The Google BigQuery integration allows you to automate the export of your Dialog Insight project to your Google Cloud project or the import of your GBQ tables toward your DI project. This makes the transfer easier for your complementary and behavioral data generated by DI. This way, you can easily exploit and analyze this data in your GBQ warehouse.
Access path: Project → Data Management → Integrations |
Prerequisites
- Create a Google Cloud account and a Google Cloud project
- Create a Google Cloud service account (dedicated to a connection with DI). It is recommended to create a custom role with the following permissions :
- bigquery.datasets.create (export)
- bigquery.jobs.create (export)
- bigquery.datasets.get (import)
- bigquery.tables.list (import)
- bigquery.tables.get (import)
- bigquery.tables.getData (import)
Step 1: Connect the GBQ Project
First, find your Google Cloud project ID:
In DI, follow the access path and next to Google BigQuery, click Add:
Enter your Google Cloud project ID:
Upload the JSON file (exported from GBQ) that contains the account service keys:
You could then add another Google Cloud project as a source (optional) by repeating step 1:
Étape 2: Configuring the Export
For the export, you can use a log export or an automated export (or both).
Be careful not to create a loop between your data sources. For example, if you configure an import from a GBQ table to a DI project, do not configure an export from this project to the same table in GBQ. This integration is not designed to support bidirectional synchronization.
Option 1: Log Export
Using log export will export all the tables from your project. When you export logs, a (non-editable) dataset is created in GBQ.
You must choose Google Big Query as the destination:
If you want to export all data starting from the creation of your DI project, check Export all data on the next run*:
Otherwise, you can specify the period in days to include in the next export (90 days as default).
*Remember to uncheck the option after the first export. Otherwise, every export will send the entirety of the data, which will impact the performance.
When you are done configuring a log export, remember to click Save at the bottom of the page.
Then, it is recommended to manually launch the first export so it is processed now (you must check Export all data on the next run). Click Activate:
Go to the History tab, select a task and click Launch now:
→ See the complete procedure for log export
Option 2: Automated Export
It's possible to configure several automated exports (unlike a log export, there is no limit). An automated export allows the export of only one object (a table) at a time. You must configure several exports if you want to export several objects.
Select Google Big Query as the destination:
When you are done configuring an automated export, remember to click Save at the bottom of the page.
After, it's recommended to manually launch the first export so it is processed now.
→ See the complete procedure for automated export
Select the GBQ project and the table to import (e.g. your contact or your product table). You must then map the fields, which means linking the fields from the source (GBQ) to their corresponding fields in the destination (DI).
After, it's recommended to manually launch the first import so it is processed now:
