Microsoft Fabric
Fabric Planning
Microsoft Fabric Planning

How Fabric Plan capacity and pricing work

Fabric Plan delivers Planning, PowerTable and Intelligence, deeply integrated with Microsoft Fabric.
Activity-Based Pricing

Only Active Users Consume Capacity

Fabric Plan is priced on user activity, not on named users. A session starts when a user engages with a Plan item.

Flexible Licensing

There are no named-user licenses to purchase, assign or renew. Users access Fabric Plan through the Fabric workspace.

Pay for Active Use

A session is billed only when a user engages with a Plan item. Unused capacity can be consumed by other Microsoft Fabric workloads.

Include Ad-hoc Users

Seasonal contributors, budget reviewers and one-time approvers can participate without driving up annual license costs.

MACC credit eligibility

Fabric Plan runs on Microsoft Fabric capacity, which is an eligible Azure service for MACC. Draw on Azure spend your organization has already committed.
Activity-Based Pricing
Fabric Planning vs. Other Products

Save 55% to 90% on Your Costs

Fabric Plan delivers enterprise planning at a fraction of the cost, with minimal dependency on IT.
Cost per user / year →
Specialist-led
Self-service
Business User Autonomy →
Fabric Plan pricing reflects reserved capacity rates for blended use by Planners and Stakeholders, and includes a 30% overhead towards other Fabric resource consumption. The Business User Autonomy axis reflects the degree to which finance and operations users can build, configure and change planning models themselves, without depending on specialists: IT staff, certified consultants, model builders or platform architects.
  1. The pricing information presented is based on publicly available data and industry research as of the date of publication.
  2. Pricing for Fabric Planning assumes 1 Planner : 5 Stakeholders in reserved capacity.
  3. Vendor pricing is estimated and may vary by year, industry, country, client size, license count, license term, bundling discounts, negotiated agreements, or product configurations.
  4. This comparison reflects Lumel's interpretation and should not be relied upon as definitive. You are encouraged to verify pricing directly with each vendor.
Fabric Plan vs. Others
Users and Sessions

Session Types

Fabric Plan meters three session types, and each consumes Fabric capacity at a different rate.

Viewer

Consumes Plan items with read-only access.
Fabric Workspace Role

Viewer

(reading view)
Capabilities
View data in reading view
Sort, filter and bookmark reports

Stakeholder

Contributes to Planning sheets, and authors PowerTable and Intelligence sheets.
Fabric Workspace Role

Viewer, Contributor or Member

(reading view and edit mode)
Capabilities
Includes all Viewer capabilities
Reading view
Fabric Viewer role · across Planning, PowerTable and Intelligence
Enter data and collaborate
Create scenarios and simulations
Provide approvals and comments
Save and write back data
Trigger audit change capture
EDIT MODE
Fabric Contributor or Member role · PowerTable and Intelligence
Author reference data and tables in PowerTable
Author reports and dashboards in Intelligence

Planner

Authors planning models, and administers security and permissions.
Fabric Workspace Role

Contributor, Member or Admin

(edit mode)
Capabilities
Includes all Stakeholder capabilities
Edit mode
Fabric Contributor, Member or Admin role · Planning
Author planning models and layouts
Author input templates for data entry
Configure planning logic, rules and allocations
Customize security settings and user permissions
Session Types
Billing follows what you do in Fabric Plan, not your workspace role
A user with Contributor or Member rights who authors PowerTable and Intelligence content in edit mode is billed as a Stakeholder. Authoring Planning models in edit mode is the single capability that moves a user to a Planner session.

Access by sheet type

Each cell shows the highest level of activity a session type can perform in that sheet type.
Session type
Sheet typeViewerStakeholderPlanner
PlanningBudgets, forecasts, scenarios, allocationsViewContributeAuthor
PowerTableReference and master data managementViewAuthorAuthor
IntelligenceReports, dashboards and analysisViewAuthorAuthor
View
Read-only
Contribute
Enter data, run scenarios, approve, comment and write back inside an existing plan.
Author
Create and edit that sheet type’s Plan items in edit mode.
Capacity Estimator

Size Your Fabric Capacity

Estimate your Fabric consumption in Capacity Units (CU), the underlying billing metric. Enter your estimated active users by session type to arrive at a total CU consumption. Use that total to size the Fabric SKU you need. Finally, pick your region and currency below to estimate what it costs.

Capacity Estimator

User Type
Active Users
Average CU (Sustained)
CU-hours per session (730 hrs)
Planner Author, modeler and admin
1.16 CU
847 CU
Stakeholder Enter data, approve, comment & writeback
0.23 CU
168 CU
Viewer View and interact locally
0.05 CU
37 CU
Subtotal
100
1.4 CU
1,051 CU
1.9 CU
1,367 CU

Your workload fits in an F2 Fabric capacity (2 CU)

Based on total CU including Fabric resource overhead.

