Digital Marketing Attribution via Multi-Order Probability Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques for evaluating digital marketing campaigns often inaccurately attribute conversion outcomes, overlooking the contributions of previous campaigns, leading to unsatisfactory results and misallocation of resources.
Innovation Solution
A system that analyzes historical data to determine first, second, and third order probabilities of conversion for each campaign in a path, attributing campaign contributions based on these probabilities to accurately assess the effectiveness of individual campaigns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional techniques credit the latest digital marketing campaign as causing conversion, then the evaluation process is simple and quick, but the measurement precision of campaign effectiveness is inaccurate
Solution Approach 1:
The patent segments the conversion attribution process into multiple probability orders (first-order, second-order, third-order probabilities) to evaluate individual campaign contributions separately rather than attributing all conversions to the latest campaign. This segmentation allows for more precise measurement of each campaign's specific contribution while maintaining a structured evaluation framework.
Solution Approach 2:
The system performs preliminary analysis by calculating first-order probabilities for each campaign before determining second-order and third-order probabilities. This preliminary action establishes a foundation for more accurate attribution by first assessing individual campaign effectiveness, then layering in interactions with other campaigns in a systematic sequence.
2Reliability
If conventional techniques ignore previous digital marketing campaigns, then the analysis is straightforward and fast, but the reliability of conversion attribution is poor
Solution Approach 1:
The system performs preliminary analysis by calculating first-order probabilities for each campaign before determining second-order and third-order probabilities. This preliminary action establishes a foundation for more accurate attribution by first assessing individual campaign effectiveness, then layering in interactions with other campaigns in a systematic sequence.
Solution Approach 2:
The patent implements feedback mechanisms by using historical conversion data to continuously refine probability calculations. The system feeds back into the attribution model by updating first-order, second-order, and third-order probabilities based on observed conversion patterns, thereby improving reliability over time while managing analysis time through iterative learning.
3Measurement precision
If the system analyzes multiple probability orders for each campaign, then the campaign contribution attribution becomes accurate, but the computational complexity increases
Solution Approach 1:
The patent segments the probability calculation process into distinct orders (first-order, second-order, third-order), where each order evaluates a specific aspect of campaign contribution. This segmentation allows the system to manage computational complexity by breaking down the overall calculation into manageable, modular components that can be computed and stored separately.
Solution Approach 2:
The system applies local quality by assigning different levels of probability analysis to different campaigns based on their specific characteristics and positions in the campaign path. Rather than uniformly applying complex multi-order probability calculations to all campaigns, the system tailors the depth of analysis to each campaign's contextual importance, optimizing computational resources while maintaining measurement precision where most needed.
Data Source
AI summary
Methods and systems for attributing contributions to digital marketing campaigns in achieving an action are described. In one or more implementations, first, second, and third order probabilities of a user taking the action are computed for each of a plurality of campaigns of a campaign path. Based on the probabilities, contributions are attributed to the campaigns of the campaign path in achieving the action.