Power BI Pro vs Premium in 2026: Which License Fits Your Viewers, Creators and Budget
- 28 min read
- on November 11, 2026
- by Gaurav Verma
TL;DR
In Power BI Pro vs Premium, choose Pro when most people build reports and Premium capacity when hundreds of people only read them. Pro costs $14 per user per month and licenses one person to publish and share. Premium is now sold as Microsoft Fabric capacity, and only F64 or larger lets users with a free license view reports. With a one-year F64 reservation, capacity beats Pro for everyone once about 358 people only read reports, or about 601 on pay-as-you-go. Premium Per User at $24 suits a small analyst team that needs 100 GB models, because every viewer of its content also needs PPU. Creators need Pro under every option, and Microsoft 365 E5 already includes Pro, so count E5 seats before you buy.
Key Takeaways
- Pro and Premium Per User license people, while Premium capacity, now sold as Fabric F SKUs, licenses compute for workspaces.
- Only F64 or larger capacity lets users with a free license view Power BI content; smaller F SKUs do not.
- Count readers, not total users: an F64 reservation beats Pro for everyone at about 358 readers, and pay-as-you-go at about 601.
- Microsoft 365 E5 and Office 365 E5 already include Pro, so those seats cost nothing extra.
- Premium Per User looks cheap until you count viewers, because everyone who opens PPU content needs PPU too.
- Microsoft no longer sells P SKUs; each one retires at the end of its agreement term, and the move to an F SKU is manual.

A Renewal Quote That Licensed Everyone to Read One Dashboard
Picture a finance director opening a Power BI renewal quote. It prices 1,280 Pro seats at $14 each, or $17,920 a month. Only 80 of those people build reports. The other 1,200 open the same few dashboards to check a number before a meeting. That quote is the Power BI Pro vs Premium question in its plainest form.
The scenario is hypothetical, but the pattern is easy to fall into wherever reporting spread faster than anyone planned licenses. Every reader got a creator’s license because that was the default. Nobody stopped to ask whether a capacity license would serve the readers for less.
This guide answers that question with Microsoft’s own list prices and rules, checked on November 10, 2026. It explains what each license covers and traces one report from author to viewer. It then works the break-even arithmetic once and gives a direct pick for six common scenarios.
What Pro, Premium Per User and Premium Capacity Actually License
Power BI sells two kinds of license, and mixing them up is where overspending starts. A per-user license attaches to a named person. A capacity license attaches to compute, which an admin then assigns to workspaces.
The two models count different things, and that difference drives the whole cost gap. Per-user pricing grows with headcount, so every new reader adds $14 or $24 a month. Capacity pricing is flat for a given size, so a thousand extra readers on F64 add nothing to the license bill.
The trade-off is a fixed floor. You pay for capacity whether ten people use it or ten thousand, so idle capacity still costs money. The table below sets out what each option licenses and who has to hold it.
| License | What it licenses | Who must hold it | List price (checked November 10, 2026) |
|---|---|---|---|
| Power BI Pro | A person, to publish, share and view content in shared workspaces | Every creator, and every viewer when content sits below F64 | $14.00 per user per month, paid yearly |
| Premium Per User (PPU) | A person, with the Premium feature set | Every creator and every viewer of content in a PPU workspace | $24.00 per user per month, paid yearly |
| Fabric capacity, F64 or larger | Compute for the workspaces assigned to it | Creators still hold Pro or PPU; viewers can use a free license | F64 in East US: $8,409.60 a month pay-as-you-go, $5,002.67 a month on a one-year reservation |
| Fabric capacity, F2 to F32 | Compute for Fabric items and Copilot | Every Power BI viewer still needs Pro or PPU | Scales with capacity units; see the pricing guide linked in the cost section |
Power BI Pro: The License to Publish and Share
Pro is the baseline license for anyone who works with reports in shared workspaces. A Pro user can publish from Power BI Desktop, share content and collaborate with other licensed users. Microsoft’s Power BI pricing page lists Pro at $14.00 per user per month, paid yearly.
Two details on that same page also change the math for many enterprises. Microsoft lists Pro as included in Microsoft 365 E5 and Office 365 E5, so users on those plans already have it. Microsoft also requires a Pro license to publish Power BI content to P or F capacity, which means capacity never removes the creator license.
