If you’re thinking of setting up a GA App+Web, there are a few things that will make your life easier (and save your time & research).
CAVEAT: The following tips are valid as of November 2019. The platform is constantly developing, so I’d expect that the ‘imperfections’ may at one point be amended.
NOTE: This idea of listing enumerated tips comes from Simo Ahava’s GTM blog & resource website. I’ve found it immensely practical, so here it is.
Tip 1: GA-FIR integration
When setting up your Firebase project, don’t disable GA integration (which is default at the moment):
This is when you will be able to create a GA App+Web property integrated with the Firebase project. If you set up a property from the GA interface, you will be left with no Firebase integration:
For now, the only way to create a Firebase-integrated GA property in while setting up a Firebase project.
EDIT: If you start setting up the property in GA, you can still link it with Firebase by adding an app data stream. It may not be instinctive for thosewho don’t have apps, but it’s possible – even if you add a data stream from a non-existing app (similarly to Tip 7).
Tip 2: show event parameters
If you haven’t worked with Google Analytics for Firebase before, you may not know this one:
In order to see the parameters sent with your Firebase/GA App+Web events, you will need to add them in the interface.
The parameter data will only be available since the moment it is added (despite the fact that you may have been sending your hits with parameters since day 1). So make sure you add your parameters as soon as they are available.
Tip 3: Custom Events
- Recommended Events are those suggested by Firebase, along with parameters. These events are different for particular verticals, e.g.
add_to_cartevent with parameters such as
quantityin the ‘Ecommerce’ vertical. Firebase documentation recommends that we try to use these events.
You will see the recommendations in the Events menu in Analytics interface:
- Custom Events are events with values of our choice, e.g.
fancy_eventwith parameters of our choice, e.g.
The documentation claims that Custom Events cannot be viewed in the web UI:
This, however, is not entirely true: right now, Custom Events seem to have the same availability in the reports as Recommended Events:
Tip 4: numeric parameters
Event parameters come in 2 formats: text (string) and numeric. Any text parameter – built-in, Recommended or Custom – can be viewed in the reports in Analytics interface. Numeric parameters, however, are a different story.
In Universal Analytics, we are used to viewing numeric values as aggregated metrics. In GA A+W, however, the only such metric is
value (apart from the automatically tracked – e.g.
value can be viewed as a sum or average in a set of instances (events). Any other numeric parameter (even Recommended) can only be viewed similarly to text parameters – as a dimension, against the
event_count Metric. In BigQuery we are free to aggregate those parameters and use them as Metrics, but if we plan to stick to the web interface, use
value as your aggregated Metric for now.
Thus, if for example you need to track product prices (and don’t want to wait for any dedicated ecommerce tracking in GA A+W), use
value rather than
Tip 5: parameter quota
The maximum number of parameters is 50 text and 50 numeric, according to Google documentation (things are changing quickly – GA for Firebase limits were up until recently only 10 text and 40 numeric).
When planning your data, however, you need to to think in terms of event-parameter combinations rather than standalone parameters. Adding the same parameter to multiple events will draw the equivalent number from the parameter quota:
Tip 6: Analytics data in BigQuery
Integrate BigQuery from the start. This way, you’ll get all your data there without missing a byte… as long as you upgrade to a paid (pay as you go) plan. This is the only way to get your Analytics data into BigQuery. (So the advertised “free integration” is not exactly free though still cheaper than GA360…)
Only once you’ve upgraded, you will be able to activate the export:
(Note that BigQuery integration will automatically work for the Firebase Messaging dataset for your website – no plan upgrade required.)
Tip 7: you need your mobile app…
(You may already know this from another blog, but just in case you don’t …)
In order to get your data sent to BigQuery, you will need a mobile app added to the Firebase project. If you don’t have a real mobile app (which you most likely don’t), just add a fake one and skip validation. No doubt this is a bug and Google will remove it at some point, but for now that’s what it is.
We’re still in beta – Google are working on changes to GA App+Web as we speak. These tips may not be as useful in a few months time, but if you’re eager to get rolling, then they could come in handy. In my opinion. If you run into any other sort of trouble during setup or early use, let me know in the comments.