Video Ad Ranking via Play-Through Rate Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online advertising systems struggle to effectively evaluate and rank video advertisements, as current metrics like Cost-Per-Action do not accurately reflect the performance of video ads, leading to poor user experience and low engagement.
Innovation Solution
A system that identifies performance features of video advertisements, uses a trained video advertisement performance model to estimate play-through rates, and ranks ads based on these estimates to improve their visibility and engagement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional Cost-Per-Action metrics are used to evaluate video advertisements, then advertiser ROI can be calculated, but the metrics do not accurately reflect video ad performance and lead to poor user experience
Solution Approach 1:
The patent changes the evaluation parameter from traditional Cost-Per-Action metrics to play-through rate metrics. By measuring the actual viewing behavior (completion rate, mid-roll engagement) rather than just conversion actions, the system achieves more accurate video ad performance measurement that directly correlates with user experience quality
Solution Approach 2:
The patent replaces the mechanical tracking of conversion actions with automated video analytics that measure actual viewing engagement. Machine learning models analyze video playback data, user interactions, and engagement patterns to evaluate ad performance, substituting manual conversion tracking with automated behavioral analysis
2Productivity
If video advertisements with high relevance to user queries are served, then click-through rates improve, but video quality may be poor leading to low play-through rates
Solution Approach 1:
The patent implements a feedback loop where play-through rate metrics are measured and fed back into the ad selection process. Video advertisements that achieve high play-through rates are prioritized for future servings, while low play-through rate videos are deprioritized. This creates a self-correcting system that continuously improves video quality selection based on actual user engagement data
Solution Approach 2:
The patent applies preliminary filtering and ranking of video advertisements based on predicted play-through rates before they are served to users. The system pre-evaluates video quality and engagement potential using historical data and machine learning models, selecting only high-quality videos that are likely to achieve high play-through rates, thus preventing poor quality videos from being served in the first place
3Loss of energy
If advertisers bid on CPC basis with target CPA, then marketing costs are controlled, but the approach does not account for video ad specific performance factors
Solution Approach 1:
The patent creates a multi-functional evaluation system that integrates both traditional ROI metrics (CPA, conversion tracking) and video-specific metrics (play-through rate, engagement quality) into a unified ad selection framework. The system can simultaneously optimize for cost efficiency while adapting to video ad performance characteristics, providing a versatile evaluation approach that handles both general advertising goals and video-specific quality requirements
Data Source
AI summary
A request for video advertisements is received and video advertisements that can be provided in response to the request are identified. Performance features associated with the video advertisements are identified and are provided as input to a video advertisement performance model trained to estimate a play through rate for each of the video advertisements. The video advertisements are ranked based on the estimated play through rates for the video advertisements provided by the video advertisement performance model, and one or more video advertisements are provided in response to the request according to the ranking of the video advertisements.