparison is not a statement of functional or licensing equivalence. Microsoft's Azure retail price list, read on November 2, 2026, shows Fabric capacity usage meters at USD 0.18 per CU per hour in the US East region, and a one-year Fabric capacity reservation at USD 938 per CU for the term in the same region. That works out to 40.5% below running pay-as-you-go for a full year, the same saving the Power BI pricing page quotes.


........


SKU

CUs

Pay-as-you-go per hour

Pay-as-you-go, 730 hours

1-year reservation per month

F2

2

$0.36

$262.80

$156.33

F32

32

$5.76

$4,204.80

$2,501.33

F64

64

$11.52

$8,409.60

$5,002.67


........


These figures are US East list prices multiplied out in Python; other regions differ, and the Azure pricing page shows the rate for each region. Pay-as-you-go capacity can be paused when nobody uses it, which lowers the bill for development capacities, but a capacity that serves reports to free users has to run whenever those users open them. OneLake storage is billed separately from compute.


For organizations still on Power BI Premium P SKUs, Microsoft's migration FAQ says it no longer sells new P SKUs, that each P SKU subscription retires at the end of its current agreement term, and that migration to an F SKU is manual. After a subscription ends there is a 30-day grace period, interactive operations are throttled from day 31, and all operations are rejected from day 91 until workspaces move to an F capacity. The FAQ maps P1 to F64 and recommends buying the F SKU first, reassigning and validating workspaces, and only then cancelling the P SKU. PPU is not affected by the P SKU retirement.


··········


POWER BI COST EXAMPLE FOR A 200-PERSON ANALYTICS AUDIENCE.


The following model is illustrative. It assumes a company with 25 report authors and 175 people who only read reports, US East capacity prices, no Microsoft 365 E5 licenses, and monthly equivalents of the annual prices. Every figure was computed in Python.


........


Option

Monthly cost

Annual cost

Pro for all 200 users

$2,800.00

$33,600.00

PPU for all 200 users

$4,800.00

$57,600.00

F32 reserved + Pro for all 200

$5,301.33

$63,616.00

F64 reserved + Pro for 25 authors

$5,352.67

$64,232.00

F64 pay-as-you-go 24/7 + Pro for 25 authors

$8,759.60

$105,115.20


........


pro, ppu = 14.00, 24.00            # USD per user per month, paid yearly
cu_hour, cu_year_res = 0.18, 938.0  # USD, US East, per CU
authors, viewers = 25, 175
users = authors + viewers

f64_reserved = 64 * cu_year_res / 12          # 5,002.67 per month
f64_payg = 64 * cu_hour * 730                 # 8,409.60 per month

print(users * pro)                             # 2,800.00
print(users * ppu)                             # 4,800.00
print(f64_reserved + authors * pro)            # 5,352.67
print((f64_reserved + authors * pro) / ppu)    # 223.03 users
print(f64_reserved / pro)                      # 357.33 viewers

For this audience, Pro for everyone is the cheapest choice as long as the models fit in 1 GB and eight scheduled refreshes a day are enough. If the team needs Premium features such as 100 GB models or 48 refreshes, PPU for all 200 users costs $4,800.00 a month, which is still below an F64 reservation plus Pro for the authors at $5,352.67. With 25 authors on Pro, the F64 reservation becomes cheaper than PPU for everyone at 224 users. Against Pro for everyone, the same F64 reservation pays off only above 357 viewers, because $5,002.67 a month covers about 357 Pro licenses.


The F32 line costs about half of F64 but leaves every viewer on a paid license, so the total ends up close to the F64 option without the free-viewer benefit.


Companies on Microsoft 365 E5 or Office 365 E5 should redo the model with Pro already paid for. In that case the incremental choice is the $14.00 PPU step-up add-on per user against an F64 capacity, and with the figures above the reservation becomes cheaper above roughly 357 Premium users. The Microsoft 365 licensing picture is covered in more detail in our breakdown of Microsoft 365 business and enterprise Copilot plans.


··········


COMMON POWER BI LICENSING MISTAKES THE DOCUMENTATION RULES OUT.


Three assumptions come up repeatedly and each one is contradicted by Microsoft's documentation. Pro users cannot read reports kept in a PPU workspace. An F2 to F32 capacity does not give free users access to reports. A capacity of any size does not remove author licenses, because publishing to Fabric requires Pro and editing Power BI items outside My workspace requires Pro or PPU. A fourth concerns renewals: outside the sovereign clouds, P1 cannot be renewed through an expiring Enterprise Agreement or Microsoft Cloud Agreement, and the supported path is an F SKU of the same CU size. Teams evaluating AI features on the same capacity can compare what Microsoft offers across Excel, Power BI and Microsoft 365 in our overview of Microsoft Copilot tools for data analysis, and how Fabric-based AI items are billed in our explainer on the Fabric AI knowledge workspace.


··········


POWER BI PRO, PPU OR FABRIC F64: WHAT TO CHECK BEFORE RENEWAL.


Microsoft confirms the per-user prices of $14.00 for Pro and $24.00 for PPU on its pricing page, the F64 threshold for free viewers on the Fabric licensing page, the PPU-only access rule in the PPU FAQ, and the retirement of P SKUs at the end of each agreement term in the migration FAQ. The capacity price depends on region and on whether the capacity is reserved, and the dollar figures in the cost model apply to US East list prices on November 2, 2026. Negotiated enterprise agreements, academic, government and nonprofit pricing, and local currencies can all change the result.


Before renewing, count three groups in the Microsoft 365 admin center and the Power BI admin portal: people who create or manage content, people who only read reports, and people whose licenses are already covered by Microsoft 365 E5 or Office 365 E5. Then check the largest semantic model and its refresh schedule against the 1 GB and eight-refresh limits of Pro, and use the Microsoft Fabric Capacity Metrics app on any existing capacity to see whether F64 is the right size or only the minimum for free viewers.


In the illustrative model, per-user licensing stays cheaper than an F64 reservation up to 223 users on PPU and up to 357 viewers on Pro at US East list prices. Above roughly 360 readers, or where the organization already runs Fabric workloads on a capacity of that size, F64 with Pro licenses for authors only is the lower-cost structure. Microsoft does not publish a single crossover point beyond its fewer-than-250-users guidance for PPU, so the decision rests on the organization's own counts and its regional price.


··········


FOLLOW US FOR MORE.


·····


DATA STUDIOS


·····


datastudios.org


bottom of page