Audience Size Projection Using Panel Cohorts to Reduce Measurement Bias
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audience measurement systems relying on return path data (RPD) and automatic content recognition (ACR) data face inaccuracies due to incomplete data collection from households with non-RPD and non-ACR devices, leading to biased audience metrics.
Innovation Solution
Utilize panel data from enrolled households with media monitoring devices to determine projection cohorts and stratification groups, comparing audience sizes between RPD/ACR-capable and non-capable devices, and projecting audience sizes accurately using projection determination circuitry to minimize bias.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If audience measurement systems rely only on RPD and ACR data from households with these capabilities, then data collection is simplified, but measurement precision deteriorates due to incomplete coverage and biased metrics
Solution Approach 1:
The patent segments the audience measurement approach into three distinct components: (1) RPD/ACR data from capable devices, (2) panel data from enrolled households with monitoring devices, and (3) projection cohorts that bridge the gap between these sources. This segmentation allows each data source to be processed differently and combined to achieve comprehensive coverage without sacrificing measurement precision
Solution Approach 2:
The patent introduces projection cohorts as an intermediary mechanism that connects RPD/ACR data with panel data. These cohorts represent households with similar characteristics but different data collection capabilities, allowing the system to infer metrics for non-RPD/ACR households based on comparable RPD/ACR households, thereby improving measurement precision while maintaining operational simplicity
2Measurement precision
If the system collects data from all household types including non-RPD and non-ACR devices, then measurement precision improves, but device complexity increases due to multiple data collection methods
Solution Approach 1:
The patent creates a universal measurement framework that handles multiple household types through a single integrated system. The projection cohort methodology serves as a multi-functional approach that works across different device capabilities (RPD, ACR, and non-RPD/ACR households), eliminating the need for separate complex systems for each household type while maintaining high measurement precision
Solution Approach 2:
The patent changes the parameter of data collection from device-specific methods to household-characteristic-based projection. Instead of implementing different complex collection mechanisms for different device types, the system transforms the approach by using demographic and behavioral parameters to identify projection cohorts, thereby reducing system complexity while improving precision through comprehensive coverage
3Reliability
If panel data from enrolled households is used to create projection cohorts, then reliability of audience metrics improves, but loss of information increases due to the need to aggregate and generalize panel data
Solution Approach 1:
The patent applies partial action by using only the necessary panel data characteristics needed for projection purposes. Rather than collecting or retaining all detailed media access data from panel households, the system extracts and uses only the demographic and behavioral parameters required to identify projection cohorts, thereby maintaining reliability while minimizing information loss through selective data utilization
Data Source
AI summary
Methods, apparatus, systems, and articles of manufacture are disclosed to optimize projection of big data beyond its footprint. An example apparatus includes memory, instructions, and processor circuitry to access panel audience sizes corresponding to subscribers of a media provider, a subset of the subscribers of the media provider, and corresponding to a media network determine a relative percent difference between the third panel audience size and the fourth panel audience size, when the relative percent absolute difference satisfies a first threshold, determine percentages of demographic groups represented in the panel audience sizes, determine differences associated with the demographic groups, and when at least one of the differences corresponding to at least one of the demographic groups satisfies a second threshold, determine the subscribers, the subset of the subscribers, and the at least one of the demographic groups as useable to determine an audience size of the media network.