移动端埋点与数据分析深度实战：从Firebase-Analytics到自定义事件与漏斗分析全解析

引言

移动应用的数据分析驱动产品决策——用户行为追踪、漏斗分析、留存分析、AB测试效果。Firebase Analytics提供免费基础分析，自定义事件实现精细埋点。本文从移动数据分析出发，系统讲解事件设计、Firebase Analytics集成、自定义埋点、漏斗分析、隐私合规。

一、事件设计

// 事件设计原则
// 1. 标准事件：Firebase预定义（view_item, add_to_cart, purchase等）
// 2. 自定义事件：业务特定（lesson_completed, level_up等）
// 3. 事件参数：附加数据（item_id, value, currency等）
// 4. 用户属性：用户分群（user_type, subscription_status等）

class AnalyticsTracker @Inject constructor(
    private val firebase: FirebaseAnalytics
) {
    // 标准事件
    fun trackViewItem(itemId: String, itemName: String, price: Double) {
        val bundle = Bundle().apply {
            putString(FirebaseAnalytics.Param.ITEM_ID, itemId)
            putString(FirebaseAnalytics.Param.ITEM_NAME, itemName)
            putDouble(FirebaseAnalytics.Param.VALUE, price)
            putString(FirebaseAnalytics.Param.CURRENCY, "USD")
        }
        firebase.logEvent(FirebaseAnalytics.Event.VIEW_ITEM, bundle)
    }
    
    fun trackAddToCart(itemId: String, quantity: Int, price: Double) {
        val bundle = Bundle().apply {
            putString(FirebaseAnalytics.Param.ITEM_ID, itemId)
            putLong(FirebaseAnalytics.Param.QUANTITY, quantity.toLong())
            putDouble(FirebaseAnalytics.Param.VALUE, price * quantity)
            putString(FirebaseAnalytics.Param.CURRENCY, "USD")
        }
        firebase.logEvent(FirebaseAnalytics.Event.ADD_TO_CART, bundle)
    }
    
    fun trackPurchase(orderId: String, items: List<PurchaseItem>, total: Double) {
        val itemsBundle = items.map { item ->
            Bundle().apply {
                putString(FirebaseAnalytics.Param.ITEM_ID, item.id)
                putString(FirebaseAnalytics.Param.ITEM_NAME, item.name)
                putLong(FirebaseAnalytics.Param.QUANTITY, item.quantity.toLong())
                putDouble(FirebaseAnalytics.Param.PRICE, item.price)
            }
        }
        
        val bundle = Bundle().apply {
            putString(FirebaseAnalytics.Param.TRANSACTION_ID, orderId)
            putDouble(FirebaseAnalytics.Param.VALUE, total)
            putString(FirebaseAnalytics.Param.CURRENCY, "USD")
            putParcelableArray(FirebaseAnalytics.Param.ITEMS, itemsBundle.toTypedArray())
        }
        firebase.logEvent(FirebaseAnalytics.Event.PURCHASE, bundle)
    }
    
    // 自定义事件
    fun trackLessonCompleted(lessonId: String, duration: Long, score: Int) {
        val bundle = Bundle().apply {
            putString("lesson_id", lessonId)
            putLong("duration_seconds", duration)
            putInt("score", score)
            putBoolean("passed", score >= 60)
        }
        firebase.logEvent("lesson_completed", bundle)
    }
    
    fun trackScreenView(screenName: String, screenClass: String) {
        val bundle = Bundle().apply {
            putString(FirebaseAnalytics.Param.SCREEN_NAME, screenName)
            putString(FirebaseAnalytics.Param.SCREEN_CLASS, screenClass)
        }
        firebase.logEvent(FirebaseAnalytics.Event.SCREEN_VIEW, bundle)
    }
    
    // 用户属性
    fun setUserProperties(user: User) {
        firebase.setUserProperty("user_type", user.type.name)
        firebase.setUserProperty("subscription_status", user.subscription.name)
        firebase.setUserProperty("registration_date", user.createdAt.toString())
        
        val uid = user.id.toString()
        firebase.setUserId(uid)
        firebase.setUserProperty("uid", uid)
    }
    
    // 用户细分
    fun trackUserSegment(segment: String, value: String) {
        firebase.setUserProperty(segment, value)
    }
}

二、自动页面追踪

// Compose自动页面追踪
@Composable
fun TrackScreenView(screenName: String) {
    val analytics = LocalAnalytics.current
    
    LaunchedEffect(screenName) {
        analytics.trackScreenView(screenName, screenName)
    }
}

// 使用
@Composable
fun HomeScreen() {
    TrackScreenView("Home")
    // 页面内容
}

// Navigation自动追踪
@Composable
fun AppNavHost() {
    val navController = rememberNavController()
    
    NavHost(navController, startDestination = "home") {
        composable("home") {
            TrackScreenView("Home")
            HomeScreen()
        }
        composable("user/{id}") { entry ->
            val id = entry.arguments?.getString("id")
            TrackScreenView("UserDetail_$id")
            UserDetailScreen(id)
        }
    }
    
    // 导航监听
    DisposableEffect(navController) {
        val listener = NavController.OnDestinationChangedListener { _, destination, _ ->
            // Firebase自动追踪
            Firebase.analytics.logEvent(FirebaseAnalytics.Event.SCREEN_VIEW, bundleOf(
                FirebaseAnalytics.Param.SCREEN_NAME to destination.route,
                FirebaseAnalytics.Param.SCREEN_CLASS to destination.route
            ))
        }
        navController.addOnDestinationChangedListener(listener)
        onDispose { navController.removeOnDestinationChangedListener(listener) }
    }
}

