CTV Audience Measurement Using Person-Level Demographic Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing CTV advertising often fails to reach the intended audience at the individual level due to household-level targeting, resulting in wasted ad dollars and inefficient inventory utilization, as households consist of diverse audiences with varying interests.
Innovation Solution
Utilizing panelist information, CTV customer viewing behavior, and machine learning models to predict person-level demographic assignments for media access data, enabling real-time audience identification and targeted advertising.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If household-level targeting is used for CTV advertising, then device complexity is reduced and ease of operation is improved, but measurement precision of audience demographics deteriorates and loss of information increases
Solution Approach 1:
The patent segments the household audience into individual person-level demographic categories (e.g., adult males, adult females, children) using machine learning models that analyze media access behavior patterns. This allows advertising to be targeted at specific persons within the household rather than treating the entire household as a single unit, thereby improving measurement precision while maintaining operational simplicity through automated modeling.
2Device complexity
If household-level targeting is used for CTV advertising, then device complexity is reduced, but loss of information about individual audience members increases
Solution Approach 1:
The patent performs preliminary actions by training machine learning models offline using historical media access behavior data and panelist information before actual advertising deployment. This pre-computation of person-level demographic probabilities eliminates the need for complex real-time processing during ad delivery, maintaining low device complexity while preserving rich individual-level audience information for targeted advertising decisions.
3Measurement precision
If person-level demographic prediction is implemented, then measurement precision of audience demographics is improved, but device complexity and computational requirements increase
Solution Approach 1:
The patent implements self-service by having the media access device itself collect and analyze its own media consumption behavior data using embedded machine learning models. The device autonomously generates person-level demographic predictions based on its observed behavior patterns, eliminating the need for external complex processing systems and reducing overall device complexity while achieving high measurement precision.
4Productivity
If real-time audience identification is implemented, then productivity of advertising delivery is improved, but use of energy and computational resources increases
Solution Approach 1:
The patent performs preliminary action by pre-training machine learning models offline using historical data before real-time advertising delivery. The models are prepared in advance to make rapid predictions during ad delivery, enabling real-time audience identification without requiring intensive computational resources or high energy consumption during the actual advertising process, thus improving productivity while controlling energy use.
Data Source
AI summary
Methods, apparatus, systems, and articles of manufacture are disclosed to generate audience metrics for connected television. An example system includes at least one memory, programmable circuitry, and instructions to cause the programmable circuitry to obtain media access data corresponding to connected television media and a user identifier corresponding to a media access device, generate, using a machine learning model, probability values for corresponding audience demographics in a household composition corresponding to the user identifier, the probability values indicative of likelihoods that corresponding ones of the audience demographics are accessing the connected television media, determine a person-level characteristic based on the probability values of the audience demographics, the person-level characteristic corresponding to an audience member of the connected television media, and assign the media access data to the person-level characteristic.