Configure Synthetic Data settings

Synthetic Data settings define the global defaults and AI-assisted discovery behavior used across all applications, datasets, generators, and jobs.

Settings are configured in two ways:

  • UI — Default Locale and Default Industry are configured from the Synthetic Data settings dialogue in the Synthetic Data home page.

  • API only — Structure Selection Threshold and Deep Search Threshold are configured through the DCT API.

Configure UI settings

To configure Default Locale or Default Industry:

  1. From the top navigation bar, click Synthetic.

  2. In the left navigation pane, click Settings.

  3. Update the Default Locale or Default Industry field as required.

  4. Click Save to apply the changes.

For Multi Data generators, some methods support a per-field Locale on the Configuration page. That per-field locale controls generated output for that field. Default Locale and Default Industry in Synthetic Data settings are global defaults that influence generator assignment and default behavior.

Settings reference

Setting Description Default value Configured via
Default Locale The global default locale for generator behavior. Influences generator assignment eligibility and default generation behavior. For Multi Data methods that support per-field locale, the method-level Locale value controls output for that field. en_US UI
Default Industry The global default industry for generator behavior. Helps align generator assignment to domain context. Generators with an industry value must match this setting to be eligible; generators without an industry value remain eligible for all industries. Technology UI
Structure Selection Threshold The minimum confidence score required for AI-assisted structure selection. When using Describe your test data, only structures that meet or exceed this threshold are included in AI-assisted recommendations. 60% API only
Deep Search Threshold The minimum confidence score (0–100%) required for structures identified through Deep Search. Only applied during dataset creation/pull. Only applies to built-in generators; custom generators use ASDD only. 70% API only