System for categorizing a seller relative to a vendor

摘要

Methods for competitive pricing are implemented in software for commerce in an electronic marketplace. A vendor may identify rivals: competitors against which the vendor most directly competes by dint of having similar reputation and market positioning. Using specified pricing models applied to filtered price data periodically gathered by spider-crawling competitors on the Internet, a vendor may adjust product prices to match rivals while accounting for overall marketplace price trends.

主权项

1. A system that is adapted to categorize a seller relative to a vendor comprising:
a processor; a memory; an overall pricing model for a vendor being stored in the memory, wherein the overall pricing model comprises a position within an ordinal range of pricing levels, and wherein said overall pricing model relates to a composite representation of item pricing models for a set of on-sale items; product-specific data for a first set of items being stored in the memory, wherein said first set comprises a plurality of items; a first set of instructions being stored in the memory that, when executed, determines a product mix comprising a plurality of items from said first set of items and stores said product mix in the memory; a second set of instructions stored in the memory that, when executed, read said product mix as a basis for search, whereby finding, via network connection, for a first set of sellers, current prices for a result set of items, wherein said first set of sellers comprises at least a first seller, wherein said first set of sellers does not include said vendor, as current prices for a plurality of on-sale items in said product mix by said vendor are available from a database without network search, and wherein said result set of items comprises a plurality of items in said product mix; storing prices for said result set item prices; a third set of instructions stored in the memory that, when executed,
(a) statistically characterize respective item price data by determining data representative of an item price curve for respective items in said result set of items,(b) store said item price curve data in the memory,(c) derive respective seller item pricing models for a plurality of items in said result set of items, wherein a said item pricing model comprises a position within a relative range of pricing levels for said plurality of items based upon said item price curve data;determining an overall pricing model for said first seller using a plurality of said first seller item pricing models;a fourth set of instructions stored in the memory that, when executed, categorize said first seller by at least in part comparing said first seller's overall pricing model to at least one threshold of similarity to said vendor's overall pricing model; and storing said first seller categorization.