| title | Configuration best practices |
|---|---|
| description | Review a list of current best practices and recommendations for the configuration metadata in Data API builder. |
| author | seesharprun |
| ms.author | sidandrews |
| ms.reviewer | jerrynixon |
| ms.service | data-api-builder |
| ms.topic | best-practice |
| ms.date | 06/11/2025 |
:::image type="complex" source="media/best-practices-configuration/map.svg" border="false" alt-text="Diagram of the current location ('Optimize') in the sequence of the deployment guide."::: Diagram of the sequence of the deployment guide including these locations, in order: Overview, Plan, Prepare, Publish, Monitor, and Optimization. The 'Optimize' location is currently highlighted. :::image-end:::
This article includes the current recommended best practices for configuration in the Data API builder. This article doesn't include an exhaustive list of everything you must configure for your Data API builder solution.
When adding an entity to the configuration file, use PascalCasing, so that the generated GraphQL types are easier to read. For example, if you have an entity named CompositeNameEntity the generated GraphQL schema has the following queries and mutations:
-
Queries
compositeNameEntitiescompositeNameEntity_by_pk
-
Mutations
createCompositeNameEntityupdateCompositeNameEntitydeleteCompositeNameEntity
-
If the entity maps to a stored procedure, the generated query or mutation would be named
executeCompositeNameEntity, which is easier and nicer to read.
When adding an entity to the configuration file, make sure to use the singular form for the name. Data API builder automatically generates the plural form whenever a collection of that entity is returned. You can also manually provide singular and plural forms, by manually adding them to the configuration file. For more information, see GraphQL configuration reference.
[!div class="nextstepaction"] Security best practices