Behavioral Cohort Identity Resolution for Cross-Platform Experience Delivery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems lack the ability to consistently deliver new software experiences across platforms and devices, fail to utilize end-user behavioral data for accurate targeting, and have limited analytics capabilities, leading to inaccurate experiment results and inconsistent user experiences.
Innovation Solution
A method and system for delivering data-driven, cross-platform experiences using behavior cohorts and identity resolution, which includes collecting user data, identifying cohorts based on behavior and properties, ensuring consistent experience delivery across devices, and using integrated SDKs for real-time tracking and analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing systems use a single user identifier to determine experience delivery, then the system operation is simple, but the user experience consistency across platforms cannot be ensured
Solution Approach 1:
The patent segments the identity resolution process into multiple components: device identifiers, platform identifiers, user account identifiers, and behavioral data. This segmentation allows the system to track users across different devices and platforms by matching multiple identifier types rather than relying on a single identifier, thereby ensuring experience consistency while maintaining manageable system complexity through modular processing
Solution Approach 2:
The patent creates a universal user profile that consolidates multiple identifier types (device IDs, platform IDs, account information) and behavioral data into a single unified representation. This universal profile serves as the foundation for delivering consistent experiences across all platforms and devices, allowing the system to recognize and track the same user regardless of which device or platform they access from
2Measurement precision
If existing systems do not access end-user behavioral data, then the system complexity is low, but the experiment targeting accuracy deteriorates
Solution Approach 1:
The patent performs preliminary actions by collecting and processing behavioral data in advance to create user profiles and behavioral cohorts before experiments are executed. User behaviors, preferences, and patterns are pre-analyzed and stored, enabling accurate targeting when experiments run without requiring complex real-time data processing during experiment execution
Solution Approach 2:
The patent introduces behavioral cohorts as an intermediary layer between raw behavioral data and experiment targeting. Instead of directly using complex behavioral data for targeting decisions, the system first groups users into cohorts based on shared behavioral characteristics, then applies experiments to these cohorts. This intermediary structure simplifies the integration complexity while maintaining high targeting accuracy
3Loss of time
If existing systems evaluate software improvements in short time horizon, then the analytics processing is fast, but the experiment analysis completeness deteriorates
Solution Approach 1:
The patent implements continuous tracking and analysis of user experiences across multiple time points and platforms. Instead of evaluating experiments only at release milestones, the system continuously monitors user interactions, collects behavioral data, and updates experiment metrics in real-time. This continuous action ensures comprehensive analysis while maintaining efficient processing through streaming data architecture
Solution Approach 2:
The patent adds temporal and cross-platform dimensions to experiment evaluation. By tracking user experiences across multiple devices, platforms, and time periods, the system creates a multi-dimensional view of experiment impact. This dimensional expansion allows comprehensive analysis without proportionally increasing processing time, as the system processes data in parallel across different dimensions
4Measurement precision
If existing systems track data only at release milestones, then the data collection complexity is low, but the delivery tracking accuracy deteriorates
Solution Approach 1:
The patent implements periodic tracking at multiple intervals: continuous real-time tracking during user sessions, periodic aggregation at defined time intervals, and milestone-based reviews at release points. This multi-interval periodic action ensures accurate delivery tracking without requiring constant complex processing, as the system adjusts tracking intensity based on the evaluation stage and user activity patterns
Data Source
AI summary
Systems and methods for delivering data-driven experiences of a product are described herein. In some embodiments, a server receives user data associated with a plurality of users using the product. The server identifies, from the plurality of users, a first user belonging to a first cohort based on at least one of user behaviors and user properties associated with the first user included in the user data. The server causes an experience of the product to be delivered to the first user. The server then detects that the first user has switched from a first device to a second device. The server further causes the experience to continue to be delivered to the first user.