Pro also has firm limits. Its semantic models cap at 1 GB, and scheduled refresh tops out at 8 times a day. Those limits, more than price, are what usually push a team toward Premium.
Premium Per User: Premium Features for a Named Group
Premium Per User gives one person the Premium feature set without anyone buying capacity. It lists at $24.00 per user per month, paid yearly, so it costs $10 more than Pro. Microsoft’s PPU FAQ sets its model size limit at 100 GB and its refresh limit at 48 a day.
The catch sits in the sharing rules. Content in a PPU workspace can only be opened by PPU users, so Pro users and free users can’t reach it. PPU therefore works well for a contained team and poorly for a wide audience.
Microsoft also lists a step-up add-on for users who already hold Pro, Microsoft 365 E5 or Office 365 E5. As of November 2026 it is priced at $14.00 per user per month. For E5 holders, who already have Pro, that means PPU features for $14 instead of $24. For a standalone Pro seat the add-on brings the total to $28, so swapping Pro for a $24 PPU seat at renewal usually costs less.
Premium Capacity Is Now Fabric F SKUs
Premium capacity used to mean a P SKU, such as P1. Microsoft now sells that capacity instead as Microsoft Fabric F SKUs, measured in capacity units, or CU. On Microsoft’s Fabric licensing page, F64 equals P1: 64 CU, or 8 v-cores.
In practice, F64 is the threshold that matters for Power BI. At F64 or larger, users with a free Fabric license and a viewer role can open Power BI content. Below F64, every viewer still needs Pro or PPU, even though the workspace runs on capacity.
Capacity also runs more than reports. The same capacity units power lakehouses, warehouses, pipelines and other Fabric items, and our Microsoft Fabric vs Power BI guide covers that wider platform. For licensing, the point is narrower: a small F SKU such as F2 adds Fabric workloads without changing who must pay to view a report.
Existing P SKU customers also face a timing decision. According to Microsoft’s Premium migration FAQ, no new P SKUs are sold, and each P subscription retires at the end of its agreement term. Customers whose Enterprise Agreement or Microsoft Cloud Agreement is expiring can’t add or renew P SKU capacity. Once a subscription ends, content gets a 30-day grace period before access is throttled. The move to an F SKU is manual, and the migration section below walks through it step by step.
Power BI Pro vs Premium Compared Side by Side
The table puts the three options side by side on the points that change a buying decision. Every row comes from Microsoft’s pricing page and Learn documentation, checked on November 10, 2026.
| Factor | Pro | Premium Per User | Fabric capacity (F64 or larger) |
|---|---|---|---|
| List price | $14 per user per month | $24 per user per month | F64: $8,409.60 a month pay-as-you-go, $5,002.67 a month reserved (East US) |
| Who can view shared content | Pro or PPU users | PPU users only | Free users with a viewer role |
| License to create and publish | Pro | PPU | Pro or PPU, still required |
| Maximum semantic model size | 1 GB | 100 GB | 25 to 400 GB, depending on SKU |
| Scheduled refreshes per model | 8 a day | 48 a day | 48 a day |
| Paginated reports | Included | Yes, included | Free users can view them on F64 or larger |
| Copilot in Power BI | Needs paid capacity | Needs paid capacity | Paid F2 or larger, or a P SKU, not trial |
| Distribution to users without a paid license | No | No | Yes |
| Power BI Report Server rights | Not listed | Not listed | Included with P capacity, and with F capacity bought on a reservation |
| Fabric items such as lakehouses and pipelines | No | No | Yes, on any F SKU |
Two rows deserve a second look. The paginated reports row corrects a common belief that pixel-perfect reports need Premium. The Report Server row matters for teams that keep reports on-premises. Microsoft’s Premium overview states that P capacity includes a license to install Report Server. The Premium migration FAQ says that on F SKUs the right comes with a Fabric capacity reservation or SQL Server Enterprise Edition with Software Assurance. Confirm you hold one of the two before you cancel a P SKU.
Limits That Force an Upgrade
Price is rarely the first reason a team leaves Pro. A limit usually is, because a limit blocks work while a price only strains a budget. The most common is model size. A Pro model caps at 1 GB, and a sales model holding several years of order lines can outgrow it. Sound Power BI data modeling practices, such as dropping unused columns, can keep a model under that cap for longer.
