(Deprecated) Google Analytics UA

This connector previously captured data from a view in Google Universal Analytics.

danger

This connector and Universal Analytics are now deprecated along with the Google Analytics Universal Analytics API - (see Google announcement.

Google Analytics 4 is supported by a separate connector.

It is available for use in the Flow web application. For local development or open-source workflows, ghcr.io/estuary/source-google-analytics-ua:dev provides the latest version of the connector as a Docker image. You can also follow the link in your browser to see past image versions.

This connector is based on an open-source connector from a third party, with modifications for performance in the Flow system.

Supported data resources

The following data resources are captured to Flow collections by default:

• Website overview
• Traffic sources
• Pages
• Locations
• Monthly active users
• Four weekly active users
• Two weekly active users
• Weekly active users
• Daily active users
• Devices

Each resource is mapped to a Flow collection through a separate binding.

You can also configure custom reports.

Prerequisites

There are two ways to authenticate with Google when capturing data from a Google Analytics view: using OAuth2, and manually, by generating a service account key. Their prerequisites differ.

OAuth is recommended for simplicity in the Flow web app; the service account key method is the only supported method using the command line.

Using OAuth2 to authenticate with Google in the Flow web app

• The View ID for your Google Analytics account. You can find this using Google's Account Explorer tool.

• Your Google account username and password.

Authenticating manually with a service account key

• The View ID for your Google Analytics account. You can find this using Google's Account Explorer tool.

• Google Analytics and Google Analytics Reporting APIs enabled on your Google account.

• A Google service account with:

  • A JSON key generated.
  • Access to the source Google Analytics view.

Follow the steps below to meet these prerequisites:

1. Enable the Google Analytics and Google Analytics Reporting APIs for the Google project with which your Analytics view is associated. (Unless you actively develop with Google Cloud, you'll likely just have one option).

2. Create a service account and generate a JSON key During setup, grant the account the Viewer role on your project. You'll copy the contents of the downloaded key file into the Service Account Credentials parameter when you configure the connector.

3. Add the service account to the Google Analytics view.

   1. Grant the account Viewer permissions (formerly known as Read & Analyze permissions).

Configuration

You configure connectors either in the Flow web app, or by directly editing the catalog specification file. See connectors to learn more about using connectors. The values and specification sample below provide configuration details specific to the Google Analytics source connector.

Properties

Endpoint

The following properties reflect the Service Account Key authentication method. If you're working in the Flow web app, you'll use OAuth2, so some of these properties aren't required.

Property	Title	Description	Type	Required/Default
/credentials	Credentials	Credentials for the service	object
/credentials/auth_type	Authentication Type	Authentication method. Set to Service for manual configuration, or use OAuth in the web app.	string	Required
credentials/credentials_json	Service Account Credentials	Contents of the JSON key file generated during setup.	string	Required
/custom_reports	Custom Reports (Optional)	A JSON array describing the custom reports you want to sync from GA.	string
/start_date	Start Date	The date in the format YYYY-MM-DD. Any data before this date will not be replicated.	string	Required
/view_id	View ID	The ID for the Google Analytics View you want to fetch data from. This can be found from the Google Analytics Account Explorer: https://ga-dev-tools.appspot.com/account-explorer/	string	Required
/window_in_days	Window in days (Optional)	The amount of days each stream slice would consist of beginning from start_date. Bigger the value - faster the fetch. (Min=1, as for a Day; Max=364, as for a Year).	integer	1

Bindings

Property	Title	Description	Type	Required/Default
/stream	Stream	Data resource from the Google Analytics view.	string	Required
/syncMode	Sync Mode	Connection method. Always set to incremental.	string	Required

Custom reports

You can include data beyond the default data resources with Custom Reports. These replicate the functionality of Custom Reports in the Google Analytics Web console.

To do so, fill out the Custom Reports property with a JSON array as a string with the following schema:

[{"name": string, "dimensions": [string], "metrics": [string]}]

You may specify default Google Analytics dimensions and metrics from the table below, or custom dimensions and metrics you've previously defined. Each custom report may contain up to 7 unique dimensions and 10 unique metrics. You must include the ga:date dimension for proper data flow.

Supported GA dimensions	Supported GA metrics
ga:browser	ga:14dayUsers
ga:city	ga:1dayUsers
ga:continent	ga:28dayUsers
ga:country	ga:30dayUsers
ga:date	ga:7dayUsers
ga:deviceCategory	ga:avgSessionDuration
ga:hostname	ga:avgTimeOnPage
ga:medium	ga:bounceRate
ga:metro	ga:entranceRate
ga:operatingSystem	ga:entrances
ga:pagePath	ga:exits
ga:region	ga:newUsers
ga:socialNetwork	ga:pageviews
ga:source	ga:pageviewsPerSession
ga:subContinent	ga:sessions
	ga:sessionsPerUser
	ga:uniquePageviews
	ga:users

Sample

This sample reflects the manual authentication method.

captures:
  ${PREFIX}/${CAPTURE_NAME}:
    endpoint:
      connector:
        image: ghcr.io/estuary/source-google-analytics-v4:dev
          config:
            view_id: 000000000
            start_date: 2022-03-01
            credentials:
              auth_type: service
              credentials_json: <secret>
            window_in_days: 1

      bindings:
        - resource:
            stream: daily_active_users
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

        - resource:
            stream: devices
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

        - resource:
            stream: four_weekly_active_users
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

        - resource:
            stream: locations
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

        - resource:
            stream: monthly_active_users
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

        - resource:
            stream: pages
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

        - resource:
            stream: traffic_sources
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

        - resource:
            stream: two_weekly_active_users
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

        - resource:
            stream: website_overview
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

        - resource:
            stream: weekly_active_users
            syncMode: incremental
          target: ${PREFIX}/${COLLECTION_NAME}

Learn more about capture definitions.

Performance considerations

Data sampling

The Google Analytics Reporting API enforces compute thresholds for ad-hoc queries and reports. If a threshold is exceeded, the API will apply sampling to limit the number of sessions analyzed for the specified time range. These thresholds can be found here.

If your account is on the Analytics 360 tier, you're less likely to run into these limitations. For Analytics Standard accounts, you can avoid sampling by keeping the window_in_days parameter set to its default value, 1. This makes it less likely that you will exceed the threshold. When sampling occurs, a warning is written to the capture log.

Processing latency

Data in Google Analytics reports may continue to update up to 48 hours after it appears.

To ensure data correctness, each time it reads from Google Analytics, this connector automatically applies a lookback window of 2 days prior to its last read. This allows it to double-check and correct for any changes in reports resulting from latent data updates.

This mechanism relies on the isDataGolden flag in the Google Analytics Reporting API.