Overview

The Device Analytics feature provides a granular analysis of push notification delivery and engagement. Unlike standard analytics, which focus on the overall campaign funnel, Device Analytics investigates how performance metrics vary across specific device attribute segments after a gateway (FCM or APNS) accepts the notification.

Device Analytics allows you to diagnose delivery issues, set realistic campaign expectations, and gain granular insights into the device landscape of your audience.

Use Cases

• Diagnose performance variations: Identify specific hardware vendors (OEMs) or OS versions that cause delivery drops. For instance, pinpoint if aggressive battery optimization settings on specific Android OEMs prevent notification display.
• Set realistic expectations: Analyze user activity (Device Recency) to interpret delivery rates based on the actual reachability of the user base.
• Understand app issues: Monitor performance across specific app version groups to identify and diagnose regressions or stability problems.
• Diagnose browser-specific issues: Identify if a specific browser (for example, Mozilla Firefox or Apple Safari) is experiencing lower delivery or click-through rates compared to others, which may indicate rendering issues or browser-specific restrictions.

Analyze Device Data

Step 1: Access and Filter

1. On the left navigation of the MoEngage dashboard, select Engage > Campaigns.
2. On the All Campaigns page, click the row that displays the information about the campaign for which you wish to see the analytics information.
3. On the Campaign Analytics and Info page, select the Device analytics tab.
   
   
4. On the Device Analytics tab, configure the following values:
   • Date Range: Select the time period for analysis.
   • Attribution Type: Select the conversion goal attribution model (for example, Click-through attribution).
   • Platform: Choose the preferred platform from the drop-down list.

Step 2: Analyze Performance of Factors

The dashboard displays widgets to visualize performance.

• Pie charts represent Sent counts, while horizontal bar charts represent Delivery Rate, Click-through Rate (CTR), and Conversion Rate (CVR).
• Click any chart to open a detailed chart/table view that breaks down Sent, Impressions, Clicks, and Conversions for the selected factor. For example, for the OEM factor, the detailed chart view appears as shown in the image below.
  

MoEngage tracks the following factors based on the selected platform:

Factor	Platform	Description	Image
Device Recency	All	Categorizes users based on how recently the device accessed the background state and sent data to MoEngage. Note: The Active 29+ days bucket represents dormant users; expect low delivery rates for this segment. The Active 0–14 days bucket represents highly active users and typically shows high performance.
OEM	Android only	Categorizes performance by the device manufacturer (for example, Samsung, Xiaomi, or Vivo). Note: You can override this default OEM list in the settings page. For more information, refer to Device Analytics Settings and Configuration. Analysis of OEMs helps identify whether specific manufacturers aggressively terminate background processes, which can lead to lower delivery rates. Different OEMs have different views for notification displays, which may affect CTRs. For more information, refer to Push Templates FAQs.
App Versions	Android and iOS	Tracks performance across specific releases of an application (Grouped into buckets like Major version 1, Major version 2). Note: You can use this to correlate drops in CTR or Delivery Rate with specific app updates.
OS Versions	Android and iOS	Displays how campaign metrics vary across different Operating System versions (for example, Android 13 versus Android 15). Note: Features such as the notification AI summaries introduced in iOS 18 and the advanced focus controls and Liquid Glass UI introduced in iOS 26 may impact how your users interact with push notifications.
Opt-in Status	iOS only	Categorizes performance based on the notification permission level: Authorized: Explicitly agreed to receive notifications. Provisional: Provisional trial (notifications delivered quietly). Note: Provisional notifications are likely to have a low CTR, as they are delivered quietly and do not display actions. Others: Status is unknown.
Browser	Web only	Categorizes performance by the web browser used to receive the notification. Values: Google Chrome, Apple Safari, Mozilla Firefox, Microsoft Edge, Opera, and Others. Note: Use this to identify browser-specific delivery failures. You can configure which browsers to track in the Device Analytics settings.

Export Data

Use one of the following options to export data for offline analysis:

• Download Chart: Downloads the specific visual graph currently displayed.
• Download CSV: Downloads the data powering the currently visible widget.
• Download All Metrics: Provides a granular export of all combinations. This enables full-funnel analysis for specific cohorts (for example, Xiaomi users on App Version Bucket 1 with Recency 0-14 days).
  

Step 3: Heatmap Analysis

The Heatmap widget facilitates cross-factor analysis by simultaneously visualizing two factors against one metric (for example, Device Recency versus OEM for Delivery Rate).
By default, the analysis factors are Device Recency versus OEM for the Delivery Rate metrics.

The table generates a color-coded gradient:

• Green: Indicates the highest-performing cohorts.
• Red: Indicates the lowest-performing cohorts. Use this to find failures in expected high-performance groups.

[Android Heatmap analysis]

[iOS Heatmap analysis]

[Web Heatmap analysis]

FAQs