BigQuery

Overview

Hightouch lets you pull data stored in your Google BigQuery warehouse and push it to downstream destinations.

Connecting Hightouch to BigQuery requires some setup in both platforms. It's recommended to set up a service account with the correct permissions in BigQuery before configuring the connection in Hightouch.

BigQuery credential setup

Setup in BigQuery has three main steps:

1. Enable BigQuery for your Google Cloud project
2. Create a service account
3. Grant the Hightouch service account access to your project

Create a project and enable BigQuery API

1. Login to the Google Developers Console.
2. Configure the Cloud Platform:
   • If you don't have a project already, create one.
   • Once you have a project, enable the BigQuery API for it.
3. Copy your Project ID for later use.
4. Find the location of your BigQuery dataset or sets. You can find this by querying the INFORMATION_SCHEMA.SCHEMATA view or by visiting the Google Cloud web console and clicking on a BigQuery dataset in the Explorer panel. You need both Project ID and Data location when connecting Hightouch to BigQuery.

Create a service account

To create a service account, follow the setup instructions in our Google Cloud Provider (GCP) documentation.

Grant access

By default, your GCP service account doesn't have permission to read data from BigQuery. You can set up your service account to have full access to your project using a predefined role. Otherwise, you can create a custom role and provide limited access according to a user-specified list of permissions.

Grant full access

You can grant full access by assigning the bigquery.user and bigquery.dataViewer roles to your service account. You can do this in the Google Cloud web console or by running these snippets in the Cloud Shell.

Grant permission to read metadata and list tables:

gcloud projects add-iam-policy-binding <YOUR_PROJECT_NAME> \
--member serviceAccount:<YOUR_SERVICE_ACCOUNT> \
--role roles/bigquery.user

Grant permission to read data from tables and views:

gcloud projects add-iam-policy-binding <YOUR_PROJECT_NAME> \
--member serviceAccount:<YOUR_SERVICE_ACCOUNT> \
--role roles/bigquery.dataViewer

Grant limited access

If you don't want to grant full access to your BigQuery service account, you can opt to grant limited access instead. You can do this by assigning the bigquery.dataViewer role only to the specific datasets, tables, or views you want to use in Hightouch.

Since you are assigning the bigquery.dataViewer role only to specific resources, you need to assign the bigquery.user role and grant the bigquery.tables.get permission at the project level. For this, you can create a custom role in the Google Cloud web console based on an existing predefined role (bigquery.user), which you can name custom.bigquery.user. When setting up the custom role, click Add permissions to add the bigquery.tables.get permission to this custom role.

Then assign this role to your service account at the project level. You can do this in the Google Cloud web console or by running this snippet in the Cloud Shell:

gcloud projects add-iam-policy-binding <YOUR_PROJECT_NAME> \
--member serviceAccount:<YOUR_SERVICE_ACCOUNT> \
--role roles/custom.bigquery.user

You can then decide which datasets, tables, or views your GCP service account has access to by granting access to a resource in the Google Cloud web console. For every resource you would like to use in Hightouch, select your BigQuery service account as the Principal and the bigquery.dataViewer role as the Role.

Connection configuration

To get started, go to the Sources overview page and click the Add source button. Select BigQuery and follow the steps below.

Configure your service account

Select the GCP credentials you previously created or click Create new. To learn more about these credentials, see the Google Cloud Provider (GCP) documentation.

Configure your source

Enter the Project ID for the project you enabled the BigQuery API for and the Dataset location.

Choose your sync engine

For optimal performance, Hightouch tracks incremental changes in your data model—such as added, changed, or removed rows—and only syncs those records. You can choose between two different sync engines for this work.

The Basic engine requires read-only access to BigQuery. Hightouch executes a query in your database, reads all query results, and then determines incremental changes using Hightouch's infrastructure. This engine is easier to set up since it requires read—not write—access to BigQuery.

