Durable Workflow

A Durable Workflow is a long-running workflow artifact that can be interrupted by events, wait on timers, involve human tasks, and recover its state after a crash or restart — as opposed to a single request/response flow that runs start to finish in one execution. You wire the steps together in a visual flow, giving an explicit, predictable path — the best fit for known, fixed business logic. For logic that needs to adapt to each request at runtime instead, see Durable Agentic Workflow.

When to use it

Reach for a Durable Workflow instead of a regular integration flow when the process:

• Spans a long period of time (hours, days, or longer) rather than completing in one request.
• Needs to pause and wait for an external event, a timer, or a person to complete a step before continuing.
• Must survive a restart or crash partway through without losing its place.

In this section

• Create a workflow: Add the artifact, give it an input type, and design its steps on the diagram.
• Start a workflow: Launch a run from a service or an automation, and keep the ID that identifies it.
• Activities: The recorded units of work in a workflow — exactly-once on replay and retryable on failure.
• Durable timers: Pause for hours, days, or months with a wait that survives restarts and holds no threads or connections.
• Await data events: Wait until an external system or a person delivers the data the workflow needs, then resume with it.
• Send a data event: Deliver a value into a waiting run, using the workflow ID it was started with.
• Await human task: Pause for role-based human decisions and external data, for as long as it takes.
• Error handling and review activities: Approval gates before risky steps and human-reviewed retries after failures.
• Prebuilt activities: Durable REST, SOAP, and email calls with no wrapper to write.

What's next

• Build an Order Processing Workflow — a worked example that fills this in end to end.
• Durable Agentic Workflow — the same durable execution model, with an AI agent driving the workflow's decisions.
• AI Agents — the AI agent building block, usable inside a workflow.