Multi-Channel Attribution Using Time and Effectiveness Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the effectiveness of advertisements across multiple channels are challenging due to difficulties in attributing the influence of one channel's advertisement over another when a user interacts with both, as conventional multi-touch attribution methods rely on time between touch events and channel effectiveness without accurately accounting for both factors.
Innovation Solution
A system and method for attributing responses to multiple channels by determining conversion data for each channel based on user interactions, using both the effectiveness and time to convert, allowing for a holistic attribution methodology that considers the impact of multiple channels such as social media, television, and websites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional multi-touch attribution methods use only time between touch events to determine channel weights, then the attribution process is simplified, but the measurement precision of channel effectiveness is reduced
Solution Approach 1:
The patent combines multiple attribution factors (time between touch events and channel effectiveness metrics) into a unified attribution model. This merging allows the system to simultaneously consider both temporal sequences and channel performance data, resolving the contradiction by integrating rather than choosing between simplicity and precision.
Solution Approach 2:
The patent introduces additional parameters (channel effectiveness metrics) to the traditional time-based attribution model. By changing the parameter set from solely temporal measurements to include performance metrics, the system achieves more precise measurement without completely complicating the process, as the new parameters are systematically integrated.
2Measurement precision
If conventional multi-touch attribution methods use only channel effectiveness to determine channel weights, then the attribution accounts for channel quality, but the loss of time information reduces measurement precision
Solution Approach 1:
The patent merges time-based attribution with effectiveness-based attribution by calculating channel weights that incorporate both the temporal sequence of touch events and the effectiveness metrics of each channel. This combination ensures that neither time information nor effectiveness data is lost, but both contribute to the final attribution.
3Measurement precision
If the system attributes conversion credits based on both effectiveness and time, then the attribution accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent segments the attribution process into distinct computational steps: calculating time-based weights, calculating effectiveness-based weights, and combining these weights to determine final channel attribution. This segmentation manages complexity by breaking down the complex attribution task into manageable, systematic components.
Data Source
AI summary
Various implementations for multitouch attribution are described. One example method includes receiving a plurality of channel data associated with a first channel and second channel, determining conversion data for the first channel and the second channel using the plurality of channel data, receiving a first touch event and a second touch event associated with a first channel and a second channel, determining a first attribution for the first channel using a first touch event and the first channel data, determining a second attribution for the second channel using a second touch event and the second channel data, and determine an item conversion strategy using the first attribution and the second attribution.