Video Ad Selection Using Viewer Retention Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional metrics for measuring and selecting video advertisements, such as click-through rate and pay-per-action pricing, are inadequate for assessing viewer retention, leading to ineffective advertisement placement and revenue loss for publishers and advertisers.
Innovation Solution
Determining video advertisement quality based on the capability to retain viewers, using a combination of advertisement, video, and viewer characteristics to select and position ads, and adjusting pricing accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional pay-per-action pricing policies are used to select video advertisements, then advertisement placement can be determined based on auction bids, but viewer retention and subsequent video engagement are not improved
Solution Approach 1:
The patent changes the selection parameter from traditional metrics (click-through rate, conversion rate, auction bid amount) to a new parameter: predicted viewer retention. The system estimates retention probability for each video advertisement and uses this as the primary criterion for selection, fundamentally altering how advertisements are chosen to prioritize viewer engagement over immediate conversion actions.
Solution Approach 2:
The system incorporates feedback loops where viewer retention data from previous video sessions is collected and used to refine future advertisement selections. By monitoring whether viewers continue watching after advertisements and adjusting selection strategies based on this feedback, the system continuously improves retention outcomes while maintaining revenue optimization.
2Quantity of substance
If video advertisements with high auction bids are selected for placement, then advertisement revenue is maximized in the short term, but viewer retention and engagement with subsequent videos decrease
Solution Approach 1:
The system performs preliminary estimation of viewer retention probability before selecting advertisements for placement. By predicting which advertisements are likely to be retained (viewed completely) versus which would cause viewers to drop off, the system pre-screens advertisements to ensure selected ones align with both revenue goals and viewer engagement objectives, preventing negative outcomes before they occur.
3Measurement precision
If traditional advertisement selection metrics focus on clicks and conversions, then pay-per-action pricing can be implemented, but the specific retention power of advertisements for video sessions is not measured
Solution Approach 1:
The patent adds a new dimension to advertisement measurement by introducing temporal continuity metrics. Instead of measuring only discrete actions (clicks, conversions), the system measures the continuous dimension of viewer engagement through retention probability - whether viewers continue watching the video session after the advertisement. This dimensional addition captures previously unmeasured information about advertisement impact on overall video consumption.
Data Source
AI summary
Embodiments of the present invention relate to facilitating selection of video advertisements for presentation in association with a video. In embodiments, advertisement quality associated with various video advertisements is referenced. Generally, the advertisement quality indicates a probability a viewer will continue viewing a portion of a video following presentation of the video advertisement presented in association with the video. The advertisement quality associated with the video advertisements is used to select one or more video advertisements for presentation along with the video. An indication of the selected video advertisements can be provided for integration with the video to present to the viewer.