Deploy at scale

Set up a Data Standard with a scope to enforce it across many datasets at once.

Writing a Data Standard defines a rule. Deploying it is what makes that rule enforceable across your data landscape. When you create a Data Standard in Soda Cloud, you give it a scope, and Soda automatically links the standard to every dataset the scope selects. Then, Soda keeps that set of datasets up to date as your data landscape changes.

That is the core value of Data Standards: you maintain one rule, and it applies to every dataset that should follow it, including datasets onboarded after the standard was created.

Creating and managing Data Standards requires the Manage data standards permission. Data Standards must also be enabled for your organization.

Elements of a Data Standard

A Data Standard is made up of the following:

Field	Description
Name	A unique name that identifies the standard within your organization.
Label	A human-readable display name shown in Soda Cloud.
Description	Optional context explaining what the standard enforces and why.
Contents	The Data Standard YAML, including the checks to run.
Scope	The rule that selects which datasets the standard applies to (see below).
Status	The lifecycle state: draft, active, or paused.
Owners	The users or groups responsible for the standard. At least one owner is required.

Scope: selecting datasets

The scope determines which datasets a standard is enforced on. You define it in the editor using one of two modes:

• Rules – match datasets by pattern. Each rule targets a data source, with an optional schema pattern and dataset pattern. Patterns support wildcards (for example, customer_data_*), so a single rule can match many datasets, and you can add several rules to cover datasets across different data sources or schemas.
• Datasets – select specific datasets explicitly.

Defining scope with rules: each rule matches datasets by data source, schema pattern, and dataset pattern.

Because rules match by pattern rather than a fixed list, Soda re-evaluates the scope against your catalog and links the standard to every matching dataset. You can re-run this at any time with Refresh scope to confirm exactly which datasets the standard targets.

Scope is evaluated continuously, not just once. When a new dataset is onboarded that matches an active standard's scope, the standard is automatically linked to it, no manual step required. This is what keeps a standard enforced "at scale" as your data grows.

Status and lifecycle

A standard's status controls whether and how it is enforced:

1

Draft

The standard is saved but not enforced. Use this state to author and review a standard before it affects any data. No datasets are linked and no scans run.

2

Active

The standard is enforced. Soda links it to every dataset matching its scope and runs its checks on the configured schedule. New matching datasets are picked up automatically.

3

Paused

Enforcement is temporarily suspended. Existing dataset links are preserved, but no new scans run until you set the standard back to "active". Use this to pause a noisy or in-progress standard without losing its configuration.

Switch a standard's status from its page in Soda Cloud.

Ownership

Every Data Standard has one or more owners (users or user groups). Owners are responsible for the standard and are who Soda surfaces when a standard needs attention. Assign owners that reflect accountability for the rule, typically a data governance or platform team.

Scheduling

Once a standard is active, Soda keeps validating every linked dataset over time without manual intervention. A standard does not carry its own cron expression; instead, its checks run per dataset, on that dataset's Data Standards schedule.

• Default from the data source. Each dataset inherits its Data Standards schedule from the default schedule set on the data source connection page, the same default used for contract verification. Set it once per data source and every dataset it contains picks it up.
• Per-dataset override. A dataset owner can change the schedule, or turn scheduled verification off entirely, from the dataset's settings dialog in Soda Cloud.

This is why the same standard can run at different times against different datasets: each linked dataset runs on its own Data Standards schedule, defaulting to its data source's default.

Learn how the data source default schedule works: Scheduled execution (from the Contract Editor).

How results stay organized

Because a standard runs across many datasets, Soda tracks its results per dataset while keeping them grouped under the standard. This lets you answer both questions that matter at scale:

• Is this dataset compliant with all the standards that apply to it?
• Is this standard being met across every dataset it targets?

See Review results.

Next steps

• Run a standard from the CLI: Verify a Data Standard
• Review results across datasets: Review results

---

You are not logged in to Soda and are viewing the default public documentation. Learn more about Licensing & documentation access.

If you do have a Soda license, make sure to log in to Soda Cloud in this same browser.