> For the complete documentation index, see [llms.txt](https://docs.decube.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.decube.io/data-quality-v2/monitors/create-a-monitor.md).

# Create a monitor

Create a monitor through the three-step wizard, and change one you already have.

You create monitors from the **Monitors** tab. The wizard splits into three steps, grouped by how permanent each decision is: what the monitor watches, how it evaluates, and how it is described.

{% embed url="<https://youtu.be/HgE_39ujuC8>" %}

### Step 1: Asset and monitor type

Pick the asset you want to add the monitor to. You can use the filters to narrow down the selections.

Each monitor type applies to a specific kind of asset: Freshness and Volume run on a table or view, Field Health runs on a column of a supported type, and Custom SQL runs on a data source. Virtual tables and virtual columns are not applicable assets. See [Applicable assets](/data-quality-v2/monitor-types.md#applicable-assets) for the full list.

<figure><img src="https://1779874722-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FTw0qpCVzfrIXqS4FEg4T%2Fuploads%2FomlwXXU0EXt2Ca9ML0rh%2FSCR-20260921-oseq.png?alt=media&amp;token=a7aa1b1b-702e-4e4b-a73e-1e7a4cb3e7cf" alt=""><figcaption></figcaption></figure>

{% hint style="warning" %}
The asset and the monitor type are fixed once the monitor exists. To watch a different table, or run a different check, create a new monitor.
{% endhint %}

### Step 2: Configuration

This is where you set how the monitor runs and what counts as a failure. The fields you see depend on the monitor type, drawn from the same set each time:

* **Schedule**, how often the monitor runs, or **Manual** if you trigger it yourself.
* **Detection method**, how Decube measures the asset.
* **Threshold**, the boundary that separates a passing value from a failing one.
* **Tolerance**, how many failing rows you accept before the run fails. Row-level checks use this.

Every field is explained once in [Building Blocks](/data-quality-v2/monitors/building-blocks.md), and each monitor type page shows the fields for that type with a worked example.

<figure><img src="https://1779874722-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FTw0qpCVzfrIXqS4FEg4T%2Fuploads%2FpNbyV7wii5p2zFdy0Lel%2FSCR-20260921-otjh.png?alt=media&amp;token=93b68eaa-ac7c-4748-9fcd-074c7c0dd64e" alt=""><figcaption></figcaption></figure>

### Step 3: Properties

Name the monitor, describe what it protects, and set the properties your team filters and reports on:

* **Data quality dimension**, which determines where this monitor contributes to your DQ Scorecard.
* **Custom attributes**, your own tags and key-value pairs (severity, domain, owning team). Alert rules read these, so attributes you set here decide who gets notified.

Attributes are defined for your organisation under **Org settings > Custom attributes**, with **Monitors** as the applicable object.

<figure><img src="https://1779874722-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FTw0qpCVzfrIXqS4FEg4T%2Fuploads%2FxvgePqg5gun2Y0jwpK9R%2FSCR-20260921-otrg.png?alt=media&amp;token=f796c4c2-c6c8-431a-960c-ae743b9033e8" alt=""><figcaption></figcaption></figure>

### After you save

The monitor becomes Active and runs at its next scheduled time. Open it from the Monitors tab to see run history, the cases it raised, and the alert rules that currently match it.

### Change an existing monitor

Open the monitor and use **Monitor actions > Edit**. You land on the step that holds the field you want, and the other steps stay as they are.

Some changes alter what the monitor measures, so its past runs no longer compare against the new configuration. Decube warns you at the point of the change and creates a new monitor version, keeping earlier runs against the version that produced them.

{% hint style="success" %}
Changing the same field across many monitors? Use bulk update instead of editing them one at a time.
{% endhint %}