Observability
PLuG Observability is a comprehensive toolkit for unraveling user behavior within DevRev platform. A session is captured when a user launches your application, which initializes the Session Analytics SDK. The SDK then tracks and captures all screens, events, API service calls, exceptions (such as crashes and ANRs), errors, and gestures (including scrolls, taps, and rage taps).
Session Replays
You can search for Session Replays in Explore page where you can see the sessions for your Web app and sessions for Android and iOS on mobile.
- View sessions by platform and time.
- Apply simple filters such as Platform, Session duration, Date range, etc.
- Jump into session replays for deeper investigation.
In the top-center, next to the app name, you find a toggle for Android, iOS, and Web. Use this toggle to view the data for each platform.
Playback of user interactions
Clicking a specific user replays sessions for that user, which lets you dive deep into each session in detail.
The timeline interface offers a chronological breakdown of user interactions, highlighting clicks, events, and associated properties.
For web sessions, tabs reflect the user's browsing history, while the comments section enables team collaboration and discussion.
For mobile sessions, recordings shows the users screen recording and navigation.
DevTool technical insights (Web only)
DevTool provides developers with logs, errors, and network details to diagnose and fix issues.
-
Click DevTools to the right of the playback window.
-
Below the playback window, the network information of all the web requests made and console errors from the browser are displayed in Network and Console tabs.
Session terminology
-
Session types
Frustrated sessions are those where one or more metrics indicate user frustration during the session.
Tolerating sessions are those where one or more metrics indicate users are just tolerating the experience.
Satisfied sessions are those with no tolerating or frustrated metrics, indicating a positive user experience.
-
Hot launch
When a user brings the app from the background to the foreground.
-
Cold launch
When a user opens the app for the first time and not from a minimized state.
-
Rage taps or clicks
Multiple taps or clicks on a screen element in quick succession, typically indicating user frustration with a malfunctioning element. Rage taps are identified when there are three or more taps within a 5-pixel area, each occurring within 600ms of the previous tap.
-
Dead taps or clicks
Taps or clicks on seemingly interactive elements that have no response, causing user confusion and potential frustration.
-
Active app versions
The different versions of the app currently being tracked and used by users.
Session Analytics: Web
A web session consists of the following:
- Screen recording of the application, captured by the PLuG SDK through DOM mutation tracking.
- Auto-captured events (also called Stock events), such as Click, Dead Click, Rage Click, Form Change.
- Custom events, fired during SDK integration to log specific product interactions.
- Network calls, captured and associated with the session.
- Console errors, such as uncaught exceptions or warnings.
Session lifecycle:
- A web session begins when a user opens a webpage.
- It can last for up to 4 hours, but ends if the user remains inactive for 30 minutes.
- If the user resumes activity after this inactivity period, a new session is started—even if they stay on the same page.
- A single session can include multiple tabs, representing different browser tabs opened by the user during that session.
Session Analytics: Web dashboard consists of the following dashboard widgets and tabs.
Tabs
-
Web
Widgets and visualizations for all web sessions.
-
Network
APIs with frequent errors, slow response times, and highest usage.
-
Events
Most frequently emitted custom events, and distribution of dead and rage clicks across pages.
Dashboard widgets
Each tile in the Dashboard has a download icon. Click this icon to download the data.
-
Total Sessions
The total number of user sessions recorded during the selected date range. This reflects overall traffic volume and how often the platform is being accessed.
-
Total Unique Users
The number of distinct individuals who accessed the platform in the specified time frame. This helps measure audience size and platform reach.
-
Average Sessions Per User
The average number of sessions generated by each unique user. Higher values suggest stronger engagement or repeat usage patterns.
-
Sessions Per Day
A daily breakdown of the number of sessions recorded. Useful for identifying usage patterns, peak activity days, and seasonality.
-
Most Active Verified Users
Logged-in users with the highest number of sessions. These are your most engaged users, often ideal for feedback, early feature testing, or targeted communication.
-
Most Active Customer Workspaces
Workspaces or customer accounts with the most sessions. Indicates which customers are deriving the most value from the platform and may warrant closer support or expansion opportunities.
-
Popular Pages
The most frequently visited URLs across the platform. These are typically your highest-performing pages and can reveal what users find most valuable.
-
Sessions with % Rage Clicks
The percentage of sessions that included at least one rage click—rapid repeated clicking on an element. This is a strong signal of user frustration or unresponsive UI elements.
-
Sessions with % Dead Clicks
The percentage of sessions that included at least one dead click—clicks on elements that don’t produce a response. Dead clicks often point to broken links or confusing interface design.
-
Average Rage Clicks Per Session
The average number of rage clicks made during a session. A rising average may indicate a growing UX issue or user dissatisfaction.
-
Average Dead Clicks Per Session
The average number of dead clicks per session. This metric helps surface broader UI reliability issues even when individual dead clicks are minor.
