Event

Google Analytics 4, or GA4 (formerly known as “App + Web”), is a new kind of property, with different reports than what you’re used to seeing in legacy Universal Analytics (UA) and Firebase properties. GA4 is an analytics service that enables you to measure traffic and engagement across your websites and mobile apps using customizable reports. One advantage of a GA4 property is that you can use it for a website, an app, or both a website and app together. UA properties only support websites and the Firebase Console only supports native app data.

Data collection is enabled through SDK instrumentation. Once your app is properly configured, data is ingested into the mParticle platform, which maps inbound data to Google Analytics features and their required formats, and then forwards the data to Google Analytics.

Setup Steps

You will need a GA4 account and an app property for every app that you want to track. To set that up, follow Google’s instructions here. mParticle supports sending data to GA4 via web, native iOS and Android apps, and server to server. Each platform will require a data stream. See Google’s documentation for how to create a GA4 data stream.

On mParticle you need to create a new Configuration. To do so, find Google Analytics 4 on the Directory, click on Setup, give your new Configuration a name and decide if you want the same credentials for Development and Production. Finally, click Save.

Web

Data can be sent to GA4 via Web both client and server side. If you are self-hosting, you can import either @mparticle/web-google-analytics-4-client-kit or @mparticle/web-google-analytics-4-server-kit depending on your use case. Note that if you want to send data server side, you must import the server kit AND enable Forward Web Requests Server Side. Enabling this setting and importing the client side kit will not work.

A GA4 Measurement ID is automatically generated when a web data stream is added. To find your Measurement ID, see here. Our GA4 web integration allows you to send your web data both client and server side.

Sending Data via Web Client Side

To send data client side on web, simply create a new Google Analytics 4 output in the mParticle UI and add it as a Connected Output to the Web Input. Add the Measurement ID to the appropriate field in the GA4 Connection Settings in the mParticle UI. Sending data server side requires mParticle to load Google’s GA4 web SDK (gtag.js).

Sending Data via Web Server Side

As required by the Google API, you must use the GA4 SDK for server-side integration. mParticle supports forwarding with both a client-side kit to forward to GA4 and S2S to the GA4 Measurement Protocol API. However, both have requirements that require the GA4 SDK be integrated natively. If you turn on the GA4 integration in our UI AND are using the mParticle web SDK, mParticle does this for you. If you want to forego the web SDK altogether, you must include the GA4 SDK yourself (GA4 docs). If you are currently forwarding data S2S to mParticle’s Events API, you must update the payloads to support forwarding to GA4, which requires the GA4 SDK to grab the relevant client_id.

You may prefer to send web data server side in order to reduce both the number of calls the browser makes, and the size of your website. In this scenario, events are sent to mParticle’s servers, and then forwarded from mParticle’s servers to GA4.

To send data server side, check Forward Requests Server Side in the Connection Settings. Add the Measurement ID and you will also need to include a Measurement Protocol API Secret for Web Stream. On GA4, each data stream can have one or more Measurement Protocol API Secrets. To create one:

1. Locate your data stream where you viewed your Measurement ID from above.
2. Click on Measurement Protocol API Secrets
3. Click Create.
4. Provide a Nickname, and click Create again.
5. Copy the newly generated Secret value and paste it into the mParticle setting into the mParticle connection setting for GA4.

Sending Data via Web Server Side with the mParticle web SDK

Google’s server side API for GA4 requires a client_id which still necessitates loading Google’s Global Site Tag (gtag.js), but our web SDK automatically loads gtag.js for you. mParticle sends the client_id to our servers which then forward to Google server side.

Sending Data via Web Server Side without the mParticle web SDK

Optionally, if you are not using the mParticle web SDK, you can resolve the client_id by directly calling the Global site tag API.

When you send the payload to the mParticle endpoint, include client_id as part of integration attributes under the key 160. For example:

"integration_attributes": {
    "160": {
        "client_id": "your_client_id"
    }
}

Note that screen_view events are not sent, even if you include client_id as part of the integration_attributes.

Native

You can set up your native app to process GA4 data client side from your users’ devices.

