Abstract
A digital advertising platform constructs a per-user product ownership graph by ingesting third-party conversion signals from multiple advertisers, including purchase events and registration-related events, and resolving each signal to a canonical product entity using catalog matching, fuzzy attribute matching, and cross-merchant deduplication. The platform assigns an ownership confidence score and stores owned product nodes with acquisition metadata, taxonomy, attributes, lifecycle state, and relationship edges such as complementary, successor, substitute, and sequence relationships. During ad selection and auction, the platform suppresses ads for owned products and for satisfied categories, and adjusts ranking to promote complementary products, timed replacements, version upgrades, similarity-based discovery under contextual gating, and next-step products in learned purchase sequences. Ownership-based audience segments may be produced in aggregated form with minimum size thresholds and user opt-out support.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Anonymous, "Product Ownership Inference from Conversion Signals for Advertising Optimization", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/10748