Refresh cadence then becomes the second trigger. Pro allows 8 scheduled refreshes a day, which suits a morning finance pack. An operations dashboard that must update every half hour through a working day needs PPU or capacity, and both allow 48. Setting up incremental refresh can also shorten each run, which helps a tighter schedule fit on any license.
The third trigger is the Premium feature set itself. For example, Microsoft’s PPU FAQ lists deployment pipelines and XMLA endpoint connectivity among the features PPU and capacity share. Teams that manage separate development, test and production releases tend to want these first. Check each feature against Microsoft’s license tables before you assume Pro lacks it.
Paginated reports, however, are not a reason to upgrade. The paginated reports FAQ confirms they are included in Pro, with no capacity needed to publish them. Teams retiring older pixel-perfect tools, as our guide to Crystal Reports alternatives describes, can often start on Pro. The same holds for many reports rebuilt during a Cognos to Power BI migration.
Plenty of report design work never touches these limits. Techniques such as field parameters in Power BI run on Pro and let one report serve many views, which can delay an upgrade by years. Most Power BI dashboard development works the same way, needing Pro to publish and share rather than capacity.
Copilot and AI: Which Capacity You Need
Copilot changes the license question because it is tied to capacity rather than to the user license. Per Microsoft’s Copilot in Fabric overview, Copilot needs a paid F2 or larger capacity, or a P SKU, in a supported region. It does not run on trial capacity.
That makes Copilot a separate decision from viewer licensing, so it needs its own budget line. A team of 40 Pro users can add a small F SKU for Copilot without changing how its readers are licensed. Free viewing still waits for F64, so a small SKU bought for Copilot leaves the per-reader math where it was.
Region support also applies, so check that your capacity sits in a supported region before you budget for it. For Copilot across lakehouses, notebooks and other Fabric items, see our Microsoft Fabric Copilot guide. Our Power BI Copilot guide covers the setup and the tenant switches in more detail.
Kanerika Service
Power BI Consulting and Implementation
Kanerika designs Power BI semantic models, workspaces and licensing setups that hold up at enterprise scale, from Pro and PPU teams to Fabric capacity rollouts.
Explore Power BI ServicesWho Needs a License: Trace One Report From Author to Viewer
The fastest way to learn the licensing rules is to follow one report as its audience grows. Each move to a new kind of workspace also changes who must pay to open it. The path below starts with a single analyst and ends with customers outside the company.
Say an analyst builds a margin report in Power BI Desktop. While it sits in My workspace, only the author sees it. Once the analyst shares it, the licensing rules apply to everyone involved. They tighten or relax depending on where the report lives.
- My workspace. Personal content for the author. Sharing it with others requires Pro licenses.
- Pro workspace. A shared workspace without capacity. The author and every viewer need Pro or PPU.
- PPU workspace. Premium features for the group. Every member and every viewer needs PPU.
- F64 or larger workspace. Capacity licenses the compute. Creators need Pro or PPU, and viewers with a free license and a viewer role can read.
- External audience. Customers or partners outside your tenant. Use embedding for apps, or check guest rules, both covered below.
The license burden moves from readers to capacity only at step 4. Steps 1 to 3 all price readers per head, which is why PPU looks cheap for a team and expensive for a company.

The same logic explains why a report can feel free in one workspace and costly in another. Nothing about the report itself changes between steps, though. Only the container changes, and the container decides the bill.
External Guests and Embedded Reports
Reports for people outside your organization follow two separate routes. Guest access lets B2B users from another tenant sign in to your Power BI service. Embedding puts reports inside your own application, where those users never see the Power BI service at all.
Guest licensing is easy to get wrong, so verify each case before you plan around it. Guests follow your tenant’s capacity and license rules, and some may bring a license from their home tenant. Confirm the exact case against Microsoft’s B2B guidance before you plan a large partner audience around it.
Embedding, by comparison, is clearer. With app-owns-data embedding, Microsoft’s licensing page states that end users don’t need a Power BI license. The content can then run on any F, A, EM or P SKU. Our Power BI Embedded guide explains the SKUs and the setup. For customer-facing reporting, this route takes per-user math off the table entirely.
The Cost Math: One Break-Even, Shown With the Arithmetic
The cost side of Power BI Pro vs Premium comes down to a single question: how many people only read reports? Creators pay the same either way, since they need Pro under every option. So the decision turns on whether capacity costs less than giving every reader a Pro seat.
This section works that number out once. Later sections refer back to it instead of restating it, so the figures stay consistent.
