Video Reach Estimation via User Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Content providers face challenges in predicting the number of video views for their budgeted campaigns on online systems, lacking clear insights into how much to spend to reach a sufficient audience effectively.
Innovation Solution
The online system employs a method to estimate a total video reach count by sampling users based on historical data and machine learning models, allowing content providers to determine the necessary budget for guaranteed video views by extrapolating the sampled video reach count to the target audience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional systems present videos to users without view prediction, then videos can be presented to users, but content providers cannot determine if they will receive an acceptable number of views for their budget
Solution Approach 1:
The system performs view count prediction before actually presenting the video to users. By calculating the predicted total video reach count in advance using sampling and machine learning models, the system enables content providers to make informed budget decisions before committing resources, thus resolving the contradiction between measurement precision and system complexity.
Solution Approach 2:
Instead of analyzing all users in the target audience, the system uses sampling to select a subset of users for analysis. This partial action approach provides sufficiently accurate predictions while significantly reducing computational complexity, allowing the system to maintain both measurement precision and manage system complexity.
2Quantity of substance
If content providers increase budget to reach more users, then audience reach increases, but it becomes difficult to determine the optimal spending amount
Solution Approach 1:
The system provides feedback to content providers by delivering predicted total video reach count estimates based on their budget parameters. This feedback loop enables content providers to understand the relationship between budget allocation and expected audience reach, allowing them to optimize their spending to achieve desired reach targets without wasting resources.
Solution Approach 2:
By calculating predicted reach counts before campaign launch, the system enables content providers to plan their budget allocation in advance. This preliminary information allows providers to determine the optimal spending amount needed to reach their target audience size, eliminating the uncertainty about whether their budget will achieve acceptable results.
3Measurement precision
If the system analyzes all users in the target audience, then prediction accuracy improves, but computational resources and time increase significantly
Solution Approach 1:
The system analyzes only a sampled subset of users from the target audience rather than all users. This partial analysis approach maintains sufficient prediction accuracy for budget planning purposes while dramatically reducing computational time and resources required, thus resolving the contradiction between measurement precision and time consumption.
Solution Approach 2:
The system creates a sample copy of the target audience population for analysis purposes. By working with this representative subset rather than the complete population, the system achieves adequate prediction accuracy while minimizing computational overhead and time requirements.
Data Source
AI summary
An online system receives a video-presentation request from a third party system. The video-presentation request comprises a video and a target audience specification for the video. The online system selects a plurality of users as the target audience of the video based on the target audience specification. From the target audience, the online system generates a sample subset of users and determines a sampled video reach count for the subset of users. A sample user who, as determined by the online system, would have viewed the view for at least a threshold duration, is included in the sampled video view count. The online system estimates a total video reach count for target audience by extrapolating the sampled video reach count for the subset of users to the target audience. The total video reach count can be used to determine a parameter for presenting the video on an online system.