Power Platform cost optimization
Kinga
Posted on November 7, 2024
Power Platform Cost Estimates
Using Microsoft Power Platform for application modernization not only reduces Total Cost of Ownership (TCO) but also helps minimize technology debt.
Unlike traditional coding approaches, low-code solutions require fewer resources for maintenance and accumulate less technical debt, as the platform provides the necessary infrastructure. Power FX, despite its ongoing evolution, remains backward compatible and utilizes a "backward compatibility converter" that automatically adjusts formulas if an incompatible change is detected.
However, depending on the complexity of the application, specific features, and the resources being connected, you may need to purchase premium licenses.
If you're modernizing an existing application and have access to reference data, you can estimate the licensing costs by analyzing historical usage.
The report
License Cost Estimation
The License Cost Estimation page provides cost estimates based on the assumption that all users have the same type of license.
For the Per User and Per App estimates, the total number of unique users per year is considered. In the case of PAYG (Pay-As-You-Go), the count is based on version history, covering all users who have created or edited items during a specific calendar month, excluding readers.
The number of months for which data is available is shown on the cards. If the data does not cover the full year, the yearly user count or cost is not extrapolated.
License Cost Optimization
To find out, which users should be assigned PAYG, it's important to find out the maximum number of months the user is accessing the app.
This threshold is determined by comparing the costs: (number of Months x 1 User x PAYG) < (12 Months x 1 User x Per App)
.
The simulation assigns less active users to the PAYG (Pay-As-You-Go) plan and calculates the associated costs.
These users are then removed from the list of unique active users for the year. The system also identifies users who have never accessed the app, such as optional approvers. These users, listed under "Assigned users" but absent from the "Active users" list, will not be assigned a license. Instead, they can access the app through PAYG when needed. You may also consider notifying them via email about the request and its contents.
The remaining list will include only the "more active" users with assigned roles.
Additionally, there may be users who, while not assigned as approvers, are responsible for system operations or user support. These users will appear in the active users list but not in the assigned users list. Based on their activity, they are categorized either as less active users eligible for PAYG billing or more active users who should be assigned a license.
Depending on how many premium Power Platform applications users access, the appropriate license - Per User or Per App - should be chosen. Work closely with the Power Platform administrator to determine the best approach.
It's also possible that users initially assigned PAYG access in the simulation may require access to additional applications, in which case a Per User license would be more suitable. PAYG is "stackable," meaning that users who already have a license won’t be charged under PAYG.
You can use the "%users with per user license" slider to simulate different scenarios and view the corresponding cost.
The calculations above apply exclusively to Power Apps premium licenses. It's common to use Power Automate for more complex logic or to send notifications. If the flow shares the same data sources as the app, it can be linked to the app to avoid the need for additional licenses.
Power Platform Cost Estimations template
To use this Power Bi Template, download the PowerPlatformCostEstimations.pbit file from GitHub and follow the https://learn.microsoft.com/en-us/power-bi/create-reports/desktop-templates#using-report-templates procedure to import it to your Power BI desktop.
Follow the steps outlined in the PowerPlatform CostEstimates GitHub repo to configure the report.
Posted on November 7, 2024
Join Our Newsletter. No Spam, Only the good stuff.
Sign up to receive the latest update from our blog.