SDK Engagement Monitoring Through Privacy-Preserving Event Buckets

The patent introduces an intermediary mechanism that aggregates and anonymizes engagement data before transmission to the SDK provider system. The system uses bucket identifiers derived from calendar dates rather than precise timestamps, and aggregates event counts at the bucket level rather than tracking individual user interactions. This intermediary layer preserves engagement monitoring capability while filtering out PII and precise user-level data.

The patent transforms the data parameters from precise event counts and timestamps to aggregated bucket identifiers. Instead of communicating exact numbers of user interactions or precise timing data, the system uses rounded calendar date buckets (e.g., start of month, quarter, or year) as the temporal parameter, and aggregates event counts within these buckets. This parameter transformation maintains analytical utility while eliminating PII.

The patent segments the engagement data into aggregate buckets based on calendar dates rather than tracking individual user events. The data is divided into time-based segments (buckets) and aggregated within each segment, so that no single user's interaction pattern can be identified. This segmentation approach maintains the ability to track engagement trends while preventing identification of specific users or their PII.

The patent creates a simplified copy of the engagement data structure that preserves the analytical relationships between different metrics while removing sensitive details. The bucket identifier system creates a representative sample of the data that maintains statistical validity for engagement analysis without containing the original precise user-level information. This copying approach allows continued analysis while eliminating PII.