Add the Kit

mParticle’s GA4 integration requires that you add the mParticle GA4 Kit to your iOS/tvOS or Android app.

mParticle publishes the GA4 kit as separate iOS/tvOS and Android libraries which have a transitive dependency on the mParticle core libraries. You can add them to your app via Carthage, Cocoapods, or Gradle:

target '<Your Target>' do
    pod 'mParticle-Google-Analytics-Firebase-GA4'
end

dependencies {
    // Ensure the Kit version matches that of the mParticle Core SDK that you're using
    compile 'com.mparticle:android-googleanalyticsfirebasega4-kit:VERSION'
}

Reference the Apple SDK and Android SDK guides to read more about kits.

Sending Data via Native Apps Client Side

Firebase is still used to send data client side to GA4. As a result, there are several references to Firebase and Firebase docs throughout this page.

Before GA4, mobile data was analyzed within the Firebase Console. If you have a legacy Firebase property, Google provides step-by-step instructions to upgrade a Firebase instance so that data will also flow to GA4. After upgrading, your mobile data will be available for analyzing within the GA4 dashboard alongside web data. Once your Firebase properties are upgraded to GA4, the data will show up in both the Firebase Console and the GA4 console.

To send data to GA4 client side in an app, first add a platform-specific data stream`

1. Follow the steps here under “Add a data stream” to create a data stream for Android or iOS/tvOS. These instructions include downloading either the google-services.json for Android, or the GoogleService-Info.plist for iOS/tvOS.
2. Set up Google Analytics for Firebase as an output and connect it to iOS, tvOS or Android in the mParticle UI.
3. Add the mParticle Firebase kit to your app (see platform-specific docs for adding kits for iOS/tvOS and Android).

Android

Due to a known issue in the Firebase Android SDK, it is impossible to programatically initialize Firebase at runtime - you must follow the Firebase documentation for adding Firebase to your application. mParticle will be tracking this issue and if it is resolved, the integration will be updated to support runtime initialization.

The Firebase-GA4 kit will detect if you have initialized Firebase, and use the existing instance in your app if present. Despite this, all typical mParticle controls such as data filtering and user-filtering are available as expected to protect the flow of event data from mParticle to GA4. However, by directly including the Firebase SDK and configuration files in your app, mParticle cannot prevent it from collecting other data automatically.

Please see Firebase’s Android setup guide here.

iOS/tvOS

Our iOS/tvOS implementation also requires you to manually instrument and initialize the Firebase SDK. Data will be automatically forwarded to that instance - mParticle will not create an additional instance.

You must follow the Firebase docs to create a Firebase project and download your GoogleService-Info.plist configuration file. You must then include the plist directly in your app.

Please see Firebase’s iOS/tvOS setup guide here.

Sending Data via Native Apps Server Side

You may prefer to send this data server side in order to reduce both the number of calls the device makes, and the size of your app. In this scenario, events are sent to mParticle’s servers, and then forwarded from mParticle’s servers to GA4.

To send data server side, check Forward Requests Server Side in the Connection Settings. You will also need to include your Firebase App ID and a Measurement Protocol API Secret for Firebase to forward web requests server side. Each data stream can have one or more Measurement Protocol API Secret. To create one:

1. locate your data stream where you viewed your Firebase App ID and then:
2. Click on Measurement Protocol API Secrets
3. Click Create.
4. Provide a Nickname, and click Create again.
5. Copy the newly generated Secret value and paste it into the mParticle setting into the mParticle connection setting for GA4.
6. Copy the Firebase App ID from the Data Stream details page into the connection settings as well.

Sending Data via Native Apps Server Side with the kit

Google’s server side API for GA4 requires an app_instance_id which comes from the Firebase SDK. The mParticle Firebase for GA4 kit automatically sends the app_instance_id to our servers to then forward to Google. This means you will need to include the mParticle Firebase Kit and Firebase SDK in your app.

Sending Data via Native Apps Server Side without the kit

Optionally, you can resolve the app_instance_id by directly calling the Firebase SDK. When the payload is sent to our endpoint, it will require the app_instance_id as part of the integration attributes under the key 160 as the following example:

"integration_attributes": {
    "160": {
        "app_instance_id": "your_app_instance_id"
    }
}

Feeds

To connect a feed, you must provide at least a Firebase App ID or a Measurement ID, and select your Preferred Platform ID that it’s only available for this type of output. If your preferred platform has an associated instance ID in the processed batch, it will be automatically selected to forward the information. If not, the system will attempt to fall back to the other platform if it’s set up and the corresponding instance id is present in the batch.

Other Platforms

Because Google has the requirement for either the app_instance_id or the client_id property in all requests, which come from Google’s SDKs, we are not supporting any other platforms at the time. We are working closely with Google to support other platforms when they provide a strategy that doesn’t require an SDK.