  1. Microsoft meters Fabric Planning as 30-day sessions (730 hours). Once a session is triggered for a user, the meter runs for the full 730 hours regardless of usage.
  2. Consumption of Fabric SQL, OneLake, Power BI XMLA API, and other native Fabric items are additional CU meters. Budget an additional 30% of estimated CU towards this, as indicated above. Actual consumption will vary based on data volume, query patterns and active usage.
  3. Automation jobs (optional) for advanced use cases like connected planning and automated cascading updates are metered separately and not included above. Each automation job consumes 2 Capacity Units.
  4. CU rates are set by Microsoft and apply uniformly across regions. Dollar cost of each CU varies by region – see the latest Fabric billing and metering documentation.
Capacity Unit Consumption by Session Type1
Session typeActive UsersCU/secondCU-hours (730 hrs)
Planner
1.16 CU
846.8 CU-hrs
Stakeholder
0.23 CU
167.9 CU-hrs
Viewer
0.05 CU
36.5 CU-hrs
Subtotal31.44 CU1,051.2 CU-hrs
Add: % for other Fabric resources 21.87 CU1,366.6 CU-hrs
Your workload fits in an F2 Fabric capacity (2 CU)
Based on the CU/second total in the table above, including the Fabric resource overhead. A SKU's CU rating is itself a per-second rate (an F2 supplies 2 CU every second), so your total compares directly with the F number, with no conversion.
Pay-as-you-go
—
per month
Reserved capacity
—
per month, about 41 percent less

  1. Microsoft meters Fabric Plan in sessions. A session starts when a user engages with a Plan item and runs for 30 days, whether or not the user returns.
  2. Consumption of Fabric SQL, OneLake, Power BI XMLA API and other native Fabric items are additional CU meters. Budget an additional 30% of estimated CU towards this, as indicated above. Actual consumption will vary based on data volume, query patterns and active usage.
  3. Automation jobs (optional) for advanced use cases like connected planning and automated cascading updates are metered separately and not included above. Each completed job consumes a fixed 2 CU-hours of capacity, whatever its runtime.
  4. CU rates are set by Microsoft and apply uniformly across regions. Dollar cost of each CU varies by region. See the latest Fabric billing and metering documentation.
Capacity Estimator
Sizing reference

Capacity and Sessions

The estimator above sizes a capacity from your active user counts. This section shows how much capacity each session consumes, and how many sessions a given capacity holds. A session, once started, reserves its share for 30 days.
What each session consumes

Sessions come in three types: Planner, Stakeholder and Viewer. Fabric meters them in timepoints, a unit lasting 30 seconds. An F2 makes 2 × 30 = 60 CU-seconds available per timepoint, an F64 makes 64 × 30 = 1,920 CU-seconds per timepoint. When a user engages with a Plan item, a session starts and draws its rate from every timepoint for the next 30 days, whether or not the user is active.

RateIn CU-secondsIn CU-hours
Session typeCU/second
the session’s rate
Per timepoint
rate × 30
Per hour
rate × 3,600
Per month (730 hrs)
per hour × 730
CU-hours (730 hrs)
per month ÷ 3,600
Planner
1.1634.84,1763,048,480846.8
Stakeholder
0.236.9828604,440167.9
Viewer
0.051.5180131,40036.5

CU-seconds is the unit the Fabric Capacity Metrics app reports. CU-hours is the same quantity divided by 3,600, and it is the unit Microsoft publishes session and automation figures in. Both measure capacity consumed, not time elapsed, the way a kilowatt-hour measures energy rather than an hour.

How many sessions each capacity holds
CapacityCU
Planners
1.16 CU
Stakeholders
0.23 CU
Viewers
0.05 CU
EST. CU
30% Fabric overhead incl.
F221121.9
F4415103.7
F8828408.0
F16163156012.9
F323273514030.1
F6464157529063.9
F12812830150580127.8
F256256603001,150254.9
F5125121206002,300510.0
F10241,0242401,2004,6001,019.7
F20452,0484802,4009,2002,039.4
  1. Based on an approximate 1:5:20 mix of Planners, Stakeholders and Viewers, with all users active concurrently. Where that exact ratio would exceed a capacity, the mix is trimmed to fit, most noticeably on F2, F4 and F8.
  2. A 30% CU allowance is reserved for other Fabric resources. Actual capacity required will vary by user activity and workload mix.
  3. Automation jobs are additional and not included above. Each completed job consumes 2 CU-hours, so 12 jobs a day is 24 CU-hours, the equivalent of 1 CU held for the whole day.
  4. The estimates above assume all listed users have active concurrent sessions. If not all users trigger a session in a given month, actual CU consumption will be lower, meaning you may be able to (a) support more users within the same capacity, (b) redirect unused capacity towards other Fabric workloads or (c) consider a lower capacity SKU.
Capacity Planner
Recommendations

