Microsoft is addressing that gap in Power Automate Process Mining by rolling out the ability to create and visualize custom KPIs (key performance indicators) directly within the web-based process intelligence experience. Tracked under Microsoft 365 AI at Work Roadmap ID 576868 and part of the Power Platform 2026 release wave 1 plan, the capability is marked as Launched for the Web platform in the Worldwide (Standard Multi-Tenant) cloud, having entered public preview on May 31, 2026 (tracked via Message Center post MC1310386) and reached general availability in July 2026 (specifically noted in admin communications as July 26, 2026). The feature is enabled automatically for admins, makers, and marketers.
What Custom KPIs Bring to the Process Intelligence Workspace
At its core, Power Automate Process Mining extracts event data from your systems of record to map out how work actually happens across your organization. While standard baseline metrics give you a decent starting pulse, Microsoft’s roadmap entry for ID 576868 outlines a more flexible framework that lets business teams author their own formulas without waiting on IT or external vendors to build custom reporting pipelines.
According to the roadmap specifications and Microsoft 365 Message Center documentation, the update introduces several core capabilities to the process intelligence experience:
- Formula-Driven Metric Creation: Makers and analysts can build custom KPIs through a configurable formula editor that combines existing process data, standard KPIs, case attributes, event data, timestamps, and mathematical operations.
- Centralized Editing and Maintenance: As business targets or compliance rules shift, users can modify, refine, or delete their custom metrics. Any adjustments automatically cascade across every visualization that references that metric.
- Native Visualization Support: Custom metrics are treated as first-class citizens alongside built-in KPIs. They can be surfaced across process intelligence tiles and components, including process maps, trend charts, statistical breakdowns, and comparative analyses.
- Reusability and Dynamic Evaluation: Once defined, a custom KPI can be reused across filters and analytical scenarios throughout the process intelligence environment. Calculations are evaluated dynamically against the underlying process data so that metrics stay accurate after every data refresh.
- Domain-Specific Measurements: Teams can codify proprietary logic—such as weighted performance scores, custom cost calculations, customer-specific SLA thresholds, compliance rates, or industry-specific efficiency ratios.
Section Summary: Rather than forcing analysts to export event logs into external spreadsheets just to calculate a bespoke SLA or cost ratio, Roadmap ID 576868 brings formula-based metric authoring directly into the Power Automate web portal, complete with automatic updates across dashboards and filters.
How This Fits Into the Web Interface (and Desktop App Precedent)
To understand why this matters to daily Power Platform practitioners, it helps to distinguish between Microsoft's older desktop tooling and the modern web interface.
For years, heavy-duty formula creation in Microsoft's process mining ecosystem was primarily the domain of the Process Mining desktop application (rooted in Microsoft's Minit acquisition). As documented on Microsoft Learn, the desktop app has long offered a Process context screen where analysts can write calculated metrics using syntax highlighting, IntelliSense with function overload hints, and data-type validation, applying them to desktop process maps, the Statistics view, and root-cause analysis (RCA).
Meanwhile, on the web side, Microsoft introduced the process intelligence experience—a next-generation, card-based analytical workspace inside Power Automate Process Mining. According to Microsoft Learn's overview of the experience, this web interface replaces the older static process overview with a customizable grid layout featuring:
- Active process views and a Tab bar (hosting views like Overview, Case Duration, and Variant Analysis, plus custom tabs).
- A Filter bar where active filters appear as removable pills that apply across all cards on the active tab.
- A Control bar with real-time data refresh controls and an Edit layout mode for arranging and resizing dashboard cards.
Bringing custom KPI creation and visualization directly into this web-based studio closes a major workflow gap. Previously, web dashboard users configuring a KPI card were documented as selecting from standard metric categories—such as Frequency, Performance, Rework, or Finance—and optionally dividing a primary metric by a secondary metric with a slash (/). By integrating custom KPIs natively into the web experience's tiles, maps, and filters, business users can tailor their web dashboards on the fly without constantly round-tripping back to the desktop client.
| Feature Aspect | Process Mining Desktop App (Existing Precedent) | Process Intelligence Web Experience (Roadmap ID 576868) |
|---|---|---|
| Primary Environment | Windows desktop client application | Power Automate web portal (Worldwide Standard Multi-Tenant) |
| Metric Authoring | Defined in Process context with syntax highlighting, IntelliSense, and validation | Configurable formula editor using case attributes, event data, timestamps, and existing KPIs |
| Where Metrics Appear | Process map (activities/edges), Statistics view columns, Root Cause Analysis (RCA) | Process intelligence cards/tiles, process maps, trend charts, statistical breakdowns, comparative analyses |
| Filtering Support | Supported across most filters (excluding Timeframe, conditional Edge, and Variants) | Reusable across process intelligence filters and analysis scenarios |
Section Summary: While the Process Mining desktop app established the precedent for calculated metrics, bringing custom KPI authoring and visualization into the web-based process intelligence experience makes self-service analytics much more accessible to everyday makers and stakeholders.
Navigating the Documentation Gap: What Admins Should Know
Now for a dose of veteran IT realism: if you head over to Microsoft Learn's support page for KPI cards in the process intelligence experience (process-mining-studio-kpi-card) expecting a comprehensive tutorial on the new formula syntax, you are going to come up short.
As of its June 12, 2026 update, the official Microsoft Learn KPI card documentation still only outlines the basic steps for labeling a card, picking a standard metric type (Frequency, Performance, Rework, or Finance), choosing a metric from the dropdown, and toggling on a secondary metric. It does not yet detail the new web formula editor's exact operator syntax, governance permissions, or formula complexity limits. Similarly, Microsoft’s public AI at Work roadmap portal automatically purges detailed item descriptions once a feature transitions to General Availability.
For administrators and process analysts exploring the newly launched web capabilities in their tenants, here is how to navigate the workspace based on Microsoft Learn's documented layout:
- Access the Workspace: Sign in to the Power Automate web portal, select Process mining from the left navigation pane, and open an existing process under Create new process (or via All processes).
- Enter Customization Mode: Use the Control bar in the top-right corner of the process intelligence dashboard to select Edit layout or Edit tabs.
- Configure Cards and Filters: When adding or editing KPI cards, trend charts, or filter pills on the Filter bar, look for your custom-defined metrics alongside standard baseline categories so they can be applied consistently across your active tab.
Why does this self-service push matter? Because in enterprise IT, a dashboard that requires a ticket to the data engineering team every time a department head wants to tweak a weighted SLA score is a dashboard that eventually gets abandoned. At the same time, IT leaders should keep an eye on metric sprawl: when makers, marketers, and admins can all spin up custom formulas automatically, teams will still need internal agreement on which custom KPIs represent the "single source of truth" during quarterly reviews.
Are your business analysts already building custom calculations in Power Automate Process Mining, or are you still wrangling event logs in Excel and Power BI? Share your experiences and dashboard tips with the WindowsForum community!