Data Processing Notes

While mParticle forwards all data in real time, GA4 has different processing times depending on the data you are sending. See the Google documentation for more information on latency and data limits.

GA4 has limits around the number of custom dimensions and custom metrics as noted here.

Restricted character replacement

GA4 enforces specific limits when it comes to event names and properties on iOS/Android/server-to-server. To prevent GA4 from rejecting any data, mParticle will make the following modifications to your payloads before sending them to GA4. These modifications are automatic and do not require any configuration.

• Replace each space and non-alphanumeric character with a single underscore: _. The replacement scheme is 1:1, so a single character (:) is replaced by a single underscore (_) and two special characters (::) are replaced with two underscores (__).
• Event names and parameters must begin with an alphabetic character. Any leading underscores or numbers will be removed from the beginning until an alphabetic character is found.
• Remove any prefixes from specific parameters, according to the limitations defined by Google.

Web - Enable Data Cleansing

Unlike Android/iOS/server-to-server, events forwarded to the GA4 web SDK are not impacted by the specific limits that Google outlines. This means that if you are using the same event names/properties across platforms, your web data will differ from other sources.

Use the Enable Data Cleansing setting when configuring your GA4 web connection to avoid these potential name discrepancies. When Enable Data Cleansing is set to true, mParticle automatically modifies names for events ingested from the Web SDK so that they follow the same limitations imposed on the iOS, Android, and server-to-server inputs.

mParticle’s GA4 web kit will follow the above modifications, and also offers a callback that allows customers to modify the individual strings themselves before hitting our cleansing logic.

For connections created before September 25 2023, Enable Data Cleansing is defaulted to false. You must manually enable this setting for these connections. For all newer connections, Enable Data Cleansing defaults to true.

Custom name standardization

You can customize the replacement scheme by defining your own name standardization logic in your app’s code with a callback function immediately before initializing the mParticle SDK. Currently, this is available for the iOS and Android SDKs only.

// The following snippet should be executed prior to initializing the mParticle SDK
[MPKitFirebaseGA4Analytics setCustomNameStandardization:^(NSString* name) {
    // Add your customization logic here to standardize your event names and to eliminate any rejected characters
    return @"your-modified-event-name";
}];

GoogleAnalyticsFirebaseGA4Kit.setClientStandardizationCallback(callback)

// The following snippet should be executed prior to initializing the mParticle SDK
MPKitFirebaseGA4Analytics.setCustomNameStandardization { name in
    // Add your customization logic here to standardize your event names and to eliminate any rejected characters
    return "your-modified-event-name"
}

GoogleAnalyticsFirebaseGA4Kit.setClientStandardizationCallback(callback)

// The following snippet should be executed prior to initializing the mParticle SDK
var callback = object : GoogleAnalyticsFirebaseGA4Kit.MPClientStandardization {
    override fun nameStandardization(name: String): String { 
        // Add your customization logic here to standardize your event names and to eliminate any rejected characters
        return "your-modified-event-name"
    }
}
GoogleAnalyticsFirebaseGA4Kit.setClientStandardizationCallback(callback)

mParticle.ready(function() {
    window.GoogleAnalytics4Kit.setCustomNameStandardization = function(name)  {
        // In this example, the first letter is removed the event or attribute names
        return name.slice(1);
    }
})

Migrating to Google Analytics 4

Depending on if you are migrating from a web or a native property, there are different considerations.

Migrating from the mParticle’s Legacy Google Analytics Web Kit

From a code perspective, there are a few changes you will need to make when migrating from UA to GA4. You may want to familiarize yourself with the differences between GA4 and UA by reading Google’s in-depth comparison here.

Specifically, a few core changes Google made from UA to GA4 that impact our kits are as follows:

• UA supports 5 potential content groups. GA4 uses event scoped custom dimensions to map these old content groups.
• UA supports non-interaction flags. GA4 drops this support.
• UA supports Category, Action, and Label. GA4 drops this support.
• UA supports User Timing. GA4 drops this support.
• UA supports Hit Types. All events in GA4 are considered Events. GA4 does not support Hit Types.

All the associated custom flags related to the above are no longer relevant and should not be included when implementing mParticle’s GA4 web kit.

