Skip to main content

Sharing Data

Datafi makes it straightforward to grant access to your datasets, whether you are sharing with a colleague down the hall or a partner at another organization. All sharing is governed by roles, so you stay in control of what each person can see and do.

Data Owner vs Data User​

Every person with access to a dataset holds one of two roles:

Data Owner​

The Data Owner role grants full administrative authorization over the dataset. Data Owners can:

  • View and query all tables and columns in the dataset.
  • Invite or remove other users.
  • Assign and change roles for other users.
  • Configure row-level and column-level policies.
  • Edit dataset metadata (name, description, tags, region).
  • Delete the dataset entirely.

A dataset can have multiple Data Owners. The person who originally connects or uploads the dataset is automatically assigned the Data Owner role.

Data User​

The Data User role provides limited, read-oriented access. Data Users can:

  • View and query tables and columns they have been granted access to.
  • Browse the schema within their permitted scope.
  • Use the dataset in Data Views, Data Apps, and chat queries.

Data Users cannot modify sharing settings, edit metadata, change policies, or delete the dataset.

When to Use Each Role

Assign the Data Owner role to team leads, data engineers, or administrators who need to manage the dataset on an ongoing basis. Assign the Data User role to analysts, business users, and anyone who needs to consume the data without managing it.

Internal Sharing via Email​

To share a dataset with someone inside your organization:

  1. Open the dataset details page in the Data Catalog.
  2. Click Share or navigate to the Access tab.
  3. Enter the person's email address.
  4. Select the role to assign -- Data Owner or Data User.
  5. Click Invite.

The recipient receives an email notification with a link to the dataset. Once they accept, the dataset appears in their Data Catalog.

You can invite multiple people at once by entering several email addresses separated by commas.

Existing Accounts Required

The recipient must have a Datafi account associated with the email address you enter. If they do not have an account yet, ask them to register first, then send the invitation.

External Partner Sharing​

Datafi supports sharing datasets with people outside your organization -- vendors, clients, consultants, or partner companies. The process is the same as internal sharing:

  1. Open the dataset details page.
  2. Click Share.
  3. Enter the external partner's email address.
  4. Assign a role.
  5. Click Invite.

The external partner receives an email invitation and gains access to the specific dataset you shared. They do not gain access to any other datasets in your workspace unless you explicitly share those as well.

External Sharing Considerations

When you share data externally, the partner can query the dataset within the bounds of the role and policies you define. Review your row-level and column-level policies before sharing to ensure that sensitive data is appropriately masked or restricted.

User Groups​

For organizations with many users, managing access one person at a time becomes impractical. Datafi supports user groups to simplify this.

A user group is a named collection of users. When you share a dataset with a group, every member of that group receives access at the role you specify. Changes to group membership automatically propagate -- if you add someone to the group, they gain access; if you remove them, they lose it.

To share with a group:

  1. Open the dataset details page.
  2. Click Share.
  3. Search for and select the group name instead of an individual email address.
  4. Assign a role.
  5. Click Invite.
Managing Groups

User groups are created and managed in the Administration section of Datafi. Work with your workspace administrator to define groups that align with your teams, departments, or project structures.

Modifying and Revoking Access​

Data Owners can change a user's role or remove their access at any time:

  • Change role -- Open the Access tab on the dataset details page, find the user, and switch their role between Data Owner and Data User.
  • Revoke access -- Click Remove next to the user's name. They immediately lose access to the dataset.

Revoking access does not delete any queries, Data Views, or Data Apps the user previously created against the dataset, but those assets will no longer return results.

Next Steps​