Skip to content

Latest commit

 

History

History
51 lines (36 loc) · 3.24 KB

README.md

File metadata and controls

51 lines (36 loc) · 3.24 KB
Raw

Google Analytics 4 Export

This is an example Meltano project tailored to extracting GA4 event data from BigQuery, having set up a BigQuery export from Google Analytics. You will need to provide credentials for the BigQuery connection, and optionally Postgres or Snowflake (should you choose either as a destination).

Checkout our Matatika article on the huge cost savings possible with this approach.

Prerequisite

First thing to do is set up the Google Analytics 4 to BigQuery export for event data. Follow this video or the steps outlined below:

IMAGE ALT TEXT HERE

  1. Go to the Google Analytics dashboard
  2. Select the property/app you want to configure the BigQuery export for
  3. Go to Admin > Property settings > Product links > BigQuery links
  4. Click the Link button and follow the wizard to configure a new link with BigQuery

For information on the export schema (what datasets, tables and columns will be created), refer to this guide.

Once this has been done and at least one export has completed successfully, we can use this Meltano project standalone or in Matatika to extract the Google Analytics data from BiqQuery and load it somewhere.

Performance

To properly take advantage of this setup, the BigQuery tap supports an alternate batch-extraction mode when providing google_storage_bucket as config, which will be signinficantly faster when paired with a loader that supports BATCH messages (in this case, target-snowflake).

Important

Postgres targets do not support BATCH messages (yet), so providing google_storage_bucket for the tap will cause the target here to fail with

Unknown message type BATCH in message

In this case, you should omit Google Storage Bucket and continue with the standard RECORD-based sync - this will be markedly slower if you have a lot of GA4 event data.

Running with Matatika

You can load this repository into Matatika as a workspace to configure and run the BigQuery GA4 export Meltano jobs as pipelines:

  1. Go to https://app.matatika.com (sign up/log in if you haven't already)

  2. On the "Setup your workspace" view, provide a workspace name and provide the repository url

    https://github.com/Matatika/example-ga4-export.git

    under the "Advanced" section

    • If you already have a Matatika account and workspace(s), you can find the "New workspace" option under the workspace context menu (top-left)

    image

  3. Hit continue and wait for the workspace to create

  4. Follow the prompts to configure the export pipeline(s) you are interested in running

  5. Save the pipelines(s)

  6. Click "Close" and then run the pipeline(s) you just configured

Note

Your credentials are always encrypted and stored securely by Matatika