Migrating from mParticle’s Legacy Native Firebase Kits

Because the GA4 data model is driven by Firebase, no changes are needed to keep your current code working. However, we do support the new GA4 ecommerce events: add_shipping_info and add_payment_info. See Custom Flags for more information.

The Firebase event select_content is select_item in GA4.

Recommended Events and Parameters

Google can auto-generate reports based on recommended event names and recommended event parameters when sent to GA4. You may already have a data model that does not line up exactly with GA4’s new recommended event names and parameters. As such, mParticle and Google’s UIs allow you to map data being sent to them to match Google’s data model for recommended events. Note that mParticle automatically maps commerce events for you, though you may want to customize the mapping.

Depending on your use case, mapping mParticle events to Google’s recommended events is done in either the mParticle UI or the Google UI.

Client-Side Event and Parameter Mapping Modifications (Google UI)

Google specifies in their documentation that modifications in their UI are executed client side before the data reaches Google. Additionally, modifications are generally updated within an hour, but may take longer to take effect. The following is a summary of how to set up event modifications:

1. In the left pane, click Configure, then Events
2. Click Modify event or Create event
3. If your property has more than one data stream, you will be asked to select a data stream
4. Click Create and follow Google’s docs to modify an event or parameter.

Server-Side Event and Parameter Mapping Modifications (mParticle UI)

Google’s UI does not support modifications to their server. However, our Custom Mappings feature does support mapping event names and parameters server-side.

Troubleshooting

If you don’t see data in your GA4 UI, there could be a couple of issues:

1. Your payload does not adhere to GA4 standards. Google has strict payload limitations around number of parameters, user properties, etc. These limitations can be found here.
2. For eCommerce events, if a revenue amount is included, Google requires a currency code.
3. For eCommerce events, if you are not receiving data in GA4, make sure your transaction attributes, product attributes, and event attributes are properly set for your eCommerce event.
4. All transaction IDs should be unique, or they will be deduplicated by Google.
5. For s2s connections, forwarding of Page View events are not currently supported. Either use client-side forwarding or a stand-in Custom Event for s2s connections.
6. Google Analytics 4 only supports the following character set: a-z, A-Z, 0-9 and underscore. If you use other special characters, mParticle will perform a transformation to forward successfully to GA4. If you cannot find your data, look for transformed names in your Google dashboard.
7. Make sure the Measurement API secret used for your GA4 configuration matches that found in your Google dashboard.

8. If you see events being forwarded from mParticle to GA4, but you don’t see an increase in Active Users in GA4, you might need to include a non-zero value for Google’s engagement_time_msec parameter.

   • You can set a default value for engagement time to be included with every event forwarded to Google in your Connection Settings or you can set the GA4.EngagementTimeMsec custom flag at the event level.

Supported User Identities

You can configure the integration to automatically map the following identities to GA4:

• Customer ID
• mParticle ID
• Other-Other10

Supported Events

The mParticle integration for GA4 automatically forwards some event types to Google by default, without requiring any manual configuration. See Events mapped to Google Analytics 4 by default below for more information.

You can use custom mappings to forward different events, or events that are not automatically mapped by mParticle, to Google. See Custom mappings for events below for more information.

Events mapped to Google Analytics 4 by default

This section describes which mParticle event types are supported and mapped to GA4 by default. To see which event types can be mapped to GA4 using custom mappings, see Custom mappings for events

Custom Events

mParticle forwards all custom events to GA4 — including the event name and any custom attributes. For kit integrations, this is via the logEvent API. These follow the GA4 naming rules for valid characters, reserved names and reserved prefixes. Note:

• All invalid characters at the start of an event name are skipped until a valid one is found, and any following invalid characters are replaced by a _.
• Every resultant name with more than 40 characters will be trimmed to allow it to be processed by GA4 according to their standards.

Crash Report Events

Crash report events are forwarded s2s and include 2 outgoing parameters:

• description: This is determined by the CrashReportEvent's stack_trace field, if present, or the message field otherwise.
• fatal: This is determined by the CrashReportEvent's severity and exception_handled fields.

Commerce Events

mParticle automatically maps your data to the following GA4 commerce events. Select a GA4 event type to see the corresponding mParticle event and attributes.

