Firebase Traffic Source Attribution Guide

Conversion events in Firebase analytics measure how well your app fulfills your target objectives. It represents a completed user activity that contributes to the success of your business. Examples of conversion events include making a purchase (buying a ticket) or adding events to a bookmark.

By default, the following events are set as conversion events:

• first_open
• in_app_purchase
• app_store_subscription_convert
• app_store_subscription_renew
• purchase

Custom Conversion Events

We can also add our own events as conversion events which can allow us to measure how our marketing efforts contributed to important business goals.

To enable or disable an event as a conversion:

STEP 01

Navigate to Analytics.

Go to the Events section under Analytics in the Firebase console.

Firebase navigation panel, by EmCube

STEP 02

Mark as conversion.

Turn on the switch in the 'Mark as conversion' column in the row for the event that you want to set as the conversion event.

Firebase conversion events, by EmCube

Click on any one of your conversion events in the conversion section of the right panel to view the traffic sources given credit for those conversions. You'll see a trendline of how frequently that event has occurred, along with a traffic source breakdown below it.

The Attribution Models

Attribution models are a set of rules that define which marketing activity gets credited for the conversion. The following attribution models are used to attribute the traffic source to the conversion events.

In addition to choosing an attribution model, you can also choose to see the origin of the events by Source, Medium, and Campaign, Network, Adgroup, and Creative.

Campaign Type in Firebase Analytics

Every marketing effort is targeted toward either making users download the application or re-engage them for an important business action. That is, an install-campaign or a re-engagement campaign.

Install Campaigns

The goal of an install campaign is to drive new installations of your app. Users who do not already have your app on their device are targeted and routed to the appropriate app store to download and install the app.

The attribution window for install campaigns is 30 days, and it is based on the first_open event, tracked by Firebase automatically.

If a first_open event is attributed to a campaign, then all other conversion events moving forward are attributed to that same campaign until the attribution period expires in one year. The attribution window is renewed for one year with each subsequent conversion that occurs before expiration.

Note: You cannot adjust attribution windows in Firebase.

Re-Engagement Campaigns

Re-engagement campaigns are designed to encourage customers to continue using your app after they have installed it. A conversion from a re-engagement campaign gets attributed twice:

1. One to the campaign that got credit for the first_open event.
2. Second to the re-engagement campaign.

Re-engagement campaigns can also send the user to specific screens of the application, to show them certain promotions or cross-sell them products using dynamic links. We can easily create dynamic links in the Firebase console.

Traffic Sources & Channels

Unlike Google Analytics, the Analytics console in Firebase doesn't have channel grouping. Instead, source, medium, and campaign values are stored and viewed directly in Firebase conversion reports. Following are some of the common source mediums you will find in Firebase analytics:

Campaign Traffic Information

Firebase can automatically capture traffic from only a limited number of sources. Everything else lands in Direct / None. The old fix was to carry UTMs into the app with Firebase Dynamic Links and read them at runtime with getDynamicLink() — but that path no longer exists.

There are two reliable ways to attribute traffic for all sources today. Which one fits depends on whether you already run a mobile measurement partner (MMP).

Option 1. Route your MMP attribution into GA4 / Firebase

If you already use an MMP (AppsFlyer, Adjust, Singular), the cleanest path is to let it forward attributed installs and events straight into your Firebase project. Firebase Analytics is a GA4 property, so once the two are connected, media source, medium, and campaign show up alongside your conversion events — for every source the MMP resolves, not just the handful Firebase handles on its own.

// 1. Read the GA4 / Firebase app instance id
FirebaseAnalytics.getInstance(context).appInstanceId
    .addOnSuccessListener { appInstanceId ->
        // 2. Hand it to your MMP so its events attach to the same GA4 user.
        //    (exact call per your MMP's Firebase integration guide)
        AppsFlyerLib.getInstance()
            .setAdditionalData(hashMapOf("firebase_app_instance_id" to appInstanceId))
    }

Store these somewhere you can reuse across environments — you'll need them again for CI, iOS, and any Measurement Protocol calls:

Note: How much of this data actually flows through depends on each partner's data-sharing policy. Self-attributing networks like Meta, Google, and TikTok restrict how granular the attribution they share with third parties can be, so some campaigns will arrive aggregated — or not at all — even when the integration is set up correctly.

Option 2. Capture UTMs with OneLink and push them to Firebase yourself

Every major MMP offers the same building block — a deep-linking product plus an SDK callback that hands you the attribution data: AppsFlyer has OneLink, Adjust has Adjust Links, Singular has Singular Links, and Branch has Branch Deep Links. The mechanics below are identical whichever you use; we'll walk through it with AppsFlyer OneLink.

This is the closest replacement for the old Dynamic Links flow, and the right fit when you want to own the event that carries the UTMs. AppsFlyer's OneLink takes over the deep-linking job Dynamic Links used to do; the AppsFlyer SDK hands you the attribution data, and you log it to Firebase exactly like before.

// Direct deep link — user tapped a OneLink
AppsFlyerLib.getInstance().subscribeForDeepLink(object : DeepLinkListener {
    override fun onDeepLinking(result: DeepLinkResult) {
        if (result.status != DeepLinkResult.Status.FOUND) return
        val dl = result.deepLink
        val params = Bundle().apply {
            putString(FirebaseAnalytics.Param.SOURCE,   dl.getStringValue("utm_source"))
            putString(FirebaseAnalytics.Param.MEDIUM,   dl.getStringValue("utm_medium"))
            putString(FirebaseAnalytics.Param.CAMPAIGN, dl.getStringValue("utm_campaign"))
        }
        firebaseAnalytics.logEvent("campaign_details", params)
    }
})

// Deferred deep link / install attribution
AppsFlyerLib.getInstance().init(DEV_KEY, object : AppsFlyerConversionListener {
    override fun onConversionDataSuccess(data: MutableMap<String, Any>) {
        val params = Bundle().apply {
            putString(FirebaseAnalytics.Param.SOURCE,   data["media_source"] as? String)
            putString(FirebaseAnalytics.Param.CAMPAIGN, data["campaign"] as? String)
        }
        firebaseAnalytics.logEvent("campaign_details", params)
    }
    override fun onConversionDataFail(error: String?) {}
    override fun onAppOpenAttribution(data: MutableMap<String, String>?) {}
    override fun onAttributionFailure(error: String?) {}
}, context)
AppsFlyerLib.getInstance().start(context)

Worth bookmarking: the AppsFlyer OneLink docs for building the links, and the AppsFlyer SDK deep-linking guide for the conversion and deep-link listeners.

End Node

How traffic is attributed in any analytics platform is key to understanding how users get to our application. Conversion events in Firebase analytics help to store and analyze traffic sources. Hence, conversion events that are important to the business should be turned on to understand marketing efforts for that conversion.

Also, Firebase itself doesn't track traffic attributes from unknown sources. However, we can send the UTM parameters as event parameters to capture traffic attributes from unknown sources. Doing so not only lets us monitor known traffic but also allows us to monitor unknown traffic sources in our analytics solution.