# Configure a Classification Scan (/docs/classification/scans/configure-classification-scans)

This guide describes how to configure discovery and classification scans, so you can create a living data sensitivity profile for your organization.

### 1. Choose a scan type [#1-choose-a-scan-type]

There are three types of scans available for Classification customers: Discovery Scans, Discovery + Classification Scans, and Classification Scans.

![View Scan Configuration and select a scan type](/images/sanity/2026-04-21-configure-scans-scan-type-options.png)

- **Discovery Scan:** scans only the schema of a data system without classifiying its contents as sensitive or not. Best for supporting DSRs without requiring classification of data.
- **Discovery + Classification Scan:** both scans the schema of a data system and classifies its sensitive contents. Recommended for any system that may contain sensitive data.
- **Classification Scan:** classifies sensitive contents of data system without updating information on a schema. Only recommended for one of re-classifications where schema and data content is stable.

<Callout type="info">
  **Note:** Classification-only scans cannot be performed until at least one Discovery or Discovery + Classification Scan has been performed.
</Callout>

For Classification customers, we recommend setting the scan type to **Discovery + Classification**, to give you the best view into your sensitive data.

### 2. Optional Configurations [#2-optional-configurations]

The optional configurations available will depend on the scan type and the integration with the third party.

Common options include:

- **Narrow scan scope:** exclude (or only scan) certain databases, tables, files, or schemas to reduce time and cost of scans.
- **Configure sampling:** enable non-null sampling, to ensure quality samples for classification. (Note: this generally has higher costs - reach out to your Customer Experience representative to learn more.)

![View Scan Configuration optional configurations](/images/sanity/2026-04-21-configure-scans-sampling-options.png)

<Callout type="info">
  **Note on Updating Scan Configuration**

  If a scan is currently In Progress or Paused, updates to the scan configuration will not take effect until a new scan starts. If you wish the updates to take effect immediately, you can **Restart** the scan from the beginning.
</Callout>

### 3. Run an Ad-Hoc Scan or set up Recurring Scans [#3-run-an-ad-hoc-scan-or-set-up-recurring-scans]

#### **Run an Ad-Hoc Scan** [#run-an-ad-hoc-scan]

Before setting up recurring scans, you may wish to run a test scan.

- Click **Run Ad-Hoc Scan** in the Most Recent Scan widget in the upper right corner of the Scan Configurations tab
- The status should change from Never Run to **In Progress**

![View Scan Configuration run ad-hoc scan](/images/sanity/2026-04-21-configure-scans-status-in-progress.png)

<Callout type="info">
  **Note:** if there is already a scan in progress, this option will not appear. Instead you may elect to "Restart" the current scan.
</Callout>

#### **Set Up Recurring Scans** [#set-up-recurring-scans]

Once you are satisfied with the results of your test scan, you can enable recurring scans to run automatically, enabling you to monitor data for sensitive content continually.

- On **Integration > Scan Configuration** enable the **Recurring Scans** toggle
- **Set a scanning schedule** - most customers select daily for frequently updated systems, or weekly for more static systems
- Click **Save**