The Formula
Start with the monthly price of F64, then divide it by the Pro price. The result is the number of readers at which capacity and Pro cost the same.
break-even readers = F64 monthly cost / Pro price per user
F64 pay-as-you-go (East US)
64 CU x $0.18 per CU-hour x 730 hours = $8,409.60 a month
$8,409.60 / $14 = 600.7 -> 601 readers
F64 one-year reservation (East US)
64 CU x $938 per CU per year / 12 = $5,002.67 a month
$5,002.67 / $14 = 357.3 -> 358 readersThe inputs come from the Azure retail price list for Microsoft Fabric for East US, checked November 10, 2026. Pay-as-you-go runs $0.18 per CU-hour, and a one-year reservation costs $938 per CU per year. That puts the reservation about 40% below pay-as-you-go, and prices differ by region, so swap in your own rates.
Read the result plainly, then compare it with your own reader count. Below about 358 readers, Pro for everyone is cheaper than a reserved F64. Above about 601, capacity wins even without a reservation.
Between the two, the answer depends on whether you can commit to a year of capacity.

Worked Example: 80 Creators and 1,200 Readers
Return to the renewal quote from the opening, where 80 people build reports and 1,200 only read them. The table then prices four ways to license all 1,280 people.
| Option | Monthly cost | How it adds up |
|---|---|---|
| All Pro | $17,920.00 | 1,280 x $14 |
| All Premium Per User | $30,720.00 | 1,280 x $24 |
| F64 pay-as-you-go plus 80 Pro | $9,529.60 | $8,409.60 + $1,120 |
| F64 one-year reservation plus 80 Pro | $6,122.67 | $5,002.67 + $1,120 |
The pick is the F64 reservation with Pro for the 80 creators. It comes to about $73,472 a year against $215,040 for all Pro. That pick is conditional, though, because a sizing pilot must first show that F64 carries the peak refresh and query load. If it doesn’t, the comparison moves up to a larger SKU.
Premium Per User for everyone is the most expensive line in the table. Against PPU-for-all the crossover comes much sooner, because each reader costs $24 instead of $14. To test it, compare 24 times your total users with 14 times your creators plus the F64 price.
What the Formula Leaves Out
The formula prices licenses only, so four adjustments belong in any real budget. The first is E5. If readers already hold Microsoft 365 E5, their Pro license is paid for. Capacity then has to earn its place through features such as large models or Copilot.
The second is the billing model. Pay-as-you-go capacity can be paused, which helps development or batch workloads that sit idle overnight. According to Microsoft’s pause and resume guidance, pausing stops capacity use and resuming restarts billing. Any outstanding smoothed usage is billed at the moment you pause, so a pause is not a free reset.
The third is sizing, because F64 is only a starting point. F64 is the smallest SKU that frees readers, but a heavy refresh schedule may need more. Our Fabric capacity planning guide covers how to size past F64.
The fourth is everything outside licenses: gateways, data storage and the people who run the platform. Our Microsoft Fabric pricing guide prices every F SKU on both pay-as-you-go and reservation terms, and covers the reservation math in more depth.
Talk to Kanerika
Want This Math Run on Your Own Tenant?
Kanerika counts your real creators and readers, removes E5 overlap and prices Pro, PPU and Fabric capacity side by side, so the renewal decision rests on your own numbers.
Schedule a Demo →Decision Framework: Power BI Pro vs Premium by Scenario
Licensing choices fall into a handful of recognizable shapes, and each shape has a clear answer. The table gives a direct pick for six of them, using the break-even from the cost section rather than restating it.
Read across from the situation that sounds most like yours. Then test the reason column against your own creator and reader counts, since the thresholds move with region and with E5 coverage.
