Power Platform Licensing: The Complete Enterprise Guide

The Power Platform Licensing Problem No One Explains Clearly

Power Platform licensing is the most confusing corner of the Microsoft enterprise licensing catalogue — and that confusion is not accidental. The platform spans four primary products (Power Apps, Power Automate, Power BI, Power Pages), each with multiple licensing models (per-user, per-app, pay-as-you-go, capacity), plus Dataverse storage add-ons, premium connector entitlements, and AI Builder credits layered on top. The interactions between these elements are not documented in a single location, and Microsoft's account teams routinely oversell or misconfigure Power Platform licensing at enterprise level.

The result: enterprises we engage with are typically paying 25–40% more for Power Platform than a well-structured deployment would cost, while simultaneously running compliance exposure because key user populations are under-licensed. This guide provides the complete commercial and technical licensing framework for enterprise Power Platform decisions — covering all four products, the critical licensing model decision, audit risk patterns, and EA negotiation strategy.

Power Apps Licensing: Per-App vs Per-User vs Pay-As-You-Go

Power Apps is licensed at three tiers, and the right choice depends almost entirely on how many apps each user needs to access and how frequently they use them.

Per-App Plan

The Power Apps Per-App Plan licenses a single user to access a single app. As of 2026, the ERP list price is approximately $5/user/app/month, but enterprise EA pricing typically falls in the $3.50–$4.50 range with D-level discounts. The Per-App plan covers both canvas and model-driven apps. It includes Dataverse access for that specific app. It does not include rights to any other Power Apps application — each additional app requires an additional Per-App licence.

This model is well-suited to targeted deployments: a field service app for 500 technicians, an expense reporting app for a finance team, or a specific line-of-business workflow. For users who need access to 2–3 apps, the per-app cost multiplies quickly and the Per-User plan becomes more economical.

Per-User Plan

The Power Apps Per-User Plan provides unlimited app access for a single user across all canvas and model-driven apps in the environment. ERP list price is approximately $20/user/month; enterprise EA pricing typically falls in the $14–$18 range. This covers all Power Apps applications the user accesses, plus full Dataverse access.

The break-even between Per-App and Per-User is four apps: at EA-discounted rates, four Per-App licences at $4/each = $16/month vs Per-User at $16/month. In practice, we recommend Per-User for power users and business analysts building or consuming multiple apps, and Per-App for broader functional populations accessing a single operational tool.

Pay-As-You-Go

Pay-As-You-Go for Power Apps is billed through Azure subscriptions at $10/unique user/app/month, metered monthly and charged to the connected Azure subscription. It requires no upfront commitment and suits scenarios where app usage is seasonal, project-based, or genuinely unpredictable. However, for any sustained usage — even as few as 150 unique user-app sessions per month — the EA Per-App plan at $3.50–$4.50 is more cost-effective than PAYG at $10. PAYG should not be the default licensing model for production apps with consistent user populations.

Microsoft 365 included rights

Microsoft 365 E3 and E5 licences include limited Power Apps access — specifically, the ability to use Power Apps in the context of Microsoft 365 data (SharePoint, Teams, OneDrive, Exchange). They do not include Dataverse access and do not cover model-driven apps. The M365 included rights are routinely misunderstood. If your Power Apps deployment reads from SharePoint and runs in Teams, the M365 licence covers it. If it writes to Dataverse or runs as a standalone model-driven app, it does not — and licensing failure in this area is a common audit finding.

Power Automate Licensing: Standard vs Premium Flows

Power Automate licensing hinges on a critical distinction that confuses most enterprise buyers: standard connectors versus premium connectors. This distinction determines whether a flow is covered by a Microsoft 365 licence, a per-user plan, or a per-flow plan.

Standard connectors and M365 coverage

Microsoft 365 licences include Power Automate rights for flows using standard connectors (SharePoint, OneDrive, Teams, Outlook, Excel Online, and approximately 300 other standard connections). If your automation estate consists primarily of M365-integrated flows — routing SharePoint approvals, sending Teams notifications, processing email-triggered workflows — the M365 included rights cover them.

Premium connectors require a Power Automate licence

Premium connectors — which include Dataverse, SQL Server, Salesforce, SAP, ServiceNow, Dynamics 365, HTTP triggers (in production), and several hundred others — require a dedicated Power Automate licence. The failure mode here is architectural: a developer builds a flow using SharePoint (standard, covered by M365), then adds a SQL Server write step (premium, requires licence). The entire flow is now premium-licensed, affecting every user who triggers it.

The Power Automate Per-User Plan (approximately $15/user/month at EA rates) covers unlimited flows for a single user, including premium connectors. The Power Automate Per-Flow Plan (approximately $100/flow/month at EA rates) covers unlimited users running a single flow — suited to enterprise-wide shared flows accessed by large populations where per-user licensing would be cost-prohibitive.

