In the second of his four-part New World for Mobile App Analytics series, Phil looks at the tool selection process, some of the differences between Firebase and Google Analytics… and how both need GTM.
Just so you are aware, I’m going to avoid any how-to guides for hybrid apps because, as mentioned in part 1, it is the same as a website, which is far too old world for this blog.
This is the New World, so the next bit is solely about native app implementations.
Firebase or Legacy GTM/GA?
So, now we know which tool to implement. Well, sort of… What’s this Firebase thing?
Google recently bought Firebase and stuck all their branding on it. I’m sure, because they spent all that money, they recommend Firebase as the preferred solution for mobile analytics. According to Google, Firebase is not solely an analytics tool but a combination of tools to “build better apps and grow successful businesses”.
Here’s a nice infographic to show you the other things Firebase is leveraging:
Obviously, all we really care about is the analytics, which is built in to the Firebase SDK as standard. I must stress that this is Firebase Analytics, not Google Analytics. For more detail and some quick differences between the two, read this.
For me, it took quite a bit of research to see what the core Firebase objective is, other than analytics. It does seem quite an expansive tool, and its complexity might put web-based analysts off pushing it to clients.
However, if you have no barriers, my recommendation is to implement Firebase, mainly because it’s (relatively) future-proof. Google will continually invest in it and that seems like a pretty good reason to go with it.
Of course, there are reasons that you might not be able to use Firebase and in that case, you will have to use the legacy GTM SDK.
Still don’t understand the difference?
*To confuse matters, you can push your data from the Firebase SDK via GTM to either Firebase Analytics or Google Analytics. We’ll get on to that bit later.
Mobile app analytics without GTM?
I know I am throwing another level of detail in here, but hopefully you’re still with me and not jumping ship.
The answer is yes, but the same pros for GTM for web are applicable here, for example: more control for the analyst, less developer input, updating tags outside of release cycles.
If you insist on going for a standard implementation of Google Analytics SDK, it will provide the following level of data out of the box, for example:
- Number of users and sessions
- Session duration
- Operating systems
- Device models
Also, for our new friend Firebase, there are similar levels of detail that are automatically collected without GTM. Some highlights are:
- App version
- Device brand
- Device category
- Device model
- First open time
Next week, I will describe the implementation process, including data layer naming conventions, guides to send to the app developer and an example of GTM tag setup.