February 2026 Release Notes

The new Synchronization tab introduces object-centric performance metrics in the Multi-Object Process Explorer component, which allow you to view the range of event lead times, execution times, and synchronization times for all objects in a process. These new metrics include:

These metrics are computed for all shared events in the process and can be viewed on the Synchronization tab by clicking the event node in the process graph. The Synchronization tab shows the range of times for each metric and how frequently each time value occurred across the events in this process. In addition, these metrics also provide the average, median and trimmed median values.

These metrics are available out-of-the-box and can help to identify critical bottlenecks in your processes such as:

For more information, see Event node details.

You can now use View modules across different Studio packages.

Previously, you could only reuse modules within your own package. This new functionality gives increased flexibility by allowing View module reuse across packages. Published changes to the reused View Module can be updated as required.

For more information, see Reusing View modules across packages.

Data engineers and analysts can now create event types directly from raw source data. This enhancement streamlines event modeling by automatically generating the SQL transformations that map event types and relationships to the data. It also enables bulk event creation and configuring event-to-object relationships.

Using the import wizard, an event will follow the raw data if it fits one of the three supported structures:

To learn more about creating events from tables, see: Events.

Action Flows are being upgraded with a new module that can be used to trigger your existing Process Copilots. This new “Call Process Copilot” module is used to create custom messages that are passed in real-time to your Process Copilot assets. Adding this module removes the need to create workaround solutions using custom code in order to connect your Action Flows to your Process Copilots.

The module can be configured by selecting your connection and the Process Copilot you want to connect to from this Action Flow. You can then create the custom message you want to send using both plain text and Action Flow variables. When the Action Flow is executed, the responses from the Process Copilot are displayed automatically and include the response, the reasoning, and the conversation ID. These calls can also be tracked on the Process Copilot Monitoring screen.

For more information, see Calling Process Copilots from Action Flows.

You can now manually control the display order of attributes for your objects and events, which were previously ordered alphabetically. After updating the attribute order and deploying your changes to development or production, the new order will be reflected in both your transformation jobs and the Knowledge Model.

While attribute ordering doesn't impact processing logic, it improves readability, usability, and governance by allowing you to present your data in a logical and consistent structure.

To learn more about creating and managing objects and events, see: Objects.

You can use annotation lines in charts to indicate thresholds and targets. Annotation lines are displayed by default for the primary axis of your chart. If your chart has both primary and secondary axes, you can now toggle annotation lines between these axes using the Annotation line settings panel:

The annotation lines will track correctly against the scales used for each axis, ensuring they are always contextually accurate.

For more information, see Chart annotation lines.

We've released a new version of the Task Mining Client software. This version includes general improvements.

In addition, we’ve added the remote suspension of Task Mining data capture for Private Preview customers only.

For more information, see the Task Mining install file release notes.

And for more information about our private preview releases, see Feature release types.

With this public preview release, you can now add a Performance Spectrum component to your Studio Views.

Traditional process explorers show aggregated views that can overlook real process dynamics. The Performance Spectrum adds a time-based perspective, revealing how tasks interact and evolve. It highlights variations, bottlenecks, and performance shifts that are often hidden in aggregations, giving you clearer insights into how your processes truly run.

In this example, the Performance Spectrum helps a government agency to visualize how individual fines move through a defined sequence over time. This helps them to identify bottlenecks and validate that specific process patterns (like batching) are adhered to and stay consistent over time (e.g. throughout the year).

To understand more about this use case and to learn how to configure and use the Performance Spectrum, see: Performance Spectrum (Public Preview)

And to read more about our private and public preview releases, see: Feature release types.

For source systems where uniqueness is defined by multiple attributes, Objects and Events now supports composite primary keys. Composite primary keys allow you to define a primary key using more than one attribute directly in the data model. This removes the need to manually concatenate attributes in transformation logic and helps improve query performance.

Previously, all object and event types in the object-centric data model required a single ID attribute with a String data type. This attribute served as the PK, uniquely identifying each row and enabling joins across objects and events. For source systems with composite PK (multiple columns), users had to manually concatenate these columns into a single ID using SQL transformations.

You can now define the PK for objects and events in one of two ways:

For more information, including a step-by-step guide to defining composite primary keys, see: Composite primary keys.

The Agent Tools (MCP) feature is a new Studio Asset that enables AI agents to securely access and interact with Celonis data and insights using the Model Context Protocol (MCP). The Agent Tools asset provides ready-to-use tools that can be used by AI agents to search for relevant information, retrieve real-time process context, trigger actions in external systems, and write decisions back to the Celonis Platform.

Agent Tools can be created as new assets in Studio and then configured with the tools you want to make available to AI agents through the MCP. These tools can include data retrieval functions, insight generators, knowledge model queries, and more. Once the new asset is published, it will generate a unique URL that serves as an endpoint that can be connected to by other MCP clients.

For more information, refer to Agent Tools (MCP) in the Celonis Developer Center.

We've updated the Process Cockpit to be more intuitive and user friendly.