For a detailed breakdown of the Standard vs Premium flow decision, cost models, and governance frameworks, see our dedicated article on Power Automate Licensing: Standard vs Premium Flows.

Power BI Licensing: Pro, Premium Per-User, and Premium Capacity

Power BI licensing has three tiers at enterprise level, and the transition between them creates significant cost leverage at EA renewal. Understanding all three is essential before committing to any Power BI contract.

Power BI Free (included)

The free tier of Power BI allows users to consume published reports and dashboards shared with them, provided the content is hosted in Premium capacity. Free licence users cannot publish reports, create workspaces, or share content — but they can view. This is commercially important: in a Power BI Premium deployment, large view-only populations can consume reports under Free licences rather than Pro licences, which is a direct cost saving of $10/user/month for every reader who does not need to create or share.

Power BI Pro

Power BI Pro (approximately $10/user/month at EA rates) allows users to publish, share, and collaborate on reports. It does not include the compute capacity for large data models, AI features, or paginated reports. Pro is the baseline licence for anyone creating or distributing reports. It is included in Microsoft 365 E5 and can be added to E3 as an add-on. The M365 E5 inclusion makes Pro licensing for E5 populations redundant — a common source of duplicate spend in mixed-estate environments.

Power BI Premium Per-User (PPU)

Power BI Premium Per-User (approximately $20/user/month at EA rates) provides all Pro features plus access to Premium features: large dataset storage, deployment pipelines, paginated reports, AI features, and XMLA endpoint access. PPU is suited to heavy analytical users who need Premium features but whose organisation has not committed to the Premium Capacity (P-SKU) model. The downside of PPU: it does not enable Free licence users to consume content — all consumers need PPU or Pro licences, losing the view-only audience cost saving available under Premium Capacity.

Power BI Premium Capacity (P-SKUs)

Premium Capacity is a shared infrastructure model billed per capacity node (P1 through P5), not per user. P1 costs approximately $4,995/month ($60K/year) at list price; EA pricing typically falls 15–25% below list for committed capacity. Premium Capacity allows unlimited Free licence users to consume reports hosted in that capacity — creating a model where a single P1 capacity node supports an unlimited consumer audience at no per-user cost for view-only access.

The commercial decision between PPU and Premium Capacity is almost entirely a user count calculation: if your view-only audience exceeds approximately 250–300 users, Premium Capacity P1 is more cost-effective than PPU for the entire population (250 users × $20/month PPU = $5,000/month vs P1 at ~$4,000/month at EA pricing). Beyond that threshold, Premium Capacity cost remains flat while PPU cost scales linearly with users.

Power Pages Licensing

Power Pages (formerly Power Apps Portals) licenses authenticated and anonymous external user access to web portals. The licensing model is capacity-based: authenticated user sessions and anonymous user logins are both purchased as monthly capacity, not per-named-user. Authenticated user logins are approximately $200/100 users/month at list price; anonymous logins are approximately $500/500 logins/month. EA pricing typically applies 15–20% D-level discounts.

Power Pages licensing is frequently over-provisioned at initial deployment because usage estimates are difficult to predict before go-live. A common pattern: an enterprise licenses 10,000 authenticated users at launch, discovers 2,000–3,000 are actually active monthly, and carries 7,000+ units of redundant capacity for the remainder of the EA term. Pre-deployment user activity modelling, combined with a 6-month post-launch review provision negotiated into the licence amendment terms, is the advisory position we recommend for any Power Pages deployment exceeding £100K annually.

Dataverse Storage: The Add-On Trap

Dataverse storage is a separate capacity entitlement that is rarely explained upfront and frequently triggers unexpected costs mid-term. Every Power Platform tenant receives a base Dataverse storage entitlement — 10GB database + 20GB file + 2GB log storage by default, plus incremental allocations per licences (approximately 50MB database and 400MB file per Per-User licence). Enterprise deployments routinely exhaust these entitlements and face overage charges or forced add-on purchases.

Dataverse add-on storage is sold as T-Dataverse capacity packs: Database Add-on ($40/GB/month list) and File Add-on ($2/GB/month list). The cost disparity between database and file storage is extreme — enterprises that store large files in Dataverse database storage instead of Dataverse file storage pay 20x more than necessary for the same data. Architecture choices made during app development have direct licensing cost implications that are not visible until the storage bill arrives.

Audit Risk Patterns for Power Platform

Power Platform is increasingly included in Microsoft audit scopes, particularly for organisations that have had significant citizen developer growth since 2020. The four audit patterns we see most frequently are: premium connector flows running under M365 licences without a Power Automate plan; Power Apps model-driven apps accessed by M365-only users; Power BI content shared outside a Premium Capacity to Free licence users; and Power Pages authenticated portals with miscounted user populations.

The premium connector flow finding is by far the most common — and the most expensive. A large flow library built by citizen developers over 2–3 years, where individual flows accumulated premium connector steps organically, can create licence exposure across hundreds of users. Discovery typically occurs during a SAM engagement or renewal documentation review, and remediation costs (catch-up licensing for the claim period) typically run 1.5x–2x the prospective annual licence cost.