| Scenario | Pick | Why |
|---|---|---|
| Under 100 users, and most of them build reports | Pro, or E5 where it already includes Pro | Readers sit far below the 358 break-even, so capacity costs more than it saves |
| A small analyst team needs 100 GB models or 48 refreshes, with few viewers | Premium Per User for that team only | Premium features without a capacity floor, as long as the audience stays inside the team |
| 80 creators and 1,200 readers | F64 one-year reservation plus Pro for creators, after a sizing pilot | 1,200 readers sit well above 358; about $6,123 a month against $17,920 for all Pro |
| Everyone already holds Microsoft 365 E5 | Stay on Pro through E5 | Pro is already paid for; add capacity only for features such as large models or Copilot |
| Copilot or Fabric data workloads wanted, under 358 readers | Pro for readers plus a small F SKU (F2 or larger) | Copilot needs paid capacity, but free viewing needs F64, so Pro stays cheaper for readers |
| Reports for customers outside the company | Power BI Embedded with app-owns-data, as the capacity side’s route for external audiences | End users need no Power BI license, so per-user math does not apply |
What the Six Picks Have in Common
The four routes in the graphic below are all Pro or a form of Premium. Three signals do most of the sorting among them: reader count, a small team’s large models, and customers outside the company. Premium capacity wins outright in only one shape: a large audience that reads far more than it builds. In the worked example, 1,200 of 1,280 users only read, or about 94%, and that ratio is what lets capacity pay for itself. Elsewhere, Pro carries most users, sometimes with a narrow Premium add-on for one team or one feature.

That split is worth stating plainly because renewal quotes often frame the choice as all or nothing. In reality, few enterprises need everyone on one license. A mix of Pro for creators, PPU for one specialist team and capacity for a reading audience is common and perfectly valid. In the worked example, that mix means 80 Pro seats beside one reserved F64, about $6,123 a month instead of $17,920.
One more pattern also deserves a note. Some companies keep development on Pro or PPU workspaces and publish production reports to F64, where a wide audience reads them. That split keeps experiments off the shared capacity while readers stay on free licenses.
When Premium Per User Is the Right Answer
Premium Per User fits one shape well: a small group that needs Premium features and shares mostly with itself. A data science team building a 50 GB forecasting model, for example, is the classic case. The team needs the 100 GB limit, yet its audience is mainly its own members.
The trouble starts when that team’s work becomes popular. Because every viewer of PPU content needs PPU, each new reader adds $24 a month.
For instance, take 20 analysts and 300 readers on PPU. That is 320 seats, or $7,680 a month. An F64 reservation plus Pro for the 20 analysts costs $5,002.67 plus $280, or $5,282.67 a month.
The crossover for that team therefore arrives early. Divide $5,282.67 by $24 and you get about 220.1, so from about 221 total users, capacity beats PPU for everyone. Microsoft’s own licensing page lands in the same range, describing PPU as more cost effective for fewer than 250 users. That holds only if F64 can carry the workload, which a pilot should confirm before anyone signs a reservation.
A short checklist helps decide whether PPU still fits your team:
- The team needs models above 1 GB or more than 8 refreshes a day.
- Most readers of the content belong to the same team.
- Total PPU users stay well below the crossover with F64.
- Members already hold E5, so the $14 step-up add-on costs less than a $24 PPU seat.
If two or more of these fail, plan for capacity instead. Moving a workspace from PPU to capacity is a workspace setting change rather than a rebuild, so the reports carry over. Microsoft’s PPU FAQ adds that its semantic models need a full refresh after the move.

Moving From Premium P SKUs or Pro to Fabric Capacity
A move to Fabric capacity is mostly administrative work, but it has a deadline when a P SKU is involved. Microsoft retires each P subscription at the end of its agreement term, and nothing moves automatically. The P SKU to F SKU migration overview describes a manual process: buy an F SKU, then reassign workspaces to it.
Teams moving up from Pro, meanwhile, follow nearly the same path without the deadline. Teams arriving from another tool, as in a Tableau to Power BI migration, face the same license choice before the first workspace goes live. The seven steps below cover both cases, in order.
- Check the agreement term. Find when each P SKU subscription ends, or when the Enterprise Agreement renews. That date then sets the deadline for everything else.
- Pilot on a trial or small F SKU. A Fabric trial capacity is usually F64-sized, at 64 CU. Microsoft notes some trials are assigned F4, so confirm the trial’s size in the admin portal before treating it as a load test. Copilot does not run on trial capacity, so test it separately.
- Watch the Capacity Metrics app. The Fabric Capacity Metrics app shows how much capacity refreshes and queries consume. Record the peak hour rather than the daily average.
- Buy the F SKU that fits the peak. Start at F64 if free viewing is the goal, because smaller SKUs don’t free readers.
- Reassign workspaces. Move workspaces to the new capacity in planned waves, starting with the least critical so problems surface early.
- Verify refresh and gateways. Confirm that scheduled refreshes run and gateway connections still work after each wave.
