Debiasing Media Creative Efficiency Using Weighted GLM
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for evaluating media creative efficiency in linear TV are biased due to factors like network, time, and cost structures, leading to inaccurate performance assessments across different television networks and ad spots.
Innovation Solution
A weighted generalized linear model (GLM) is used to analyze spot airing data, adjusting media creative efficiency by accounting for individual impacts beyond network effects, allowing for objective quantification of relative performance across multiple networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If media creative efficiency is calculated using aggregated lift and aggregated ad spends across multiple networks, then the calculation is simple, but the performance measurement is biased by network, time, and cost structure factors
Solution Approach 1:
The patent segments the aggregated efficiency calculation into network-specific efficiency measurements. Instead of calculating a single aggregated efficiency across all networks, the system calculates efficiency separately for each network-media creative combination, then combines these segmented measurements using weights that account for network characteristics. This segmentation eliminates the bias introduced by aggregating diverse network performances into a single metric.
Solution Approach 2:
The patent introduces weight parameters (w_ij) that modify the efficiency calculation to account for network-specific characteristics. By changing the calculation from a simple aggregation to a weighted aggregation where weights represent network influence and cost structures, the system adjusts the measurement to reflect true media creative performance independent of network biases.
2Ease of operation
If traditional efficiency measurement is used without adjusting for network effects, then the measurement approach is straightforward, but it leads to false conclusions about media creative performance
Solution Approach 1:
The patent introduces weight parameters as intermediary elements between the raw efficiency data and the final performance assessment. These weights act as mediators that adjust for network effects, time variations, and cost structures, allowing the system to maintain operational simplicity while improving reliability through the intermediary adjustment layer.
Solution Approach 2:
The system incorporates feedback mechanisms by using network-specific efficiency measurements to inform and adjust the overall performance assessment. The weighted aggregation process provides feedback about network-specific performances, allowing the system to continuously refine its measurements and eliminate biases in the performance assessment.
3Loss of time
If media creative efficiency is evaluated independently of network and time factors, then the evaluation is quick and simple, but it produces biased results that do not reflect true performance
Solution Approach 1:
The patent performs preliminary calculations of network-specific efficiency measurements before combining them into the overall assessment. By pre-calculating efficiency for each network-media creative combination and preparing the weight parameters in advance, the system reduces the time required for the final evaluation while maintaining high measurement precision through the detailed preliminary analysis.
Data Source
AI summary
A quantification system is configured for debiasing media creative efficiency. In some embodiments, the quantification system leverages a weighted generalized linear model (GLM) to determine the individual impacts of media creatives beyond network effects. To prepare input data for fitting the weighted GLM, the quantification system analyzes spot airing data, creates a specific data structure for storing observations (e.g., network-media creative combinations) that can be provided to the weight GLM as input, and computes additional input data points needed by the weighted GLM (e.g., network spend, media creative efficiency per network-media creative combination, etc.). The weighted GLM is then fitted to obtain coefficients representing the individual impacts of the media creatives. The quantification system utilizes the computed impacts to adjust the previously computed media creative efficiency for each media creative. In this way, relative performance of media creatives can be objectively quantified across networks without needing digital evidence.