Advertising Reach Estimation Using Precomputed Device Probabilities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advertisers face challenges in accurately estimating the reach and device frequency of advertising campaigns due to limitations in estimating overlap in viewership, which affects the effectiveness of strategies for proposed campaigns.
Innovation Solution
A method that generates probability data for devices presenting advertising content in specific time blocks, using historical viewing data to estimate reach and device frequency for future campaigns by calculating first and second probabilities based on device usage patterns across multiple time blocks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional exposure estimation methods are used based on previously aired campaigns, then campaign strategies can be devised, but the accuracy of reach and frequency estimation is limited
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing probability data for each device-time block combination before the advertising campaign begins. Historical viewing data is analyzed in advance to determine the probability that each device will be used during each time block, and these probabilities are stored for rapid retrieval during campaign planning. This eliminates the need for complex real-time calculations and provides accurate reach and frequency estimates before the campaign launches, enabling more reliable strategy formulation.
Solution Approach 2:
The patent segments the estimation process into distinct components: (1) dividing time periods into discrete time blocks, (2) identifying individual devices separately, (3) calculating first probabilities for each device-time block combination independently, and (4) computing second probabilities for sets of time blocks separately. This segmentation allows each component to be optimized independently and combines to provide accurate overall estimates of reach and frequency.
2Measurement precision
If overlap of viewership is estimated for future campaigns, then reach can be calculated, but it is difficult to estimate accurately
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing probability data for each device-time block combination before the advertising campaign begins. Historical viewing data is analyzed in advance to determine the probability that each device will be used during each time block, and these probabilities are stored for rapid retrieval during campaign planning. This eliminates the need for complex real-time calculations and provides accurate reach and frequency estimates before the campaign launches, enabling more reliable strategy formulation.
Solution Approach 2:
The patent introduces probability data as an intermediary between historical viewing data and reach/frequency estimates. Instead of directly analyzing complex overlapping viewership patterns, the system uses pre-computed probabilities as a mediator that simplifies the estimation process. These probability values capture the essential overlap information in a compact form that can be easily combined to estimate reach and frequency without requiring complex real-time analysis.
Data Source
AI summary
Estimating the reach and device frequency of a proposed advertising campaign based on the probability that advertising content delivery devices (e.g., a television set top box, radio) were used to deliver advertising content (e.g. television ad, radio ad) during specific time periods on specific delivery device channels.