Dynamic Content Format Selection Based on Abandonment Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Video content websites face challenges in determining the optimal format and timing for presenting additional content, such as advertisements, to users without disrupting their viewing experience, leading to potential session abandonment and revenue loss.

Innovation Solution

A data-driven approach using a content presentation system that predicts user interest and likelihood of session abandonment, selecting the least intrusive content format and timing based on historical and current session data, and user behavior analysis to dynamically customize content delivery.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If additional content is presented to users during video viewing, then revenue from content sponsors is improved, but user session abandonment increases

Engineering Contradiction:

Improverevenue lossVSAvoidsession abandonment

Core Design Contradiction:

Loss of energyVSReliability

Solution Approach 1:

The system dynamically changes content format parameters (overlay, interstitial, companion) and timing parameters (pre-roll, mid-roll, post-roll) based on predicted user abandonment likelihood, optimizing the balance between revenue generation and user retention

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The content delivery system transitions from static rule-based presentation to dynamic prediction-driven presentation, continuously adapting content format and timing based on real-time user behavior analysis and abandonment probability assessments

Inventive Principle:

Principle #15Dynamics

2Loss of energy

If content is presented in intrusive formats, then sponsor revenue is improved, but user satisfaction deteriorates

Engineering Contradiction:

Improverevenue lossVSAvoiduser satisfaction

Core Design Contradiction:

Loss of energyVSEase of operation

Solution Approach 1:

The system applies different content presentation qualities to different user contexts by selecting specific formats (less intrusive overlay or companion for high-value users, more intrusive interstitial for low-engagement users) based on individual user behavior patterns and predicted abandonment likelihood

Inventive Principle:

Principle #3Local quality

3Device complexity

If content is presented without prediction, then system complexity is reduced, but revenue optimization deteriorates

Engineering Contradiction:

Improvesystem complexityVSAvoidrevenue loss

Core Design Contradiction:

Device complexityVSLoss of energy

Solution Approach 1:

The system performs preliminary analysis of user behavior patterns and predicts abandonment likelihood before content delivery decisions are made, enabling proactive optimization of content presentation strategy rather than reactive rule-based approaches

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9451306B2Selecting content formats for additional content to be presented along with video content to a user based on predicted likelihood of abandonment

Publication Date: 2016.09.20 GOOGLE LLC

AI summary

Methods, systems, and computer program products are provided for selecting content formats based on predicted user interest. One example method includes receiving a request to present additional content to a user in association with the user viewing a video content item during a session, identifying one or more candidate content formats, predicting a likelihood that the user will abandon the session for each candidate content format, selecting a format based at least in part on the predicting, determining when to present the additional content to the user, and presenting the additional content in accordance with the selected format.