Media Recommendation Insertion Using Abandonment Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining the optimal placement of content recommendations in consumed media to avoid disrupting the user experience, as inserting recommendations too early or late can lead to user dissatisfaction.
Innovation Solution
A recommendation insertion application that generates optimal insertion points based on aggregated instances of user abandonment data and individual viewing habits, using an abandonment aggregation service, insertion point generation service, and recommendation generation service to determine the best time to present recommendations within media content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If recommendations are inserted early in media content, then recommendation visibility increases, but user experience is disrupted
Solution Approach 1:
The system performs preliminary analysis of abandonment data before inserting recommendations. By pre-determining optimal insertion points based on historical abandonment patterns, the system ensures recommendations appear at moments when users are naturally least engaged, maximizing visibility while minimizing disruption to the media experience.
Solution Approach 2:
The system continuously collects and analyzes user abandonment data to refine recommendation insertion timing. This feedback loop allows the system to learn from actual user behavior patterns and progressively optimize insertion points, balancing recommendation visibility with user experience preservation through data-driven adjustments.
2Object-affected harmful factors
If recommendations are inserted late in media content, then user experience is preserved, but recommendation visibility decreases
Solution Approach 1:
The system pre-calculates optimal insertion points by analyzing abandonment data before content consumption begins. This preliminary action identifies specific time windows where recommendations can be inserted without disrupting engaged users, ensuring both experience preservation and adequate visibility by targeting moments of natural disengagement.
Solution Approach 2:
The system dynamically adjusts the timing parameter of recommendation insertion based on analyzed abandonment patterns. By changing the insertion timing parameter to align with identified abandonment periods, the system optimizes the balance between visibility and user experience preservation without requiring manual intervention.
3Ease of manufacture
If recommendations are inserted at fixed points in media content, then implementation is simple, but effectiveness varies across different users
Solution Approach 1:
The system transitions from static fixed-point insertion to dynamic insertion based on user behavior patterns. By making insertion timing adaptive and responsive to individual abandonment data, the system maintains implementation feasibility through automated analysis while significantly improving effectiveness for diverse user groups through personalized timing.
Solution Approach 2:
The system performs preliminary analysis of user abandonment patterns to pre-determine personalized insertion points before content consumption. This preliminary customization enables effective user-specific timing without requiring complex real-time adjustments, maintaining implementation simplicity while improving reliability through data-driven personalization.
Data Source
AI summary
Disclosed are various embodiments for a recommendation insertion application. Instances of abandonment for media content are aggregated. A recommendation insertion point is calculated as a function of the instances of abandonment. A recommendation for suggested content is inserted into the media content at the recommendation insertion point.