Develop Reference

Exclude age, gender and region from firebase Analytics on Android

Recently I had a specific request, don't allow firebase analytics to collect user's data like gender, age and region (Due to GDPR). I know, we can completely disable Analytics.
Firebase Analytics says:
Analytics automatically logs some user properties; you don't need to add any code to enable them. If you need to collect additional data, you can set up to 25 different Analytics User Properties per project. Note that user property names are case-sensitive and that setting two user properties whose names differ only in case results in two distinct user properties being logged.
You can't use a small set of user property names reserved by Google:
Age
Gender
Interest
I was wondering if there is a way to only disable these users properties which are logged automatically??
Thanks a lot for your effort.

There is no option to exclude any of the automatic properties (except, apparently, on iOS, where you have to do some configuration to enable them). The only configurations are documented here. If you're required not to collect that information about the end user, your only option is to disable Analytics altogether.
Note the the data is fully anonymized. There is no way to track the properties back to a specific end user, unless you add some personally identifying information to a user record, such as a user ID.

Firebase Analytics Filter By User ID

The Firebase Android SDK for analytics provides a method named setUserId(String id), now i enabled firebase logging and every time i called the method setUserId(1234) i see this in the logs Setting user property (FE): _id, 1234.
My question is how come the Firebase dashboard does not let us filter by this property, there is no property not for userId and not _id, i even tried to add _id as a property in the dashboard but Firebase does not allow property names to start with an underscore.
Do i really have to stop calling that method and just do setUserProperty("userId", 1234) or am i missing something...

I have spent a few hours on this and found the answer i was looking for more or less, hopefully this might save someone else some time.
The only way to filter by Firebase's User Id property is by creating an audience, there you will be able to pick the user id property that is supplied from the setUserId(String id) method.
It is a downer though because you won't see any events for the user/user's you want from before the audience was created.

I was scratching my head with this issue. Initially, I felt so dumb when I could not find how to use user id for filtering in Firebase analytics dashboard that I am setting with official API setUserId(). When I try to look for answer, I realized that I am not the only one. Finally, this is how I was able to filter events based on user id.
In Android Code, I set user id using setUserId() method. When I read bigQuery has this property as user_id, I thought to give it a try by adding a new user property in Firebase Dashboard with the hope that it will fetch the reported id.
And... that worked...Now I can filter new events based on user id.

In audience you have to select contains, exactly match or regular expression for a field.
So to filter for a user you have to add all your users to the audience, which looks cumbersome.
Isn't using setUserProperty is better? Have you faced any problems with using it ?

Android Firebase User ID mapping issue

I had integrated Firebase analytics in Android few months back.
Initially I was setting the user_id wrongly, by using the following code :
mFirebaseAnalytics.setUserProperty("userId", "<value_of_type_1_for_user_id>");
As a result my user_id column in bigquery was null. Hence on my next release, I corrected it and used the following method :
mFirebaseAnalytics.setUserId("<value_of_type_2_for_user_id>");
This created a problem as all those users who had installed/used the app between those two release dates have an extra user property key named "userId". And this has some wrong values of user_id.
Now, another strange thing that has happened is that there are many new users who are onboarding the platform now with the latest version of APK. A small percent of these users still have the malicious key of "userId" under user_properties.
I need help in understanding the following :
- Understanding the reason behind why new users (a random small group of them) are still being affected by this wrong mapping ?
- How to clean up the user property field of old users by removing this property key for all users.

You can set the user property "userId" to null/empty and the user property will be removed from the future events. User property will stick with events until you explicitly remove it.

How to set User property to "Purchasers" in Firebase?

Unfortunately, Firebase documentation is incomplete.
There is a filter inside Firebase Analytics panel:
And Firebase Notifications panel:
Which allows to filter users, but there is no instructions on how to set a user property as a Purchasers in the documentation.
The only description about Purchasers in the documents is this link.
I think there must be a command like this:
firebaseAnalytics.setUserProperty("USER_TYPE", "PURCHASERS"/"NONPURCHASERS");
But I have no clue about the Property name or the Value. Does anybody know how to do this?

The Purchasers Audience is one of the two predefined Audience available (the other one being All Users), wherein a Purchaser is defined as:
Users who have completed an in-app purchase or ecommerce purchase.
-- source
I think that pretty much explains it on how a user is labeled as a Purchaser.
You can however, create your own Audience if you prefer (from the same source above):
Create an audience
You create an audience by defining criteria that reference any of the user properties and events (along with any of their parameters) logged by the app. Once created, an audience accumulates users who meet the specified criteria from that point onward. You can create up to 50 audiences.
To create a new audience:
In Firebase Analytics, click the Audiences tab.
Click NEW AUDIENCE.
Enter a name and description for the audience. This name and description will allow you to identify the audience in the management table.
Click Select Event or User Property.
Create one or more conditions that define who should be included in this audience. You can create conditions that include the users who have taken specific actions (Event) or who share a property (User Property). Combine multiple conditions with OR or AND.
Click Create to save your conditions and create the audience.

Here is how you can create your Purchasers audience:
Click audience on the left-side menu
Click new audience on the top right
Select a filter > choose event > choose in_app_purchase
Click add a parameter
count > greater than 0

Is possible to set firebase custom properties analytics as array?

I'm planning to use Firebase Analytics for my Apps. According to Firebase Docs is possible to set custom user properties docs here Android and IOS. Latter one can use the properties to create audiences in Firebase Analytics Dashboard as described here Analytics DashBoard
I want to use a user custom property as an Array.
For example:
Setting user property to TagsUserLikes=["hashtagBlue","hashtagRed","hashtagGreen"] so latter one I'll be able to track an audience of all the users that follow a particular tag, meaning users that contain the tag in the array.
So latter in the Analytics dashboard a audience defined solely as TagsUserLikes="hastagBlue" will match all users that contain this tag ["hashtagBlue","hashtagRed","hashtagGreen"] and ["hashtagBlue","hashtagDogs"],?
Is this supported in the Google Analytics Dashboard (create audience)?
Is this supported in the Google Big Query?
If Yes How can I achieve this?
tks

Although you could theoretically concatenate a list of tags in one User Property value (and use it for audience creation), the limit on the length of User Property values (36 characters) might make this approach impractical for you.
Alternatively, you can log an event such as "tag_followed" with a parameter "tag_name=" and then you can create an audience of users who log tag_followed with a particular tag_name value.