GA4 commerce event types

• add_payment_info
• add_shipping_info
• add_to_cart
• add_to_wishlist
• begin_checkout
• purchase
• refund
• remove_from_cart
• screen_view
• select_item
• select_promotion
• view_item
• view_item_list
• view_promotion

Product attribute mappings

The product field on mParticle commerce events maps to items in GA4. Here are the corresponding GA4 item parameters for each mParticle product attribute.

ad_impression

add_payment_info

Requires custom flags on native mP commerce events. Alternatively, this is also supported via s2s Custom Mappings.

add_payment_info event attribute mappings

add_shipping_info

Requires custom flags on native mP commerce events. Alternatively, this is also supported via s2s Custom Mappings.

add_shipping_info event attribute mappings

add_to_cart

add_to_cart event attribute mappings

add_to_wishlist

add_to_wishlist event attribute mappings

begin_checkout

begin_checkout event attribute mappings

purchase

purchase event attribute mappings

refund

refund event attribute mappings

remove_from_cart

remove_from_cart event attribute mappings

screen_view

screen_view event attribute mappings

select_item

select_item event attribute mappings

select_promotion

Each promotion is mapped to a distinct GA4 event, as opposed to product-based events which can include that info in an items array.

select_promotion event attribute mappings

view_item

view_item event attribute mappings

view_item_list

view_item_list event attribute mappings

view_promotion

Each promotion is mapped to a distinct GA4 event, as opposed to product-based events which can include that info in an items array.

view_promotion event attribute mappings

Custom flags

Custom flags are used with native mParticle event types to send partner-specific data points:

add_shipping_info custom flag example

To map to a Firebase add_shipping_info event, pass a custom flag of GA4.CommerceEventType equal to add_shipping_info and an optional custom flag of GA4.ShippingTier with a string value. The following examples show constants being used for iOS/tvOS and Android:

import com.google.firebase.analytics.FirebaseAnalytics;
import com.mparticle.kits.GoogleAnalyticsFirebaseGA4Kit;

CommerceEvent event = new CommerceEvent.Builder(Product.CHECKOUT_OPTION, new Product.Builder("Spa Essentials", "spa-1", 100.00).build())
                .addCustomFlag(GoogleAnalyticsFirebaseGA4Kit.CF_GA4COMMERCE_EVENT_TYPE, FirebaseAnalytics.Event.ADD_SHIPPING_INFO)
                .addCustomFlag(GoogleAnalyticsFirebaseGA4Kit.CF_GA4_SHIPPING_TIER, "overnight")
                .build();
MParticle.getInstance.logEvent(event);

import com.google.firebase.analytics.FirebaseAnalytics

val event = CommerceEvent.Builder(Product.CHECKOUT_OPTION, Product.Builder("Spa Essentials", "spa-1", 100.00).build())
    .addCustomFlag(GoogleAnalyticsFirebaseGA4Kit.CF_GA4COMMERCE_EVENT_TYPE, FirebaseAnalytics.Event.ADD_SHIPPING_INFO)
    .addCustomFlag(GoogleAnalyticsFirebaseGA4Kit.CF_GA4_SHIPPING_TIER, "overnight")
    .build()
MParticle.getInstance().logEvent(event)

@import mParticle_Apple_SDK;
@import mParticle_Google_Analytics_Firebase_GA4;
@import FirebaseAnalytics;

MPProduct *product = [[MPProduct alloc] initWithName:@"Spa Essentials" sku:@"spa-1" quantity:@1 price:@100.00];
MPCommerceEvent *event = [[MPCommerceEvent alloc] initWithAction:MPCommerceEventActionCheckoutOptions product:product];
[event addCustomFlag:kFIREventAddShippingInfo withKey:kMPFIRGA4CommerceEventType];
[event addCustomFlag:@"overnight" withKey:kMPFIRGA4ShippingTier];
[[MParticle sharedInstance] logEvent:event];

import mParticle_Apple_SDK
import mParticle_Google_Analytics_Firebase_GA4
import FirebaseAnalytics

let product = MPProduct(name: "Spa Essentials", sku: "spa-1", quantity: 1, price: 100.0)
let event = MPCommerceEvent(action: .checkoutOptions, product: product)
event.addCustomFlag(AnalyticsEventAddShippingInfo, withKey: kMPFIRGA4CommerceEventType)
event.addCustomFlag("overnight", withKey: kMPFIRGA4ShippingTier)
MParticle.sharedInstance().logEvent(event)

const product1 = mParticle.eCommerce.createProduct(
    'Spa Essentials',  // Name
    'spa-1',           // SKU
    100.00,            // Price
    4                  // Quantity
);
const customAttributes = {sale: true};
const customFlags = {
  'GA4.CommerceEventType': 'add_shipping_info',
  'GA4.ShippingTier': 'overnight'
};
mParticle.eCommerce.logProductAction(
  mParticle.ProductActionType.CheckoutOption,
    [product1],
    customAttributes,
    customFlags);

add_payment_info custom flag example

To map to a Firebase add_payment_info event, pass a custom flag of GA4.CommerceEventType equal to add_payment_info (Firebase provides a constant for this), and an optional custom flag of GA4.PaymentType with a string value:

import com.google.firebase.analytics.FirebaseAnalytics;

