> For the complete documentation index, see [llms.txt](https://handbook.n8n.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://handbook.n8n.io/start-here/what-is-automation.md).

# What is automation

## Our definition of automation

In this handbook, we define automation as **a system that performs or coordinates work with less manual intervention**. It connects apps, services and data so that repetitive work can run consistently. An automation may use deterministic workflow logic, AI models or agents, human approval steps, or a combination of these.

### Automation is not the same as a workflow

![](https://1295977995-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FikYvt5xWzvK7NtuHBUPD%2Fuploads%2Fgit-blob-5993f8944ca60f58607ab49089d6a6d39c1d1936%2Fwhat-is-automation-editorial.png?alt=media)

An automation describes the whole system. A workflow is one possible component used to implement it.

An automation may combine:

* Workflows that connect systems and move data
* Code that applies specialized logic
* AI agents that classify, extract, or generate information
* Human steps that provide judgment, approval, or oversight

Thinking about the whole automation helps us account for the process, people, and responsibilities around the technical implementation.

### Automation can combine rules, AI, and human judgment

Different processes require different levels of predictability.

* **Deterministic steps** follow explicit, fixed rules and should produce the same output for the same input. Examples include formatting a date, validating an email address, checking a number range, or routing based on a status field. They are generally fast, inexpensive, and straightforward to test and debug.
* **Non-deterministic steps** use AI models or agents to handle ambiguity, interpretation, or generation. Examples include summarizing a document, classifying intent, extracting structured data from unstructured text, or generating a contextual reply. Their outputs can vary with context, phrasing, and model behavior, making them flexible but less predictable and often slower or more expensive.

Reliable production workflows often **combine both types**, using each where it is strongest. For example, a support-ticket automation might:

1. Receive a new ticket through a trigger.
2. Use an AI step to classify its urgency and category.
3. Use deterministic Switch logic to route it to the correct team.

It also makes debugging easier: rule-based failures can be inspected directly, while AI-related failures can be addressed through prompt, model, context, or evaluation changes.

## Automation includes the surrounding system

A production automation is more than its nodes or code. It also includes:

* The business process
* The data sources and credentials
* The people who use, own, or maintain it
* Documentation
* Monitoring and error handling
* The process for changing or retiring it

This broader definition matters because a lot of the failures happen at the boundaries. A technically correct workflow can still fail if the process changes, an owner leaves, a source system changes, or nobody acts on an alert. More on this in [How we build](/how-we-build/how-we-build.md).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://handbook.n8n.io/start-here/what-is-automation.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.