Role based access

Last updated:

PostHog gives you tools for feature access control and team member groupings with role based access inside of PostHog.

Organizational defaults

By default, team members in your organization have view and edit access to all resource features (feature flags, experiments, session recordings).

These can be changed under organization settings -> role based access tab. You can set organizational defaults to just "view only" and restrict edit access to certain features.

Roles

Roles are groupings of team members. You can assign a member to multiple roles. For more granularity, you can also define whether a role has view or edit access per resource.

Only organization members with admin+ access level can create or edit roles.

Resources

Resources are PostHog features. These include but are not limited to feature flags, experiments, session recordings, and cohorts.

Resources that you create will always be viewable and editable. Access control restricts editing for resources that you don't own.

Custom edit access per resource

For even more granularity, you may set only certain individual resources to be editable by certain roles. You can set these on the resource itself.

For example, say I have a feature flag "top-secret" that I only want the "product" team role to have edit access for. By default the product team role only has view access to feature flags.

In this case, I'll set the organization default to "view only", and then add the product team role to top-secret flag, which ensures they can edit the flag, despite having only view access as the role default for all feature flags.

Note: Roles based access is an enterprise offering. To use this feature, the enterprise package should be added on..

Questions?

Was this page useful?

Next article

Apps

Apps extend PostHog's functionality by either pulling data into or sending data out of PostHog. They allow anyone to extend and customize PostHog in order to better fit their needs. We have a comprehensive library of apps , including data warehouses (Snowflake, BigQuery, Redshift) and marketing tools (HubSpot, Sendgrid, Customer.io, Salesforce). Apps can be used for a wide variety of use cases, such as: Sending the event data to a data warehouse. If you have a data lake or data warehouse…

Read next article