-
URLs with Rage Clicks
Pages where users frequently rage click. These pages likely contain frustrating or misleading elements that require UX improvements.
-
URLs with Dead Clicks
Pages where users often click on non-responsive elements. These highlight parts of the interface that may not be working as expected.
-
Sessions with % JS Errors
The percentage of sessions that experienced at least one JavaScript error. A high percentage may suggest performance instability or poor code execution affecting user experience.
-
JS Errors
The most frequently occurring JavaScript errors across all sessions. These errors often lead to broken functionality and should be prioritized for engineering fixes.
-
URLs with High Response Times
Pages with the slowest loading or response times. These may be causing user drop-off or dissatisfaction due to delayed interactions.
-
Platform Distribution
A breakdown of sessions by device type (desktop, mobile, tablet). Understanding this helps with responsive design prioritization and tailoring the experience across platforms.
-
OS Distribution
Sessions categorized by operating system, such as Windows, macOS, Android, or iOS. This helps QA and development teams optimize and test for the right environments.
-
Browser Distribution
Sessions segmented by browser type (for example, Chrome, Safari, Firefox). This helps ensure compatibility and performance across the most-used browsers by your users.
Session Analytics: Mobile
A mobile session includes:
- Screen recording, captured by the PLuG SDK through periodic screenshots of the user interface.
- Auto-captured events, such as Tap, Rage Tap, Swipe, etc.
- Custom events, triggered through SDK integration.
- Network calls
Session lifecycle:
- A mobile session starts when a user opens the app.
- Unlike web sessions, there is no time limit on session length.
- A session contains multiple recordings:
- A recording ends when the app is backgrounded.
- A new recording starts when the app comes to the foreground again.
- The session ends when the app is killed, or it crashes, or it becomes non-responsive, or after 30 mins of inactivity.
- Sessions are uploaded the next time the app is launched.
Session Analytics: Mobile dashboard consists of the following dashboard widgets and tabs.
Tabs
-
Sessions
Widgets and visualizations for all the mobile sessions.
-
Screens
Top screens and screens with crashes, errors, rage taps, and ANR.
-
Events
Most frequently emitted custom events across screens.
-
APIs
APIs that frequently error out, respond slowly, and are most popular.
Dashboard widgets
Each tile in the Dashboard has a download icon. Click this icon to download the data.
-
Total Sessions
The total number of recorded user sessions within the selected period, showing overall platform usage and activity levels.
-
Total Distinct Users
The number of unique individuals who initiated at least one session. This reflects the reach of your platform during the timeframe.
-
Average First User Interaction Time
The average time taken by users to make their first interaction (for example, tap, scroll) after a session begins. A shorter time indicates faster user engagement; a longer time may suggest loading delays or poor onboarding flow.
-
Average Session Duration (in seconds)
The mean duration of user sessions, measured in seconds. Longer durations often signal higher user engagement, while extreme values (for example, 8M+ seconds) may suggest a tracking issue or inactive sessions left open.
-
Session % with Crashes
The percentage of sessions that ended due to an app crash. Ideally close to zero, this metric helps track overall app stability.
-
Session % with Rage Taps
The percentage of sessions where users tapped on the same spot—typically out of frustration. High values may indicate unresponsive elements or confusing design patterns.
-
Session % with UI Freeze
The percentage of sessions in which the UI froze or became unresponsive. Even a small percentage here warrants performance optimization to prevent churn.
-
Sessions Per Day
A day-by-day count of sessions across the selected date range. Helps visualize usage trends, spikes, or lulls in activity.
-
App Version Trend
Session volume by app version over time. This helps monitor the adoption of new releases and detect any performance regressions or bugs introduced in specific versions.
-
Exceptions Time Trend
The frequency of app exceptions over time. This helps identify whether error rates are rising and if they correlate with new updates or feature rollouts.
-
Most Active Users
A ranked list of users (by anonymized ID) based on session count. These are your most engaged users and can be valuable for feedback loops or early feature testing.
-
Most Active App Versions
Shows which app versions are used most frequently. This is crucial for targeting QA efforts, rolling out fixes, or analyzing behavior differences across versions.
-
Average App Launch Time (Cold & Hot)
Measures average launch time in milliseconds for cold (first-time) and hot (relaunch) starts. Lower times contribute to a better user experience, especially in high-frequency use apps.
-
App Launch Distribution (Hot vs Cold)
The breakdown of app launches by type—hot or cold. Helps understand how frequently users return to the app and informs caching or load strategy.
-
Network Type Distribution
The distribution of sessions by network type (WiFi, 4G, 5G, etc.). Helps assess real-world app performance across varying network conditions.
-
Device Manufacturer Distribution
The share of sessions by device manufacturer (Apple, Google, Samsung, etc.). Useful for ensuring device compatibility and prioritizing testing resources.
-
OS Version Distribution
Distribution of user sessions by operating system version. Essential for tracking legacy usage, planning deprecations, and debugging OS-specific issues.