UI Changes

Performance enhancements

For more information, refer to Activating and configuring live data and Process Cockpit.

This release focuses on "under-the-hood" excellence and highly requested fixes for escalated customers:

UI updates:

Our Business Miner service has been removed from the platform and has already been replaced by our AI-driven Insight Explorer and Studio features.

With these advanced capabilities, you’ll benefit from:

As a result of the removal of this service, the Business Miner navigation item has been removed from the navigation bar and no users can access the service. As our Quickstarts feature rely on the use of Business Miner features, we have also removed these from the platform.

To get started with Insight Explorer, see: Insight Explorer.

Transformation Center has been removed from the platform as its features transition into our more powerful Transformation Hub. With improved AI-driven insights and more precise KPI tracking, Transformation Hub is included in your existing Celonis Platform subscription at no additional cost.

As a result of the removal of this service, the Transformation Center navigation item has been removed from the navigation bar, with no access available to users.

To get started with Transformation Hub, see: Transformation Hub.

Following the deprecation of the Business Miner service, the associated data integration Quickstarts have been removed from the platform. As these Quickstarts were only used to establish initial connections to source systems, this change does not affect existing workflows - any connections created with a Quickstart will continue to function as before.

To learn how to connect your source system to the Celonis Platform using our Extractor Builder options, see: Connecting data sources

You can now create a “router” Process Copilot in Studio and use it to choose the correct tool for your specified task. To take advantage of this router Process Copilot, you can build a series of focused Process Copilots with specific skills or for a defined scope and then surface them as tools to the router. Then your users can interact with the router and use it to perform any of the tasks provided by the associated tools.

For example, you can build one Process Copilot per domain or one Process Copilot for each task. This allows each Process Copilot to keep its focus narrow and maintain high quality. You can then expose each of these focused Process Copilots as individual tools to the router Process Copilot. The router can then search these tools to find the Process Copilot that best fits the needs of each request.

When creating the Process Copilots to be used as tools, it is important to provide clear and accurate descriptions to help the router identify its functionality. Each of the Process Copilots being surfaced as tools must be published in the same Space as the router.

With this private preview release, the colors assigned to event logs will remain consistent across all your process mining components and assets. This means that the color assigned in the Event Log Builder or visual editor will be used for that event log in all of your process mining components, such as Process Explorer, Variant Explorer, Process Adherence Manager (PAM), Event Explorer, Case Explorer, Throughput Time (TPT) Explorer and any associated process filters. If the assigned color is changed in the Event Log Builder or visual editor, the change will be updated automatically across all components.

As part of this release, the color palette used for event log colors has been refreshed to make the available colors easier to distinguish from one another and improve the accessibility of the colors used in the process mining components.

To learn more about our Private Preview releases, see: Feature release types.

The new Synchronization tab introduces object-centric performance metrics in the Event Explorer component, which allow you to analyze the arrival times of various objects at specific interaction points in the process. These new metrics include:

These metrics are computed for all shared events in the process and can be viewed on the Synchronization tab in the Event Explorer by clicking the arrow to the right of the Event logs column. The Synchronization tab shows the range of times for each metric and how frequently each time value occurred across the events in this process.

These metrics are available out-of-the-box and can help to identify critical bottlenecks in your processes such as:

For more information, see Event Explorer Component.

You can now choose the color theme used in the Celonis Platform, allowing you to adjust the user interface colors and contrast to suit your needs.

You can select the following preset themes from your user profile management screen:

• System: Uses the browser preferred theme settings.

• Light (default): Classic, clean light-themed interface.

• Dark: Optimized for low light and reduced eye strain.

• High Contrast: Designed for maximum readability and accessibility.

You can also switch between light and dark color themes from Studio Package settings:

With the addition of the dark mode platform theme, we've also expanded Studio color themes to support dark mode-specific color options. This means you can ensure your custom brand color themes remain readable even when users have dark mode enabled.

The dark mode colors match light mode by default. Users can manually override these colors to provide better contrast or visibility against dark backgrounds. This customization specifically applies to Custom Brand Color Themes; Celonis default themes already include automatic translation for dark mode support.

For more information about choosing the theme for your Celonis Platform experience, see Managing your user profile. And to find out more about colors in Studio Views, see Color themes.

We've fixed an issue with the incorrect labeling of some applications in Task Mining.

Previously, Microsoft applications such as Microsoft Outlook were sometimes automatically and incorrectly labeled as Microsoft Teams, potentially giving an inaccurate view of the amount of time users spend using Microsoft Teams.

With this fix, all Microsoft applications will now be labeled correctly during Task Mining. This fix will be automatically applied to all current Task Mining projects and may result in changes to your Workforce Productivity app data for those Task Mining projects.

For more information, see Task Mining default Labels: Microsoft Office. And to find out more about the Workforce Productivity app, see Working with Task Mining data.

Our Business Miner service is being deprecated and has already been replaced by our AI-driven Insight Explorer and Studio features.