- Decide pay-as-you-go or reservation. Once the load is steady, compare the two prices from the cost section and commit only to the size the metrics support.
Reading the Capacity Metrics App: F64 or F128
Step 3 is where the size gets decided, so it helps to know what the app is showing. Microsoft’s throttling guide explains that Fabric smooths interactive operations over at least five minutes and background operations over 24 hours. Utilization above 100% therefore doesn’t mean throttling on its own, because the capacity can first borrow 10 minutes of future capacity as overage protection.
The Throttling chart on the app’s Compute page shows when smoothed usage crosses each limit. Past 10 minutes of future use, new interactive operations wait 20 seconds. Past 60 minutes they are rejected, and past 24 hours every new request is rejected.
That gives the pilot a plain sizing rule. If F64 stays inside overage protection at the peak hour, keep F64. If interactive delays recur at the peak, price F128 next, since Microsoft’s guide says consistently high throttling calls for a larger SKU or load balancing. In East US, F128 costs $16,819.20 a month pay-as-you-go or $10,005.33 reserved, exactly twice F64.
Before doubling the bill, check whether one heavy model or refresh causes the peak. Microsoft notes that slow performance is often down to an item’s design rather than capacity throttling.
Timing and Communication
Timing these steps against the agreement term matters more than any single step. A pilot left to the final month leaves no room to resize if F64 proves too small. Microsoft also gives content only a 30-day grace period after a P subscription ends. For a P1 whose term ends on June 30, for example, the six-month trigger in the governance table below starts planning on January 1.
Plan the communication as well, since creators feel the change first. Readers who move from Pro to free licenses on F64 notice nothing, but creators need to know which workspaces moved and when.
Talk to Kanerika
Planning a Move From P SKUs to Fabric Capacity?
Kanerika pilots your workload on Fabric capacity, reads the Capacity Metrics app with you and plans workspace waves around your agreement term, so the cutover is not rushed.
Talk to a Power BI Expert →Risks, Governance and Common Licensing Mistakes
Licensing errors rarely come from the price list itself. Instead, they come from habits that nobody revisits after the first purchase. The six below are easy to make, and each one has a simple control.
- Buying Pro for pure readers above the break-even. Past about 358 readers on a reservation, Pro for everyone costs more than capacity.
- Buying capacity below F64 to free viewers. F2 to F32 add Fabric workloads and Copilot, but viewers on them still need Pro or PPU.
- Forgetting that creators still need Pro. Capacity frees readers only, and every publisher keeps a Pro or PPU license.
- Paying twice for E5 users. E5 already includes Pro, so separate Pro seats for E5 users are waste. Our Microsoft licensing optimization guide covers E5 renewal economics.
- Spreading PPU content to a wide audience. Each new reader needs PPU, which turns a cheap team license into the most expensive option.
- Missing the P SKU term date. P SKUs retire at term end and the move is manual, so a late start risks a rushed cutover.
Owners, Controls and Review Triggers
Each mistake also maps to an owner and a control. The table assigns them so the review runs on a schedule instead of waiting for the next renewal.
| Owner | Control | Review trigger |
|---|---|---|
| BI platform lead | Monthly creator and reader counts from activity logs | Reader count crosses 358 (reserved) or 601 (pay-as-you-go) |
| Microsoft 365 admin | Reconcile Pro assignments against E5 holders | Any E5 renewal or reorganization |
| Capacity admin | Capacity Metrics app review of peak load | Sustained peaks near the capacity limit, or a new heavy workload |
| Finance or procurement | Track P SKU and reservation end dates | Six months before any term ends |
| Workspace owners | Confirm each workspace’s license mode: Pro, PPU or capacity | A new audience outside the team |
Kanerika Service
Data Analytics Services
Kanerika builds governed analytics on Microsoft Fabric and Power BI, from data platform design and security roles to the reporting operations that keep licenses and capacity in check.
Explore Data Analytics ServicesGovernance for the content itself, such as sensitivity labels, workspace roles and row-level security, sits beside these controls. Our Microsoft Fabric governance guide covers that side of the platform.
License Admin Mechanics: Assign, Restrict, Expire
The owners in that table also need to know where the license controls live. Per Microsoft’s Power BI licensing guide for admins, buying licenses takes the Billing admin role. License or User admins can then manage them. Assignment happens in the Microsoft 365 admin center under Billing and Licenses, or in Microsoft Entra ID through the Azure portal.
