Advocate Program Bulk Import Methods

Import jobs can be started from impact.com, via the Advocate SFTP integration, or by using the Advocate API directly.

File-based bulk imports in impact.com

Bulk imports can be started from within impact.com. To get started, select Reporting → Imports & Exports in the left navigation menu.

For full instructions on the file-based bulk import process, see our other guides:

• Import Participants in Bulk
• Delete Participants in Bulk

Bulk imports via SFTP integration

Bulk imports can be performed via Advocate SFTP integration. See our guide on the SFTP integration here for instructions on:

• Enabling and authenticating the integration
• Generating an SSH key
• Connecting to the SFTP server
• Uploading import files and checking their status

Bulk imports via API

To start a bulk import job, there are three API requests to be performed.

• Uploading the import file
• Validating the import file
• Starting the import job

Uploading the import file

The /export/upload endpoint accepts a file upload in two different ways, either with multipart/form-data encoding or as a raw file upload.

Example Request:

cURL

curl -X POST 'https://app.referralsaasquatch.com/api/v1/{tenant_alias}/export/upload' \
  -u :{tenant_api_key} \
  --form 'file=@my_file.csv'

When the file is successfully uploaded, a fileRef will be returned. The fileRef will be used in the next two API requests.

Example Response:

JSON

{
  "fileRef": "imports/test_akdq8a9wyvzba/userEvents_63323378e1edcd44b03eed9a.jsonl"
}

Validating the import file

Before starting your import job, you can use the validateJobInput GraphQL mutation to validate the import file before attempting to start the import job. There are two inputs required, the fileRef from step 1, and the job type. These are the job types available:

• Import Users: MUTATION/USER
• Delete Users: MUTATION/DELETE_USER
• Import User Events: MUTATION/USER_EVENT

Example:

cURL

curl -X POST 'https://app.referralsaasquatch.com/api/v1/{tenant_alias}/graphql' \
  -u :{tenant_api_key} \
  -H "Content-Type: application/json" \
  -d '{
    "operationName": "validate",
    "variables": {
      "jobInput": {
        "fileRef": "{file_ref}",
        "type": "{job_type}"
      }
    },
    "query": "query validate($jobInput: JobInput!) {validateJobCreation(jobInput: $jobInput) { errors }}"
  }'

If there are any errors found in the import file, they will be returned as an array in errors. If errors is empty, then the same fileRef can be used to start an import job in step 3.

Example Response:

JSON

{
  "data": {
    "validateJobCreation": {
      "errors": []
    }
  }
}

Starting the import job

To start the import job, use the createJob GraphQL mutation. Provide the fileRef and job type from the upload step.

Example:

cURL

curl -X POST 'https://app.referralsaasquatch.com/api/v1/{tenant_alias}/graphql' \
  -u :{tenant_api_key} \
  -H "Content-Type: application/json" \
  -d '{
    "variables": {
      "jobInput": {
        "type": "{job_type}",
        "outputFormat": "CSV",
        "fileRef": "{file_ref}"
      }
    },
    "query": "mutation ($jobInput: JobInput!) { createJob(jobInput: $jobInput) {id type requester dateCreated}}"
  }'

The id returned by the mutation can be used to query the status of the job.

Example Response:

JSON

{
  "data": {
    "createJob": {
      "id": "633b1cf34efc053cb50a3f6d",
      "type": "MUTATION/USER_EVENT",
      "requester": "API",
      "dateCreated": 1664818419661
    }
  }
}

Check the status of the import job

To check the status of the import job, you can use the job GraphQL query and provide the job id returned in step 3.

Example:

cURL

curl -X POST 'https://app.referralsaasquatch.com/api/v1/{tenant_alias}/graphql' \
  -u :{tenant_api_key} \
  -H "Content-Type: application/json" \
  -d '{
    "variables": {
      "id": "{job_id}"
    },
    "query": "query ($id: ID!) { job(id: $id) { status stats { recordsProcessed }}}"
  }'

Example Response:

JSON

{
  "data": {
    "job": {
      "status": "COMPLETED",
      "stats": {
        "recordsProcessed": 22
      }
    }
  }
}