The Lightning engine requires read and write access to BigQuery. The engine stores previously synced data in a separate schema in BigQuery managed by Hightouch. In other words, the engine uses BigQuery to track incremental changes to your data rather than performing these calculations in Hightouch. Therefore, these computations are completed more quickly.

Hightouch recommends using the Lightning sync engine when syncing more than 100 thousand rows of data.

If you select the Basic engine, you can switch to the Lightning engine later. Once you've configured the Lightning engine, you can't move back to the Basic engine without recreating BigQuery as a source.

To learn more, including migration steps and tips, check out the Lightning sync engine docs.

Basic versus Lightning engine comparison

The Lightning sync engine requires granting write access to your data warehouse, which makes its setup more involved than the Basic sync engine. However, it is more performant and reliable than the Basic engine. This makes it the ideal choice to guarantee faster syncs, especially with large data models. It also supports more features, such as Warehouse Sync Logs, Match Booster, and Identity Resolution.

Lightning engine permissions

To enable the Lightning engine, you need to provide your service account additional permissions to create schemas read/write data.

Run the following snippet to provision the hightouch_planner and hightouch_audit schemas, which are used for storing logs of previously synced data.

CREATE SCHEMA IF NOT EXISTS `hightouch_audit`;
CREATE SCHEMA IF NOT EXISTS `hightouch_planner`;
GRANT `roles/bigquery.dataViewer`, `roles/bigquery.dataEditor` ON SCHEMA `hightouch_planner` TO "serviceAccount:<YOUR_SERVICE_ACCOUNT>";
GRANT `roles/bigquery.dataViewer`, `roles/bigquery.dataEditor` ON SCHEMA `hightouch_audit` TO "serviceAccount:<YOUR_SERVICE_ACCOUNT>";

Test your connection

Next steps

Once your source configuration has passed the necessary validation, your source setup is complete. Next, you can set up models to define which data you want to pull from BigQuery.

The BigQuery source supports these modeling methods:

• writing a query in the SQL editor
• using the visual table selector
• leveraging existing dbt models
• leveraging existing Looker Looks
• leveraging existing Sigma workbooks

The SQL editor allows you to query data from all projects that your GCP service account has access to. The table selector only supports querying from the project specified in the source configuration.

Tips and troubleshooting

If you encounter an error or question not listed below and need assistance, don't hesitate to reach out. We're here to help.

Hashed values are displayed as objects of bytes

When previewing your model or when syncing your data to the destination, you might see an object of bytes, such as:

{"0":123,"1":456,"2":789,"3":012,...}

This happens if you use certain hashing algorithms in BigQuery to hash your data, such as SHA256. To resolve this, open the SQL editor and use TO_HEX or TO_BASE64 to query this model column:

SELECT TO_HEX(model_column_name) as string_model_column_name, * FROM dataset.table

No matching signature for operator {...} for argument types: {...}, {...}.

An example of the error message is No matching signature for operator != for argument types: DATE, TIMESTAMP.

This error means that your model definition is using the named operator, for example, != to compare two incompatible types, for example, a date to a timestamp. To resolve the issue, ensure your model columns are properly typed or revise your model query.

Unable to process number due to [big.js]

An example of the error message is Unable to process number due to [big.js] Imprecise conversion: [big.js] Imprecise conversion

This error occurs when one or more values have a very high precision, for example 123.12345678910. Hightouch attempts to convert the value to a number (based on the data type selection of your model). An error occurs if the number can't be precisely represented by a JavaScript number.

To resolve the issue, ensure the values in your model columns have a lower precision. Additionally, if your value doesn't need to be a number datatype, casting the value to a string datatype can also resolve this issue.

Dataset was not found in location

You may encounter this error if you customized the lightning cache location. You must use fully qualified table names - project_id.dataset.table in model queries. Note that you cannot use dbt, Looker, Sigma or the Table Selector to create models when a custom project is specified for the lightning cache. You will have to use the SQL query editor for the model query.