Self-service settings are what most often throw a license count off. Billing admins decide whether users can sign up for, or buy, their own licenses. Turning off self-service purchase blocks sharing, exporting and new collaboration workspaces until an admin assigns a paid license. Microsoft lists a negotiated enterprise rate as one reason to switch it off, so every Pro or PPU seat stays under that agreement.
Expiry follows the Microsoft 365 lifecycle. A license bought through volume licensing gets a 90-day grace period, and a direct purchase gets 30 days. Trials get none, and a Pro trial of up to 25 licenses turns into a charge once it expires.
How Kanerika Runs a Power BI Licensing Review
A licensing review is a short, structured project that turns usage data into a license plan. Kanerika runs it in four stages, and each stage then feeds the next one. The aim is a decision backed by your own numbers rather than by a vendor quote.
Watch on YouTube
End-to-End Microsoft Power BI Implementation and Migration with Kanerika
Kanerika walks through how it implements and migrates Power BI from start to finish. Watch it to see where licensing and capacity decisions fit into a full rollout.
Four Stages, Each Feeding the Next
Stage 1: usage audit. We start with activity logs to see who opens what and who publishes. The output is a creator and reader split per workspace, which is the main input the break-even needs.
Stage 2: cost model. We then apply the formula from the cost section to those counts. Users who already hold E5 come out of the per-user side, and regional prices replace the East US defaults.
Stage 3: pilot. We test the proposed capacity on a trial or small F SKU and track it with the Capacity Metrics app. The pilot answers the one question the formula can’t: does F64 carry your peak load?
Stage 4: rollout and chargeback. We reassign workspaces in waves, settle the reservation decision and set up chargeback, so each department sees its share of the capacity cost.
What the Review Delivers
The output is a short licensing plan you can take into a renewal meeting. It records the creator and reader counts, the chosen mix of Pro, PPU and capacity, and the pilot results. It also lists each term end date and names an owner for every control in the governance table, so the review repeats on a schedule. When the reader count crosses a threshold later, the plan already says what changes.
The review sits inside Kanerika’s wider Power BI consulting and implementation practice, which also covers model design, migration and governed rollout. Kanerika is a Microsoft Solutions Partner for Data and AI, with ISO 27001 certification and a CMMI Level 3 appraisal.
Teams that want help beyond Power BI licensing can draw on our data analytics services, from data platform design to reporting operations. Our consultants can also run the capacity pilot and the workspace moves if your own admins are stretched.
The Same Approach in Client Work
The same delivery approach also shows in client work. For Allen Distribution, a US-based distributor, Kanerika delivered Microsoft Fabric and Power BI analytics with automated refreshes. The published case study reports a 60% improvement in operational efficiency and a 95% cut in report generation time. It also records a 54% gain in data accuracy and 61% growth in data-driven decisions.
Case Study
60% Better Operational Efficiency for Allen Distribution
A US-based distributor moved its reporting onto Microsoft Fabric and Power BI with Kanerika, automating refreshes to cut report generation time by 95% and lifting data accuracy by 54%.
Read the Case Study →Choosing Between Power BI Pro and Premium
The right license follows the size and shape of your audience. Count creators and readers separately, then match the result to one of the cases below.
Small teams, under about 100 users. Stay on Pro, or on E5 when you already pay for it. Capacity costs more than it saves at this size.
Analyst groups with heavy models. Give PPU to the team that needs 100 GB models or 48 refreshes, and keep its audience small. Then watch the crossover with F64 as readers grow.
Audiences of up to a few hundred readers. Pro still wins below about 358 readers on a reservation. Add a small F SKU only if you want Copilot or Fabric workloads.
Large reading audiences. Move to F64 or larger once readers pass about 358, keep Pro for creators, and confirm the size with a pilot first. Still choosing a platform before you license one? Our Power BI vs Tableau licensing comparison covers that decision.
So the answer to Power BI Pro vs Premium turns on who reads. Pro fits creators and small audiences, Premium Per User fits a small team with heavy models, and Premium capacity on F64 fits a budget built around hundreds of readers.
Frequently Asked Questions
What is the difference between Power BI Pro and Power BI Premium?
Pro is a per-user license that lets one person publish, share and view reports in shared workspaces. Premium is now sold as Microsoft Fabric capacity, which licenses compute for workspaces. At F64 or larger, users with a free license can view reports. Authors still need Pro or Premium Per User to publish content to that capacity.
