Browse Prior Art Database

Method and System for Optimizing Price Parity for Advertisers and Partner Websites

IP.com Disclosure Number: IPCOM000197953D
Publication Date: 2010-Jul-23
Document File: 6 page(s) / 48K

Publishing Venue

The IP.com Prior Art Database

Related People

Alan Lu: INVENTOR [+3]

Abstract

A method and system for optimizing price parity for advertisers and partner websites is disclosed. An optimization formulation is provided to determine click quality on the partner websites and subsequent price parity for both the advertisers and partners under relevant business constraints.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 26% of the total text.

Method and System for Optimizing Price Parity for Advertisers and Partner Websites

Abstract

A method and system for optimizing price parity for advertisers and partner websites is disclosed.  An optimization formulation is provided to determine click quality on the partner websites and subsequent price parity for both the advertisers and partners under relevant business constraints.

Description

Disclosed is a method and system for optimizing price parity for advertisers and partner websites.  In order to attract search traffic, a search engine often places search boxes on partner websites.  When a search query is entered from a partner website, the search engine provides both algorithmic and sponsored search results.  The sponsored search results correspond to advertisements that are sponsored by various advertisers who pay an amount to the search engine and the partner websites when someone clicks on the search results.  However, in a pay per click system, it is difficult to determine the click quality.  For a same keyword, clicks from various partners may have different quality.  Typically, a conversion rate, or a number of conversions per click is used for determining the click quality.  For many reasons, the click traffic of certain keywords from certain partners may be consistently having lower conversion rates than the average.  In such a case, it is preferable to discount the price for these partner and keyword pairs, so that the advertisers do not waste money on inferior clicks.

The method and system as disclosed herein provide an optimization formulation for click quality on the partner websites and subsequent price parity to both the advertisers and partners under relevant business constraints.  Further, a macroscopic or global layer of pricing layer is provided that includes data organization, probability model and a global optimization model.

The optimization formulation minimizes the disparity under the various business constraints.  The disparity may be mathematically defined in the optimization formulation.  In a scenario, a monotonic constraint may correspond to a main business constraint.  The monotonic constraints are considered in order to ensure that publisher-keyword groups that have poorer quality receive more discounts.  In another scenario, discount and premium may be applied under certain boundaries.  A certain budget may also be specified while considering the total revenue cost as a result of the discounts.  A linear or quadratic programming may be applied to obtain an optimized discount rate for each pair of partner and keyword group.

For strong marketing reasons, search engine pricing cannot be advertiser specific, therefore the discounts are applied to partner-keywords pairs.  A weighted sum of disparity for each partner and keyword group pairs is calculated as a base of the discount.  However as each advertiser may have different marketing focus and different quality measurements, it is not rar...