Prevention is straightforward: implement a Power Platform Centre of Excellence with connector classification governance, require connector type review before flow deployment, and run quarterly licence reconciliation against your flow inventory. The Microsoft Power Platform Centre of Excellence Starter Kit (a free Microsoft resource) provides the governance framework — but the commercial discipline to apply it requires organisational commitment that citizen developer cultures often resist without top-down mandate.

Power Platform EA Negotiation Strategy

Power Platform is typically negotiated as part of a broader Microsoft EA rather than as a standalone agreement. Within the EA, it is one of the highest-growth spend categories for most enterprises — which gives Microsoft's account team strong motivation to push adoption. That motivation is leverage for buyers who know how to use it.

Use adoption data as a price anchor

Before negotiating Power Platform pricing, build your adoption data: how many users are active, across which products, using which connector types. Microsoft's opening position for Power Platform licensing is typically based on your user headcount and assumes broad adoption. If your actual Power Apps active user count is 800 out of a 5,000-person EA headcount, you have a deployment-based anchor — you are not paying for potential adoption you have not realised. This is the same scope-based negotiation technique that works for EA leverage in general: buyer-side data consistently outperforms Microsoft's estimates.

Challenge the licensing model recommendation

Microsoft's account teams have financial incentives to recommend Per-User plans over Per-App plans — the per-user unit revenue is higher. At 2,000 users accessing one app each, Per-User at $16/month = $384K/year versus Per-App at $4/month = $96K/year — a 4x pricing difference for the same functional coverage. Scrutinise every Power Platform licensing model recommendation against your actual deployment pattern before accepting it.

Power BI capacity vs PPU at renewal

If you are currently on PPU and your report consumer population has grown beyond 250–300 users, the renewal is the point to restructure to Premium Capacity. This requires a scope reduction argument (fewer per-user licences) traded against a capacity commitment (P1 or P2). Microsoft's account team may resist because per-user revenue is predictable and scalable; capacity revenue is fixed. Use the mathematical argument: at 400 PPU users, you are paying $96K/year ($20 × 400 × 12) for Power BI alone. Premium Capacity P1 at $48K/year EA covers all 400 users plus unlimited Free-licence consumers. The $48K saving is not a negotiated concession — it is a structural model optimisation you are entitled to make at renewal.

Negotiate Dataverse storage proactively

At EA renewal, negotiate a Dataverse storage buffer — additional storage units included in the EA at flat price rather than at overage rates. A common position: commit to a 3-year Power Platform deployment growth roadmap (citizen developer programme, specific app targets), and in exchange negotiate 100GB–200GB of pre-purchased Dataverse database storage at EA D-level pricing rather than list price. This eliminates mid-term storage surprises and demonstrates commercial maturity to Microsoft's account team.

For a detailed treatment of individual Power Platform components, see our dedicated guides: Power Apps Licensing: Per-App vs Per-User vs Pay-As-You-Go covers the model selection decision in depth; Power Automate Licensing: Standard vs Premium Flows covers connector classification and flow governance. For the broader Microsoft EA context in which Power Platform sits, see the Microsoft EA Negotiation Complete Guide. For understanding how Power Platform true-up obligations interact with your annual reconciliation, see the Microsoft True-Up Compliance Guide.

Power Platform Licensing Action Plan

Step 1: Inventory your actual deployment. Pull active user counts from the Power Platform Admin Centre by product and plan type. Identify premium connector flows running under M365-only users. Cross-reference Power BI content sharing against licence type of recipients.

Step 2: Build the per-app vs per-user model. For every app in your deployment, count the unique user population and applications per user. Identify populations where Per-App is more cost-effective than Per-User and vice versa. For most enterprises, a hybrid model (Per-User for developers and multi-app users; Per-App for single-application functional populations) saves 20–35% versus a uniform Per-User approach.

Step 3: Model Power BI capacity restructuring. Count your Power BI Free, Pro, and PPU user populations. Identify the view-only consumer population. Run the Premium Capacity break-even at your current scale. If above 250 users, build the renewal restructuring case.

Step 4: Establish connector classification governance. Before your next audit exposure grows, implement connector type controls in your Power Automate environment. Default to standard-connector-only for citizen developer flows; require IT review and licence provisioning for premium connector usage.

Step 5: Negotiate Power Platform commercial terms at EA renewal. Bring deployment data, model comparisons, and a growth roadmap to the renewal table. Challenge the licensing model recommendations with arithmetic, not preference. Target Dataverse storage buffer as a low-cost but valuable insurance provision.

For independent advisory on your Power Platform commercial position ahead of an EA renewal, our M365 Optimization service covers Power Platform as an integrated component of your full Microsoft licensing review. The Power Platform Licensing Enterprise Guide white paper provides the complete model selection framework for enterprise deployments.