Google Analytics Parallel Tracking is a method of duplicating all Google Analytics hits and sending them to another data processing engine and database or data warehouse.
One of the most common use cases is to send all raw Google Analytics hits to BigQuery. This provides a dataset that is very similar to the native export of Google Analytics 360.
Why use Google Analytics Parallel Tracking
Let’s take a look at the three most popular reasons why companies decide to implement Google Analytics Parallel Tracking.
Access to raw hit-level data
With the free version of Google Analytics, it is almost impossible to gain access to raw data. You can get close by using custom dimensions and the Reporting API but even that doesn’t provide the raw underlying hits.
With Google Analytics Parallel Tracking, you can access hits as they were sent from your site or via the measurement protocol – including IP address, user agent and other information that is not available within Google Analytics.
Avoid sampling
When building more advanced reports within Google Analytics UI or querying data via the API, sooner or later you will end up seeing sampled data.
When having access to raw hit-level data you never have to worry about data being sampled. This is especially useful when analyzing hits generated by a single user (or a small group of users) in a longer time period.
Working with PII data
PII or Personally Identifiable Information is not allowed in any data point you send to Google Analytics. Violating this rule can get your account terminated.
You could, of course, hash all the sensitive data and send it to Google Analytics but it is much easier to just exclude it from the hits going to Google Analytics and only include it in the hits that are going straight to your data warehouse.
Having access to this kind of data makes it much easier to join your analytics data with data from CRM, CMS and even offline data.
How does Google Analytics Parallel Tracking work
1. Tracking code modification
A small modification to the Google Analytics tracking code is required in order to enable streaming all hits to another data processing pipeline. This modification is easy to implement and is doable with gtag.js, analytics.js or within Google Tag Manager.
2. Data processing
Data processing engine receives and processes all hits similar to Google Analytics. This system can scale almost infinitely and is free from processing limits existing in Google Analytics.
3. Data is stored in the warehouse
Our recommended data warehouse is Google BigQuery but we can work with any other solution as well. All data is processed and available for reporting in near-real-time.
Useful links
- Google Analytics Parallel Tracking Service
- How to Query and Analyze Google Analytics Data with BigQuery
- Unsampled Hit-Level Google Analytics Data Without 360
- BigQuery documentation
Last modified: April 27, 2020
Want to see more articels like What is Google Analytics Parallel Tracking? Check out all definitions in the Analytics Dictionary.