CommerceEvent event = new CommerceEvent.Builder(Product.CHECKOUT_OPTION, new Product.Builder("Spa Essentials", "spa-1", 100.00).build())
    // how to import GoogleAnalyticsFirebaseGA4Kit?  in the tests, it sows up as kitInstance
    .addCustomFlag(GoogleAnalyticsFirebaseGA4Kit.CF_GA4COMMERCE_EVENT_TYPE, FirebaseAnalytics.Event.ADD_PAYMENT_INFO)
    .addCustomFlag(GoogleAnalyticsFirebaseGA4Kit.CF_GA4_PAYMENT_TYPE, "Visa")
    .build();
MParticle.getInstance.logEvent(event);

import com.google.firebase.analytics.FirebaseAnalytics

val event = CommerceEvent.Builder(Product.CHECKOUT_OPTION, Product.Builder("Spa Essentials", "spa-1", 100.00).build())
    .addCustomFlag(GoogleAnalyticsFirebaseGA4Kit.CF_GA4COMMERCE_EVENT_TYPE, FirebaseAnalytics.Event.ADD_PAYMENT_INFO)
    .addCustomFlag(GoogleAnalyticsFirebaseGA4Kit.CF_GA4_PAYMENT_TYPE, "Visa")
    .build()
MParticle.getInstance().logEvent(event)

@import mParticle_Apple_SDK;
@import mParticle_Google_Analytics_Firebase_GA4;
@import FirebaseAnalytics;

MPProduct *product = [[MPProduct alloc] initWithName:@"Spa Essentials" sku:@"spa-1" quantity:@1 price:@100.00];
MPCommerceEvent *event = [[MPCommerceEvent alloc] initWithAction:MPCommerceEventActionCheckoutOptions product:product];
[event addCustomFlag:kFIREventAddPaymentInfo withKey:kMPFIRGA4CommerceEventType];
[event addCustomFlag:@"visa" withKey:kMPFIRGA4PaymentType];
[[MParticle sharedInstance] logEvent:event];

import mParticle_Apple_SDK
import mParticle_Google_Analytics_Firebase_GA4
import FirebaseAnalytics

let product = MPProduct(name: "Spa Essentials", sku: "spa-1", quantity: 1, price: 100.0)
let event = MPCommerceEvent(action: .checkoutOptions, product: product)
event.addCustomFlag(AnalyticsEventAddPaymentInfo, withKey: kMPFIRGA4CommerceEventType)
event.addCustomFlag("visa", withKey: kMPFIRGA4PaymentType)
MParticle.sharedInstance().logEvent(event)

const product1 = mParticle.eCommerce.createProduct(
    'Spa Essentials',  // Name
    'spa-1',           // SKU
    100.00,            // Price
    4                  // Quantity
);
const customAttributes = {sale: true}; // if not passing any custom attributes, pass null
const customFlags = {
  'GA4.CommerceEventType': 'add_payment_info',
  'GA4.PaymentType': 'credit card'
};
mParticle.eCommerce.logProductAction(
    mParticle.ProductActionType.CheckoutOption,
    [product1],
    customAttributes,
    customFlags);

Custom mappings for events

The mParticle integration for Google Analytics 4 supports optional Custom Mappings for server-to-server configurations. These optional custom mappings can help if:

• You need to forward events to GA4 that are not automatically captured by mParticle.
• You have an mParticle implementation that does not conform to mParticle’s default event schema.

Following are all event types that can be forwarded to GA4 using custom mappings. Event types that are already mapped to GA4 by default are noted in the column on the right, but remember that even events mapped by default can be mapped using a custom mapping to suit the specific needs of your implementation.

For details on which event attributes are supported by each event type, refer to the Google Analytics 4 documentation .