Best Practices

Size the capacity for Plan sessions, with room for automation jobs and other Fabric resources. That headroom keeps it responsive.
1
Give Plan its own capacity
Run Fabric Plan on a dedicated capacity rather than sharing one with other Fabric workloads. Microsoft’s guidance for Fabric recommends dedicated capacities for critical workloads.
2
Size your capacity with headroom
Keep sessions under 75 percent of a dedicated capacity, including the Fabric resource overhead. On an F32 or smaller shared with other workloads, keep them under 50 percent.
3
Size on active users, not headcount
Size for active users in a given month, and not total users. A session starts only when someone engages with a Plan item.
4
Assign Fabric workspace roles based on need
Assign the Viewer role to users who do not need to edit or create Plan items. They can still enter data, provide approvals, comment, write back and run scenarios in reading view.
5
Keep Plan on one capacity
Each combination of user and capacity starts its own session. Consolidating Plan workspaces onto a single capacity keeps each user on one session.
6
Monitor before throttling starts
Set capacity alerts and use the Fabric Capacity Metrics app to see where consumption is going. Request spare regional quota in advance, since busy regions can be slow to grant it.
7
Scale up in place, never migrate
Sessions are tied to the capacity they started on. Moving workspaces starts fresh sessions there while the originals run out their 30 days, so you pay for both.
8
Reserve the capacity you always need
A reservation cuts 41 percent off the pay-as-you-go price. You can opt to reserve part of a capacity and let the rest run pay-as-you-go.
Best Practices
Enterprise agreements with negotiated pricing may carry additional reservation discounts.

Did You Know

PowerTable and Intelligence at no additional cost
One Fabric capacity subscription covers all three. Plan users incur no additional cost for using PowerTable or Intelligence.
Users who do not engage cost nothing
A session starts only when someone engages with a sheet in a Plan item. Users who do not engage in a given month consume nothing at all.
Extra workspaces do not add sessions
A user working across several Plan workspaces on the same capacity runs one session, not one per workspace.
Did You Know 
FAQ

Common questions about session billing

A session starts in two ways. One, when a user creates a new Plan item, assigns data and saves it. Two, when a user engages with a sheet in an existing Plan item, in edit mode or reading view.
A session runs for 730 hours from the moment it is triggered, equivalent to a 30-day month. The meter runs for the full period regardless of usage.
No. Once a session starts, there is no way for the user to terminate it early. It runs to completion.
When the session expires, a new session starts the next time the user engages with a Plan item, at whatever session type that engagement establishes.
Users can upgrade (Viewer to Stakeholder, Stakeholder to Planner) but cannot downgrade within an active session. On upgrade, the earlier session closes and is billed for the consumption up to that point. A new session starts at the higher session type, running from the time of the upgrade
Each unique combination of tenant, user and capacity creates a separate session, billed independently. A user working across two capacities will have two active sessions.
If the same capacity is assigned to multiple workspaces, the user is billed at the highest session type active across all workspaces under that capacity.
Yes. Each completed job consumes a fixed 2 CU-hours of capacity, whatever its runtime and whichever session type triggered it. Only successful jobs consume capacity, and failed jobs consume nothing. Fabric spreads that consumption across the following 24 hours so background jobs do not spike the capacity.
Pausing or deleting a capacity does not stop or shorten the sessions running on it. The full session period is billed for every active session, so pausing does not avoid the charge.
Active sessions continue to run and are billed through to the end of the session even if the Plan item is deleted.
Plan sessions are not suspended and continue to consume for their full period. What changes is performance: the capacity throttles for every workload running on it.
The capacity is overcommitted: demand exceeds what the SKU delivers, and sustained overcommitment throttles it. Small SKUs such as F2 and F4 reach that point soonest, as do capacities shared with other Fabric workloads.
A capacity that goes past 100 percent of its available Capacity Units keeps running for a while by borrowing from future timepoints. Overage of up to 10 minutes is absorbed with no throttling at all. Beyond that, throttling arrives in stages: interactive actions slow down after 10 minutes of overage, interactive actions are rejected after 60 minutes and background operations are rejected after 24 hours.
No. Scale the existing capacity up in place. Plan sessions are specific to the capacity they started on, so moving workspaces to a new capacity starts a fresh set of sessions there while the original sessions continue to run and bill to the end of their 30 days. You pay twice for the same users.
FAQ
About this page

Lumel is a strategic ISV of Microsoft and the development partner behind Microsoft Fabric Planning.

This page is an overview of Microsoft's Fabric Planning capability. It should not be relied upon for decision-making without verifying with Microsoft and is subject to your license agreement or any other agreement with Microsoft. Microsoft's strategy and possible future developments for Fabric Planning are subject to change by Microsoft at any time for any reason without notice.

This document is provided without warranty of any kind, either express or implied, including but not limited to the implied warranties of merchantability, fitness for a particular purpose, or non-infringement. Lumel assumes no responsibility for errors or omissions in this document, except in cases of intentional misconduct or gross negligence by Lumel.

Microsoft and Microsoft Fabric are trademarks of Microsoft Corporation.

ON THIS PAGE

Request a demo

Learn how Lumel helps enterprises deliver real-time integrated reporting and planning applications

Get Lumel Brochure

Enhance your BI, analytics and xP&A use cases with our no-code Data App suite for Power BI.
Download now
Lumel
Look Forward. Think Ahead ®
Leader in Unified Planning & Analytics for the Modern Data Stack.
© Lumel Inc. All rights reserved.
Connect With Us
Headquarters
5920 Windhaven Pkwy,
Plano TX 75093
United States