December 2025 Release Notes

We've released a new version of the Task Mining Client software. This version includes fixes for:

For more information and a complete list of fixes included in this release, see the Task Mining install file release notes.

When importing a BPMN model into Process Adherence Manager, PAM now detects common mistakes within the model (missing start/end events, duplicate or missing task/event names, missing <incoming>/<outgoing> tags) and automatically "repairs" the model to continue the import.

In this example, the imported model had two events with the same name (Task B). PAM detected this, and solved the problem by assigning a suffix that turned them into unique names.

For more information, see Importing a BPMN model.

We’re replacing and upgrading the publishing workflow in Studio and Objects and Events with a new versioning and deployment system, giving you greater control over your content. Initially released in public preview, these features allow you to create and manage package and data models versions, track changes, deploy to development and production environments, and efficiently roll back to previous releases.

For a video overview of these features:

To learn more about our public preview releases, see: Feature release types

You can now use native Delta Sharing connections to transfer data to the Celonis Platform, enabling true zero-copy data integration from your data platforms, such as Databricks and Jira. By leveraging the open Delta Sharing protocol, you can access your raw data directly at the source with near-zero latency, ensuring immediate data freshness in your Celonis data pools.

Delta Sharing simplifies your data pipelines by eliminating complex extraction steps and the duplication of data, allowing you to connect once and begin consuming data immediately. Additionally, since your raw data remains secured in its original location, it preserves your organization’s data governance.

For a video overview of using Delta Sharing:

For more details, see Delta Sharing.

The AI Settings in the Admin module are being upgraded with new and more powerful language models. Administrators can now enable the GPT-5 and, GPT-5 mini models for use with their team’s Celonis AI products. Once enabled by the admin, users can leverage the increased intelligence of these new models for use within their AI-enabled Celonis applications.

For more information, see: AI Settings.

With this private preview release, we've added a new View component called KPI meter.

The KPI meter component displays the current value of a metric on a bullet chart, with color coding indicating different KPI thresholds. You can customize the threshold values and colors used in the bullet chart.

In this example, a dashboard contains KPI meter components that display KPIs related to orders. The color customizations of the different thresholds give a clear visual indicator of how the actual KPI data compares to the defined threshold values.

In addition, the KPI meter component supports the migration of Analyses that use these Single KPI components:

For more information, see Configuring the KPI meter component.

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

You can now download a report of all application keys and OAuth clients used in your Celonis Platform teams. The report provides a comprehensive overview of your applications and includes information on how many application keys exist in a given team, who and when each key was created, and what permissions are associated with them.

For more, see Downloading application key report.

The list of available KPI views in multi-object Process Explorer and Variant Explorer configured on object-centric data will now include two new out-of-the-box options: Object count and Event count. These new KPI views are similar to the existing "Object and Event counts" KPI view, but will allow you to focus the metrics displayed in your process graph view on the specific object or event type.

When using the new Object count KPI view, you will see the object count on the connection (the total number of objects that flow through a connection) as well as the event node (i.e. the total number of objects per object type, for which the event has occurred one or more times) of your process graph.

In the new Event count KPI view, your process graph will display the total number of times that connection occurs across all objects of that object type on the connection and the total number of times that event occurs for each object type on the event node.

The existing Object and Event Count KPI view displays object counts on the connections (i.e., the total number of objects that flow through a connection) and event counts on the event nodes (i.e., the total number of times an event occurs for each object type). The Object and Event Counts KPI view will still be available from the KPI view dropdown.

You now have greater visibility and control over your asset-to-asset dependencies within Studio, reducing the possibility of broken content and dashboard errors. Studio packages often include multiple assets that are connected and rely on each other. Because of these dependencies, action made to one asset, such as publishing, copying, or deleting it, can have a direct impact on other assets that reference it. For instance, deleting a Knowledge Model could break the Views that rely on it.

To help you manage these relationships, Studio now offers features that allow you to:

Note: Dependencies coming FROM an Action Flow are not visible yet. However, dependencies TO an Action Flow can be seen.

To learn more, see: Asset-to-asset dependencies.

We've made usability improvements to our linked Views feature, making them easier to configure from table and button components. You can now create and configure a linked View from within your existing View, whereas previously your target View could only be an existing View within the same Studio package.

In this example, when clicking data in the SALES column the app will display the 'Sales Order' target View as a side panel. There is also an active mapping for the View variable "SalesOrder" from the target View. This variable will get updated with the "Order_ID" of the row in which the click in the table is done.

To learn how to configure linked Views, see: Linked Views.

With this private preview release, you can now configure an mTLS authentication method when connecting to your source system using a custom REST API extractor.

To configure mTLS authentication, you need to choose one of the supported mTLS formats:

Dependent of the chosen format, you need to provide the required certificates and/or passwords:

For more information about authentication methods, see: Extractor builder authentication methods.

And to learn more about our private preview releases, see: Feature release types.

We've added the ability to allowlist a domain. We originally prevented this due to security reasons, but we noticed that it resulted in customers having problems with adding intranet links.

To add an allow list, go to Admin section > Process Designer settings > Filter by “Link positive list” and add your configuration.

Other bugfixes in this release include:

With this private preview release, we've added a dedicated Data Labels panel that gives you flexible and granular control of data labels in charts. It also provides feature parity with the data label customization options available in Studio Analysis (maintenance mode only).

Previously, data labels could make charts in Views look cluttered and the data labels themselves could be difficult to read. With the flexibility and greater control introduced with the new Data Labels panel, you can now:

For detailed information about the settings available, see Chart data labels.

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