How much does Power BI Premium cost in 2026?
Premium Per User lists at $24 per user per month, paid yearly, as checked in November 2026. Premium capacity is billed as Fabric F SKUs. An F64 in East US costs about $8,410 a month on pay-as-you-go and about $5,003 a month with a one-year reservation. Prices vary by region, so check the Azure price list for yours.
Can you pause Fabric capacity to cut Power BI costs?
Yes. Fabric capacity can be paused, which stops capacity use, and resuming it restarts billing. Any outstanding smoothed usage is billed at the moment you pause. Reports on paused capacity are unavailable until it resumes, so pausing suits development or batch workloads more than reports that people open every working day.
Can external guest users view Power BI reports without a license?
It depends on your setup. Guests follow your tenant’s capacity and license rules, and some guests bring a license from their home tenant. Check Microsoft’s B2B guidance for your exact case before planning a large partner audience. For customer-facing reports, app-owns-data embedding lets end users view content without any Power BI license.
Is Power BI Pro enough for paginated reports?
Yes. Microsoft’s paginated reports FAQ confirms that paginated reports are included in Pro, so you do not need Premium capacity to publish them. Capacity matters only when many readers without a paid license need to open them, since free users can view paginated reports on F64 or larger, or on P1 and above.
How do you size Fabric capacity for Power BI?
Start from measured usage, since user count alone says little about load. Run a pilot on a trial or small capacity and watch the Fabric Capacity Metrics app for peak refresh and query load. Size for the busiest hour, then decide between pay-as-you-go and a one-year reservation once the load is steady.
At what user count does Power BI Premium beat Pro?
Count readers, since creators need Pro under every option. With a one-year F64 reservation, capacity costs less than Pro for everyone once about 358 people only read reports. On pay-as-you-go the crossover is about 601 readers. Run a sizing pilot first, because a heavy workload may need a larger SKU than F64.
What is Power BI Premium Per User (PPU) and when should you use it?
PPU gives named users the Premium feature set, including 100 GB models and 48 scheduled refreshes a day, for $24 per user per month. Every viewer of PPU workspace content also needs PPU. It suits a small analyst team that needs larger models or more refreshes and shares its work mostly within the team.
Did Microsoft retire the Power BI Premium P SKUs?
Microsoft no longer sells new P SKUs. Each existing P SKU subscription retires at the end of its agreement term, and Enterprise Agreement customers can renew until that agreement ends. Moving to an F SKU is a manual process. You buy Fabric capacity, reassign workspaces to it, and then confirm that refreshes and gateways still work.
Do authors still need Power BI Pro on a Premium capacity?
Yes. Microsoft requires a Pro license to publish Power BI content to P or F capacity. Capacity removes the license requirement for viewers only, and only at F64 or larger. Anyone who builds, edits or publishes reports in a shared workspace needs Pro or Premium Per User, whatever capacity the workspace runs on.
Is Microsoft Fabric the same thing as Power BI Premium?
Fabric capacity replaced Premium capacity as the way to buy dedicated compute for Power BI, and F64 matches the old P1 at 64 capacity units. The same capacity also runs lakehouses, warehouses and pipelines. Power BI Pro and Premium Per User remain separate per-user licenses, and many enterprises use them alongside Fabric capacity.
Is Power BI Pro included in Microsoft 365 E5?
Yes. Microsoft lists Power BI Pro as included in Microsoft 365 E5 and Office 365 E5. If most of your users already hold E5, Pro costs nothing extra for them. Capacity then pays off only when you need its features, such as larger models or Copilot, or when you have many readers outside E5.
Which Fabric capacity lets free users view Power BI reports?
F64 or larger. Below F64, every viewer needs Pro or Premium Per User even when the workspace sits on capacity. F64 provides 64 capacity units, which matches the old P1 SKU. Smaller F SKUs, from F2 to F32, help with Fabric workloads and Copilot but do not let free users view Power BI reports.
What license do you need for Copilot in Power BI?
Copilot in Power BI needs a paid Fabric capacity of F2 or larger, or a P SKU, in a supported region. It does not run on trial capacity. A small F SKU bought for Copilot leaves viewer licensing unchanged, since free viewing still starts at F64. Check Microsoft’s Copilot in Fabric overview for the current list of regions.
