All Collections
Admin Dashboard
Data Sync
Mapping Your Incoming Data
Mapping Your Incoming Data

Translating people data from your source to Sift

Ryan Bickham avatar
Written by Ryan Bickham
Updated over a week ago

After setting up a Source, regardless of the sync method you choose, you will need to set up mappings. A mapping is a translation from a property on a person in your data source, to an attribute on a person in Sift. For example, in a CSV, this source property is defined by a header in your CSV file, where in something like an Azure AD source, it would be defined by the properties on the API response.

Watch this video going over how to import and map incoming data via a CSV.

Note: Any properties from your data source that are not mapped will not be consumed by Sift. We only consume what you elect to give us!

Use the links below to view examples of source properties for:

Step 1: Assign your Unique Identifier mapping

Sift requires mapping a Unique Identifier for all sources. The Unique Identifier defines a person’s identity in Sift.

We strongly recommend using something that will never change for a person, such as an employee ID. The simplest thing to use is email address. However in this case, if someone's email address were to change in your data source, it would cause thei Profile to be recreated in Sift, rather than updated. Any record from your data source with a missing or non-unique value will not be imported.

To set the Unique Identifier mapping, choose the source property (ex: CSV heading or API property) from the dropdown on the left-hand side. You should see an example of the value under the "Preview" column. The Unique Identifier mapping cannot be changed after your first sync without contacting our support team.

Step 2: Assign hierarchy mapping (watch video)

In order to use features in Sift like our Org Chart or parts of Profile pages, you'll need to assign the Org Chart mapping. This is used to build your organization structure by letting us know which attribute defines a person's direct manager.

For CSV sources:

In the first dropdown, select which CSV header in your file represents a person's direct manager (we recommend something like a manager's email address, or a manager's id)

In most cases, this should be all you need to do. However, if we still aren't able to properly associate people to their managers, we may need some further information.

If you see a second dropdown, or are in a non-CSV source:

We were not able to automatically detect how to match the first column you selected to another row in your file. In the second dropdown, select which property the above property points to on another person in your data source (Employee Header).

For example, you may be importing a CSV that includes the following columns:

  • "teamLeaderEmail": Referencing the person's manager by their email address

  • "email": The person's email address

In this case, you would choose "teamLeaderEmail" in the first dropdown, and "email" in the second dropdown, since that's what the "teamLeaderEmail" property needs to look at to match the direct report to their manager.

Note: For non-CSV source types, we provide a default mapping here according to the source type's official API schema. You should never need to change this unless you represent your hierarchy with alternate properties.

Step 3: Assign Profile attribute mappings

Use the Profile Attributes section to provide Sift with other pieces of data that you'd like to include in your Profile pages, Search, and Org Chart. Primary Sources require you to provide mappings to the first name, last name, and email properties in Sift, and each incoming person must have all three of these pieces of information. Providing other pieces of data to Sift is optional, but encouraged!

On the left-hand side, Sift will pre-populate the area with recommended properties from your data source (ex: CSV headers or API properties). Use the dropdown menu on the right-hand side of the mapping to choose which Sift attribute to map to. If you do not see an attribute here that represents the data you're attempting to map, you may need to create it.

Referring to Your Mappings Summary

On the right-hand side of the Mappings page, you will see a "Mappings Summary" panel that will scan your mappings for any missed or mis-mapped fields. You will see indicators throughout the mapping experience to inform you of your progress as you map.

Check mark indicators mean that your mapping is accurate and complete. A caution indicator means that there is an issue with a mapping but it is not severe. An error indicator means that you have a major issue with your mapping that needs to be addressed before you can proceed.

Adding Additional Mappings

If there is a property in your data source that you do not see on the left-hand side, click the "Add Mapping" button on the top right. When you do, Sift will add an additional row on the bottom that will let you select from a dropdown of source properties, similar to the primary and hierarchy mapping areas.

If you still do not see the source property you'd like to map in the dropdown, click the "Use Another Property..." button. If there’s a property coming for your source that is not present in the provided dropdown, you can add it manually. For example, this could happen if you have a CSV header that will be added to your file later, or if your objects coming from Azure AD have dynamic custom properties.

Note: For API-based sources, Sift supports dot notation for inner properties (ex: name.givenName)

Note on Google Workspace Collection Properties:

The Google Workspace API will return some properties like phone numbers and addresses in an array. To support retrieving these types of attributes, you'll need to provide us with some additional context about the type of item you're retrieving. Start providing one of these items by typing the name of the collection (e.g. phones) into the left input, and the system will ask for some further information.

Example: mapping a mobile phone number from Google Workspace

Where do I go next?

After saving your mappings and setting up your integration, you should be ready to sync your data. You may always come back and adjust mappings as needed.

Did this answer your question?