Skip to main content

About

Beyond manual execution, Interactly workflows can run automatically — either at a specific scheduled time or in response to an external event. Scheduled workflows fire once at a configured future datetime without any human intervention. Triggered workflows respond to events from external systems, enabling real-time integration with your existing infrastructure.

Execution Modes

Manual Execution

Run a workflow on demand from the dashboard or via the Execute API. Useful for testing, one-off tasks, and operator-initiated processes.

Scheduled Execution

Run a workflow once at a specific future date and time. No human trigger required — the platform fires the run automatically at the configured time.

Triggered Execution

Run a workflow in response to an external event — an SQS message or an inbound webhook — enabling real-time, event-driven automation.

Scheduled Workflows

Scheduled workflows execute automatically at a specific future datetime. Once a schedule is created, the platform fires the workflow run at the configured time via AWS EventBridge — no manual action required.
Each schedule triggers the workflow once at the specified datetime. To run a workflow repeatedly, create multiple schedules with different datetimes.

Setting Up a Schedule

Schedules are created via the API. Use the POST /workflows/{workflow_id}/schedules endpoint with the following fields:
  • scheduled_time — the exact datetime (UTC) when the workflow should run; must be in the future
  • run_input — optional input variables to inject into the workflow execution
  • version — optional version number to run; defaults to the active version if not specified
  • scheduled_by_name — optional display name of the person creating the schedule
scheduled_time must be provided in UTC. The schedule fires exactly once at the specified time.

Schedule Status

The schedule has been created and is waiting for its trigger time. It can still be modified or cancelled at this stage.
The scheduled time has arrived and the workflow run has been initiated.
The workflow run triggered by this schedule has finished successfully.
The workflow run triggered by this schedule encountered an error. Check the run details for diagnostics.
The schedule was cancelled before its trigger time. No workflow run was initiated.

Managing Schedules

Only Pending schedules can be modified. You can update the scheduled time, input variables, or target version before the trigger fires.
Only Pending schedules can be cancelled. Once cancelled, the schedule will not trigger a workflow run. Deletion is permanent.

Triggered Workflows

Triggered workflows execute in response to an external event. Interactly supports two trigger sources: AWS SQS messages and inbound webhooks.

SQS Trigger

The workflow service listens to an AWS SQS queue. When a message arrives, a workflow run is initiated with the message payload as the input. Ideal for integrating with internal event-driven architectures.

Inbound Webhook Trigger

An external system calls a webhook endpoint on the Interactly platform. The request payload is passed directly to the workflow as input. Ideal for integrating with third-party platforms and CRMs.
Triggered workflow configuration — queue bindings and webhook endpoint setup — is managed at the platform level. Contact your Interactly administrator to configure trigger sources for your team.

Concurrency & Execution Controls

All workflow execution modes share the same concurrency model. Understanding these limits is important when configuring schedules or triggered workflows that may fire frequently.
Your organisation has a maximum number of workflow runs that can execute simultaneously across all workflows. This limit is set by Interactly and visible in My Organisation → License.When the team limit is reached, new execution requests are queued until a slot becomes available.
You can set an additional cap on a specific workflow — for example, to ensure a particular workflow never consumes more than a fixed number of concurrent slots regardless of how frequently it is triggered.Configure this under the workflow’s Settings → Concurrency. Setting it to 0 removes the per-workflow cap and defers to the team limit.
The platform automatically detects and cleans up workflow runs that have been in progress for longer than expected — typically caused by a node hanging, a tool call timing out, or an unexpected platform interruption.Stale runs are marked as Failed with a timeout reason, freeing up concurrency slots for new executions. This runs automatically in the background and requires no configuration.
If you have workflows that need to run repeatedly at regular intervals, create multiple schedules in advance — one for each intended run time.

Use Cases

Daily Patient Follow-Ups

Schedule a workflow to run at a specific time to follow up with patients after their appointment.

CRM Event Processing

Trigger a workflow via SQS whenever a new lead is created in your CRM to initiate an automated intake process.

Appointment Reminders

Schedule a reminder workflow to run 24 hours before a specific appointment by setting the trigger time accordingly.

Real-Time Inbound Routing

Trigger a workflow via inbound webhook when a new support ticket arrives, automatically qualifying and routing it based on the ticket content.

API Reference

Create Schedule

POST /workflows/:id/schedules Schedule a one-time workflow run at a specific future datetime

List Schedules

GET /workflow-schedules Retrieve all schedules configured for your team

Execute Workflow

POST /workflows/:id/execute Trigger a workflow run manually via API

List Workflow Runs

GET /workflow-runs View the history of all runs across execution modes

Next Steps

Workflow Versioning

Pin schedules to a specific version for stable automated execution

Workflow Webhooks

Subscribe to run completion events from scheduled and triggered executions

Simulation

Test the workflow before scheduling a live run

Building a Workflow

Go back to the workflow builder guide