With these advanced capabilities, you’ll benefit from:

To prepare for this deprecation, the Business Miner navigation item has been moved down a level and an in-platform notification informs your users of the forthcoming deprecation. As our Quickstarts feature rely on the use of Business Miner features, we have also removed these from the platform.

To learn more about this deprecation, see: REMOVED BUSINESS MINER Deprecating Business Miner service.

And to get started with Insight Explorer, see: Insight Explorer

Transformation Center is being deprecated as its features transition into our more powerful Transformation Hub. With improved AI-driven insights and more precise KPI tracking, Transformation Hub is included in your existing Celonis Platform subscription at no additional cost.

To prepare for this deprecation, the Transformation Center navigation item has been moved down a level and an in-platform notification informs your users of the forthcoming deprecation.

To learn more about this deprecation, see: REMOVED TRANSFORMATION CENTER Deprecating Transformation Center service.

And to get started with Transformation Hub, see: Transformation Hub.

Teams using Celonis core-process catalog v3.0.0 or higher now benefit from two enhancements to both automatic and manual catalog update workflows.

To learn more about updating your catalog version, see: Core process catalog.

When scheduling and managing your View Reports, you can now choose between creating a personal connection, where the report is linked to an individual user in the team, or a technical connection.

A technical connection allows you to link a report or report trigger to a team account instead of an individual user in the Celonis Platform. By assigning View permissions to this connection, the report or trigger remains accessible and active even if the original creator leaves the team or loses Celonis Platform access.

In this example, the report is configured for a technical connection:

To create a technical connection, a team admin must create a reporting application in the Admin & Settings area. For more information, see: Technical connection for View reports.

And to learn more about scheduling and managing View reports, see: View Reports.

With this Private Preview release, you can now import translations from a Data Model to a Studio package.

Translations in an Analysis use Data Model name mappings. In Views, translations are saved in the package settings. To migrate translations, the name mappings can now be imported to the package settings.

When migrating an Analysis to a View, the translation syntax is automatically updated during the migration and uses the translation keys from the package, with no changes required in the Analysis or View. Users can then select their preferred language from the View Translations menu.

For more information about migrating Analyses, see Migrating Analysis to Views and for information specific to this functionality, see Migrating translations from an Analysis to a View.

And for more information about our private preview releases, see Feature release types.

We’re introducing OAuth 2.0 as a connection method for all Celonis Action Flow modules. This industry-standard framework allows applications to interact securely on your behalf without sharing sensitive credentials. You can find OAuth as a connection type in Action Flow modules. This connection type requires a Celonis Admin to create an OAuth client first to obtain client ID and client secret (basic type). See Adding OAuth client to Celonis Platform.

Annotation Builder, our AI-enhanced Studio asset for processing and labeling data, can now be used as a starting step for Process Orchestrations. This allows you to automatically trigger subsequent orchestration steps based on the specific findings and labels generated within Annotation Builder.

For more information, see Annotation Builder and Orchestration Engine.

We’re introducing a refreshed permissions management experience that improves how administrators view and manage access across the platform.

This release includes:

While this update focuses on improving clarity, consistency, and usability for administrators, the underlying permission model remains unchanged:

To learn more about managing your user permissions, see: Managing permissions.

The interface for the Relationships to objects tab in the Objects and Events module has been updated to provide a clearer overview of your object-to-object relationships. In this new interface, the Outgoing and Incoming tabs have been replaced with a single list that shows all your one-to-one relationships on one page.

In this new list, the positioning of the objects on the card is meant to reflect their positioning within the relationship. For example, all one-to-many relationships will display the 1-sided object on the left. For many-to-many relationships, the current object will be displayed on the left. If there is a join table in the relationship, it will be displayed between the two objects.

The list also uses the color of the chips to differentiate between the objects in the relationship. A dark chip will always be used for the current object and a light chip is displayed for the other objects. You can also click on a light chip to be taken to the page for that object.

For more information, see: Modeling objects and events.

With this Private Preview release, you can now generate Insights from within Views.

Previously, users generated Insights about their data in Insight Explorer before exporting them to Views. Now users can create Insights directly within Views containing KPI list or card components and apply the resulting Insight filters directly to their View.

For more information, see Generating Insights in a View.

And for more information about our private preview releases, see Feature release types.

In this bundled release, we’ve refreshed several View components with visual and functional updates to improve the look, feel, and usability of your dashboards and applications. These changes reduce wasted space, standardize the UI, and add new customization options for greater information density.

The highlights include:

To learn more about available Studio components, see: Components.

Previously, Celonis limited string values in Cloud Extractors to a maximum of 80 characters. Based on customer feedback, this character limit has been increased. Cloud Extractors now support string values of up to 8 KB, with an increase to 65 KB planned in the coming weeks.

Important: Celonis dynamically determines column widths based on actual data to optimize performance and avoid unnecessary processing overhead. However, manually created tables, especially VARCHAR columns defined during the transformation phase, may require manual adjustment if incoming values exceed the existing column size.

If column widths are not updated, you may encounter transformation errors such as:

String of 82 octets is too long for type Varchar(80)

For extraction task best practice, see: VARCHAR optimization.