三、漏斗分析

// 漏斗：用户从注册到购买的转化路径
// 注册→完善资料→浏览商品→加入购物车→结算→支付→完成

class FunnelTracker @Inject constructor(
    private val analytics: AnalyticsTracker
) {
    fun trackFunnelStep(step: FunnelStep, params: Map<String, Any> = emptyMap()) {
        val bundle = bundleOf(
            "funnel_name" to "registration_to_purchase",
            "funnel_step" to step.name,
            "step_order" to step.order,
            *params.toList().toTypedArray()
        )
        Firebase.analytics.logEvent("funnel_step", bundle)
    }
}

enum class FunnelStep(val order: Int) {
    REGISTER(1),
    COMPLETE_PROFILE(2),
    BROWSE_PRODUCTS(3),
    ADD_TO_CART(4),
    CHECKOUT(5),
    PAYMENT(6),
    PURCHASE_COMPLETE(7)
}

// 使用
class CheckoutViewModel @Inject constructor(
    private val funnel: FunnelTracker
) : ViewModel() {
    
    fun startCheckout() {
        funnel.trackFunnelStep(FunnelStep.CHECKOUT, mapOf(
            "cart_total" to cart.total,
            "item_count" to cart.items.size
        ))
    }
    
    fun completePayment(orderId: String) {
        funnel.trackFunnelStep(FunnelStep.PURCHASE_COMPLETE, mapOf(
            "order_id" to orderId,
            "payment_method" to "credit_card"
        ))
    }
}

四、埋点中间件

// 埋点中间件：统一管理所有埋点
class AnalyticsMiddleware(
    private val providers: List<AnalyticsProvider>
) {
    fun track(event: String, params: Map<String, Any>) {
        // 1. 数据清洗
        val cleaned = sanitize(params)
        
        // 2. 多平台分发
        providers.forEach { provider ->
            provider.track(event, cleaned)
        }
        
        // 3. 调试日志
        if (BuildConfig.DEBUG) {
            Log.d("Analytics", "$event: $cleaned")
        }
    }
    
    private fun sanitize(params: Map<String, Any>): Map<String, Any> {
        return params.mapKeys { (_, v) ->
            // 移除PII（个人身份信息）
            v.toString().replace(Regex("\\b[\\w.-]+@[\\w.-]+\\.[a-zA-Z]{2,}\\b"), "[EMAIL]")
             .replace(Regex("\\b\\d{11}\\b"), "[PHONE]")
        }
    }
}

interface AnalyticsProvider {
    fun track(event: String, params: Map<String, Any>)
    fun setUserProperty(key: String, value: String)
    fun setUserId(id: String)
}

// Firebase实现
class FirebaseAnalyticsProvider(
    private val firebase: FirebaseAnalytics
) : AnalyticsProvider {
    override fun track(event: String, params: Map<String, Any>) {
        firebase.logEvent(event, bundleOf(*params.toList().toTypedArray()))
    }
    override fun setUserProperty(key: String, value: String) {
        firebase.setUserProperty(key, value)
    }
    override fun setUserId(id: String) {
        firebase.setUserId(id)
    }
}

// Amplitude/Sentry实现
class AmplitudeProvider(...) : AnalyticsProvider {
    override fun track(event: String, params: Map<String, Any>) {
        amplitude.logEvent(event, params)
    }
    // ...
}

五、隐私合规

// GDPR/CCPA隐私合规
class PrivacyManager(context: Context) {
    private val prefs = context.getSharedPreferences("privacy", MODE_PRIVATE)
    
    fun hasAnalyticsConsent(): Boolean {
        return prefs.getBoolean("analytics_consent", false)
    }
    
    fun setAnalyticsConsent(granted: Boolean) {
        prefs.edit().putBoolean("analytics_consent", granted).apply()
        
        // Firebase
        Firebase.analytics.setAnalyticsCollectionEnabled(granted)
        
        // Crashlytics
        Firebase.crashlytics.isCrashlyticsCollectionEnabled = granted
        
        // Amplitude
        // amplitude.trackingEnabled = granted
    }
    
    // 数据删除请求
    fun requestDeletion() {
        // 删除Firebase Analytics用户数据
        // amplitude.regenerateDeviceId()
        // 后端API：DELETE /users/me/analytics
    }
    
    // IDFA/AAID管理
    fun getAdvertisingId(): String? {
        return if (hasAdvertisingConsent()) {
            // 仅在用户同意时获取
            Firebase.installations.id
        } else null
    }
}

六、总结

移动数据分析驱动产品决策。事件设计遵循标准+自定义组合——Firebase标准事件覆盖电商常见场景（view_item/add_to_cart/purchase），自定义事件覆盖业务特定场景（lesson_completed/level_up）。自动页面追踪通过Navigation监听自动记录屏幕浏览。漏斗分析追踪转化路径每一步——注册→完善资料→浏览→加购→结算→支付→完成。埋点中间件统一管理多平台分发——Firebase Analytics（免费基础）+ Amplitude（高级分析）同时上报。隐私合规遵循GDPR/CCPA——用户同意后才收集数据、支持数据删除请求、PII（邮箱/电话）脱敏。生产最佳实践：标准事件覆盖电商全流程、自定义事件参数有明确命名规范、多平台分发（Firebase+Amplitude/Sentry）、PII数据脱敏、隐私合规（同意机制+数据删除）、调试模式验证埋点正确。数据分析的核心理念是"数据驱动产品"——不是凭感觉判断用户行为，而是用数据验证假设。好的埋点应该：事件命名规范（snake_case）、参数完整（每事件有足够的上下文）、不遗漏关键步骤、不过度埋点（关键事件而非每个按钮）、隐私合规（用户同意后才收集）。数据分析的最终目标是"理解用户行为、优化用户体验、提升业务指标"——漏斗分析找出转化瓶颈、留存分析发现流失原因、AB测试验证改进效果。