Triggers
Power Automate defines a trigger as an event that starts a cloud flow. These events can be initiated by a user, are scheduled, or are generated by a connector, including custom connectors.
The triggers that are used in Power Automate are defined by connectors. Each connector has its own set of triggers. Connectors are divided into two groups:
- Tabular - These connectors are for data sources where the data is stored in tables. Dataverse is a tabular connector. Tabular connections will have triggers around creating, updating, and deleting of records.
- Function-based - These triggers are for cloud services such as Twitter or Microsoft Outlook. The triggers are related to functions in those services. For instance, the Outlook connector has a trigger for when an email is received.
Trigger types
Three types of triggers for initiating Power Automate cloud flows are:
- Automated - A flow that is started by a defined event in a connector, such as a record being created or a file being added to a file location.
- Instant - A flow that is run manually by a user, such as a button being pressed.
- Scheduled - A flow that is run on a recurring basis, such as at 9:00 AM every workday, or every hour.
This section will focus on using the triggers for Dataverse. Automated flows can be triggered from data events in the platform's event framework. The Dataverse connector has a single automated trigger when a record is created, updated, or deleted.
In the trigger step, you have to specify the trigger condition:
- Added
- Added or deleted
- Added or modified
- Added or modified or deleted
- Deleted
- Modified
- Modified or deleted
The Dataverse connector has a single instant trigger when a flow step is run from a business process flow. This feature allows a Power Automate cloud flow to be manually initiated by a user as a step in a business process flow.
The Dataverse connector has a single instant trigger when a row is selected. This feature allows a Power Automate cloud flow to be manually initiated by a user from within a form for a row in a model-driven app.
For example, a common pattern that is used with Power Automate cloud flows is to use the Scheduled trigger to run feature to retrieve all rows that are due for the day and then loop through and process them on a daily basis.
Poll vs. push triggers
Triggers represent notifications to Power Automate cloud flows that an event has occurred. Triggers are grouped into two types: poll and push. A polling trigger makes a call into the API at a reoccurring frequency to check for new messages. When new data is available, the trigger will run the flow. Examples of poll triggers include timer triggers. A push trigger responds to a push of new data from the service.
When you are using triggers, indication of whether a trigger is poll or push will be invisible to you. The deprecated Dynamics 365 connector used polling triggers; the more modern Dataverse connector uses push triggers. Push triggers are more efficient and are more real-time.
Not every connector has a trigger. In such cases, you can use a scheduled trigger and then retrieve the data that has changed since the previous run. In other words, use a polling pattern. If you use this pattern, you need to ensure that you don't run the flow too often, and you should leave several minutes between each iteration.
Use filters
Solution architects should consider minimizing the number of flow runs for each cloud flow. A common mistake that many people make with Power Automate when creating and updating triggers is to retrieve the row and then have a condition that checks if the flow needs to do anything. With Dataverse connectors, you do not need to retrieve the row; the trigger step contains the new/changed data.
Additionally, like other triggers, Dataverse connectors can have a filter applied to the trigger. A filter will prevent the flow run from being implemented, reducing unnecessary implementations of the flow.
In the previous diagram, the When a row is added, modified or deleted trigger for the Dataverse trigger has two filters that can be used:
- Column filter - If columns are specified, the flow will only run if any of the listed columns are modified.
- Row filter - If an OData query is specified, the filter expressions determine which rows can trigger the flow.
In Dataverse, if a column is included in a trigger's output, it doesn't mean that the value of the column has changed. If you need to ensure that you only process changed values, you need to use a plug-in with Pre and Post-Images.
Note
Instant triggers are not currently visible from apps if the flow is in a solution.