Web measurement should evolve in universal measurement
When it comes to measuring website behavior, Google Analytics is a very powerful solution and one of the main market leaders. Remarkably enough, Google Analytics has been neglecting the measurement of other devices such as (native) apps, wearables, IOT devices or other non-browser based channels.
Most of the out-of-the-box reports available in Google Analytics, are mainly combining dimensions such as “Landing Page” or “Source/Medium” with metrics like “Sessions”, “Bounce rate” or “Pages/ Session”. These reports are actually only making sense when you want to analyze (traditional) website behavior.
Hypothetically speaking, if you want to measure the behavior of the fridges in your company, concepts as “Bounce Rate” or the “Browser Version” of this “Session” are just not applicable. If you want to calculate the ROI of buying new fridges for your company, you want something that reflects the actual usage of the fridge: how many times was the door opened? How long did it stay open? What is the temperature inside the fridge?
In this case, decisions will be made based on specific events such as “door opens”, “door closes” or specific metrics, like temperature and kWh. This example may be a bit abstract or less relevant for your business, but the same concept is applicable for your (native) app, paid campaigns, physical store visits or whatever behavior you want to measure and analyze. This requires a different, more generalized way of measuring, analyzing and reporting.
Google Analytics for Firebase
With the latest update of “Google Analytics for Firebase” this generalized approach has now become reality. Let us take a step back and clarify the different product names of Google related to analytics (because it can be confusing).
Firebase already exists several years, it offers a variety of different cloud based services for Rapid Mobile Application Development such as: authentication, push notifications, database, storage and analytics. The analytics module of Firebase was named: “Firebase Analytics”. Firebase was acquired by Google in 2014 and is the preferred solution for app analytics in the Google ecosystem for a few years now. With the latest updates in Google Analytics, the new “Apps” and “Apps and web” properties is actually “Firebase Analytics” rebranded as “Google Analytics for Firebase”.
From a technical and data structural perspective, you are using Firebase Analytics.
So, in short: “Google Analytics App + Web” = “Google Analytics for Firebase” = “Firebase Analytics”.
At The Reference, we have been using Firebase for several years with different purposes and use cases. Some of its (non-analytics) cloud services such as real-time database, remote configuration and dynamic links have been proving to be very powerful. However, we want to focus a bit on the analytics capabilities of Firebase a.k.a. “Google Analytics for Firebase”.
1. Flexible implementation
2. Event logging
The data is structured based on event logging. This may sound a bit abstract but an “event” is applicable on any device or application. A page view, a payment, a new lead, a transaction, a fridge that opens, … anything is basically an event. As a result, you can aggregate your data from different channels. E.g. if you have a native commerce app and an e-commerce website you can aggregate events like add to basket or transactions and compare conversion rates.
3. Raw data without limitations
All this event data, together with other Firebase generated data, can be exported without limitations to BigQuery – for free. BigQuery is Google's highly scalable enterprise data warehouse. It is designed to make data analysts more productive with unmatched price-performance. Because there is no infrastructure to manage, you can focus on uncovering meaningful insights using familiar SQL. The BigQuery export of web properties is only available within the paid version of Google Analytics (a.k.a. Google Analytics 360). However, this export is a now a non-paid feature in Google Analytics for Firebase.
This means that we can have raw and unsampled data available in BigQuery. This opens a lot of new doors:
- Transform your data in a preferred format with SQL (Structured Query Language) statements;
- Aggregate and connect your data with other (backend) sources;
- Visualize your data with Google Data Studio or other reporting solutions such as PowerBI or Tableau;
- Unleash Artificial Intelligence on your data with machine learning, anomaly detection or predictive modeling capabilities in the Google Cloud Platform.
Conclusion“Google Analytics for Firebase” opens many doors but it will require a bit of mind shift to measure and analyze your data. Compared to the traditional web properties of Google Analytics you will probably miss a lot of the out of the box dimensions, metrics and reports. Google Analytics is therefore still the preferred solution if you want to analyze website behavior. But we need to think broader than measuring website behavior. Any interaction has the potential to give relevant insights. Therefore, both solutions can perfectly co-exist, but eventually (Google Analytics for) Firebase is more future proof.
This article was created by Thomas Danniau and Geert Michiels.