Cross-Device Conversion Attribution via Shared Account Intermediaries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computer systems face challenges in efficiently determining related network activity across multiple computing devices during different network sessions, particularly in attributing conversions involving third-party content when users interact with content across various devices without being logged into the same online accounts.
Innovation Solution
A method and system for estimating cross-device conversions by determining observed interactions with third-party content, identifying cross-device conversions based on devices logged into the same online accounts, and calculating an estimated total number of conversions using observed data and account login percentages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a computer system tracks network activity across multiple computing devices to determine conversions, then the accuracy of conversion attribution is improved, but the complexity of determining related activity across different devices and sessions increases
Solution Approach 1:
The patent uses online account identifiers as intermediaries to link activity across different computing devices. When a user interacts with third-party content on one device and later converts on a different device, the system uses the shared online account identifier to attribute the conversion to the original content interaction, resolving the complexity of cross-device tracking without requiring direct device-to-device correlation
Solution Approach 2:
The system segments the conversion tracking problem into distinct components: (1) tracking content interactions on individual devices, (2) identifying devices associated with the same online account, and (3) attributing conversions based on account-level relationships. This segmentation allows the system to handle cross-device attribution through manageable, modular processing steps
2Quantity of substance
If the system monitors all computing devices and network sessions to estimate conversions, then the completeness of conversion data is improved, but the computational resources and time required for analysis increase
Solution Approach 1:
The system performs preliminary actions by pre-establishing relationships between computing devices and online accounts before conversion events occur. Devices are pre-tagged with online account identifiers, and the system maintains ready-made associations between devices sharing the same account. When a conversion event is detected, the system can immediately query pre-established device-account relationships rather than performing complex analysis across all devices in real-time
Solution Approach 2:
The system applies partial monitoring by focusing computational resources on devices that are members of the same online account communities. Rather than analyzing all possible device combinations, the system only needs to check relationships within account-based device groups, significantly reducing the search space while maintaining complete conversion data for relevant devices
Data Source
AI summary
Multi-computing device network based cross-device conversion determination is described. A content selection computer server can identify cross-device conversions. A first computing device accesses third-party content via a first computer network connection. A second computing device accesses a webpage of the third-party content provider via a second computer network connection. The first and devices can be logged into a same online account. The content selection computer server can obtain, via a content tag that includes a script that executes on a webpage that includes the third-party content, data indicating that the first computing device accessed the third-party content. The selection computer server determines the conversion from the content tag and the second computing device having accessed the webpage. Based on a percentage of first and second sets of computing devices that have logged into common respective online accounts, the content selection computer server extrapolates an estimated number of cross-device conversions.