Session Replay Capture Triggered by Events to Cut Resource Waste
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for capturing software application session replay data are inefficient, as they allocate resources to all sessions, including those not of interest, leading to wastage of computation, storage, and communication resources, and often capture duplicative data.
Innovation Solution
Techniques for selectively capturing session replay data by monitoring for specific events and triggering data collection only when these events occur, using event configuration information to identify sessions of interest and efficiently allocate resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If resources are allocated to capture data from all sessions, then complete session coverage is achieved, but resource wastage increases and duplicative data is captured
Solution Approach 1:
The patent extracts and captures only the specific session data that is relevant and useful, rather than capturing all session data. This is achieved by identifying key events and characteristics that indicate session importance, then selectively capturing data only for sessions exhibiting these characteristics, thereby eliminating wasteful capture of duplicative or irrelevant data while maintaining complete coverage of important sessions
Solution Approach 2:
The patent applies different data capture strategies to different sessions based on their local characteristics. Sessions are evaluated individually against criteria such as event types, user interactions, and session metrics, and data capture is triggered only for sessions that meet specific quality thresholds, ensuring that resources are allocated proportionally to session importance rather than uniformly across all sessions
2Loss of information
If session replay data is captured for all users, then comprehensive analytics are available, but computation and storage resources are wasted
Solution Approach 1:
The patent extracts only the essential session replay data that contributes to meaningful analytics, filtering out redundant information. By identifying key events, critical user interactions, and significant session outcomes, the system captures a condensed representation of session data that maintains analytical value while dramatically reducing the volume of stored information
Solution Approach 2:
The patent implements partial data capture by selectively recording sessions based on predefined criteria such as error events, conversion actions, or unusual user patterns. This partial action approach captures sufficient data to maintain analytics completeness for decision-making purposes while avoiding the excessive storage burden of capturing every session in detail
3Speed
If data transmission is performed for every session, then real-time monitoring is achieved, but communication bandwidth is consumed
Solution Approach 1:
The patent implements periodic evaluation of sessions against capture criteria rather than continuous data transmission. Sessions are monitored in real-time through event-driven triggers, and data transmission occurs periodically only when specific conditions are met, such as completion of key actions or detection of anomalies, maintaining monitoring responsiveness while minimizing unnecessary communication
Solution Approach 2:
The patent extracts and transmits only the essential session data that requires immediate attention or analysis, filtering out routine or redundant information before transmission. This selective extraction ensures that communication bandwidth is consumed only for transmitting valuable insights and critical session information rather than all session data
Data Source
AI summary
Described herein are techniques for capturing data from devices to generate replays of sessions of a software application executed by the devices. The techniques use data obtained from the device to determine whether certain events in which a session replay is to be generated have occurred in a session of the software application. When occurrence of such an event is detected during the session of the software application, the techniques determine to trigger collection of session replay data. In this case, the session replay data may be stored and/or transmitted from the device for storage, and subsequently used to generate a session replay. If none of the events are detected, then session replay data may not be transmitted. Accordingly, the techniques can selectively allocate resources (e.g., compute, storage, and/or communication resources) for sessions in which events are detected.