Event and User Attributes mapping

mParticle will map attributes to GA4 following GA4’s limitations for valid name characters rules and length limitations for parameter names and values. All invalid characters at the start of a parameter name are skipped until a valid one is found and any following invalid characters are replaced by a _. All attributes are alphabetically sorted and sent to GA4 until the limit defined per event in the GA4’s limitations docs.

Custom Metrics and Custom Dimensions

Google has revamped how custom dimensions and metrics work and are implemented in GA4. Through mParticle, send events with any event attributes as normal. In Google’s UI, you can then pick the parameter and associate it with a specific custom dimension or metric. See here for how to create custom dimensions and metrics in Google’s UI.

Session ID Attribute

If the Forward Session ID dropdown connection setting is set to a value other than None, mP will try to forward the associated value to GA4 under the session_id field. This only applies to server-to-server data. Options:

• None

  • The session_id field will not be forwarded.

• Forward Hashed Session ID

  • This corresponds with the events[x].data.session_id long field, and generally represents mP-generated session IDs.

  • Note: Due to GA4 asynchronously rejecting negative long values for the session_id field, mParticle will first transform the value before sending it. The value is hashed using sha256 and then converted into a 64 bit hex string for readability.

    • Ex: 123 -> a665a45920422f9d417e4867efdc4fb8a04a1f3fff1fa07e998e86f7f7a27ae3

• Forward Session Uuid

  • This corresponds with the events[x].data.session_uuid string field, and represents customer-defined session IDs.

Note: This generally only applies to app-based inputs (e.g. iOS, tvOS, Android).

• For Web, the above setting doesn’t apply.
• For Feed and Roku inputs, the session_id is forwarded if the given data’s platform resolves to Firebase. One example is when the Preferred Platform ID connection setting is set to the Firebase App ID, and the Firebase App details are present in the data.

Content Groups Attribute

In GA4, you can set a single Content Group as an event attribute content_group. See here for more info about the change in Content Groups between UA and GA4.

Kit-Only Functionality

Screen View Events

Kit integrations automatically invoke setScreen APIs for every screen event passed through mParticle. This is not relevant for s2s forwarding.

Screen Tracking for Single-Page Web Apps

To log page views for single-page web applications, you’ll need to use our custom flags Google.Page and Google.Title to set the url and title of the page.

mParticle.logPageView(
	"Product Detail Page",
	{ page: window.location.pathname },
	{"Google.Location": window.location.pathname,
         "Google.Title": "The title of the page"}
);

Read more about logging page views through our Web SDK here.

Consent

Google Analytics 4 has added specific Consent Mode parameters that must be sent via gtag implementation: ad_user_data, ad_personalization, ad_storage and analytics_storage when sending events.

User-specified Consent

To configure user consent forwarding under this value, a mapping should be set-up leveraging mParticle’s notion of Consent Purposes. To learn more about handling user consent within mParticle’s platform, see the following docs: Data Privacy Controls.

Once a Consent Purpose is set-up, user consent information can be associated with it in subsequent Events. The Consent Purpose data mapping can then be configured for downstream forwarding via the User Consent Data Mapping connection setting.

Consent Defaults

In the absence of a user-defined consent value for the ad_user_data, ad_personalization, ad_storage, and analytics_storage fields via the Consent Purpose mapping, a default value can be optionally configured via a separate drop-down setting for each consent type. When no user consent is provided, the default status is used, if specified. If omitted, the Unspecified status will be sent.

Caution: It is recommended that in the long term, you set up user-specified consent through the Consent Purpose mapping, such that the user consent is correctly forwarded to Google. It is your responsibility as a Data Controller to stay compliant under the GDPR, and set up user consent collection for downstream forwarding. The consent default setting may be deprecated in the future.

Consent Behavior (Web)

mParticle follows Google’s recommendations when forwarding consent state defaults and consent state updates. However, the Web kit behaves differently depending on how your default consent states are configured:

• If your consent state default is set to Denied or Granted:

  • Whenever a user changes their consent state away from the default, the Web kit forwards the new value downstream the next time a custom event or page view event is triggered.
  • Whenever a page is refreshed or the kit is re-initialized, the default consent states are automatically forwarded downstream, along with any consent state changes the user may have made.

• If your consent state default is set to Unspecified:

  • As soon as a user accepts or denies the consent purposes specified in your connection settings, the mapped consent states are set as the defaults after a page is refreshed or the kit is re-initialized. These new consent state defaults are then forwarded downstream.

Connection Settings