Browse Prior Art Database

Method and System to Resolve Sponsored Search Advertisement Footprint at Query Level

IP.com Disclosure Number: IPCOM000248191D
Publication Date: 2016-Nov-07
Document File: 5 page(s) / 198K

Publishing Venue

The IP.com Prior Art Database

Related People

Avishek Saha: INVENTOR [+3]

Abstract

A method and system is disclosed that provides a higher fidelity resolution for maximum number of advertisements (ads) to include in prominent positions within a search results page, specifically in north section of the search results page. The method and system computes offline parameters using historic logs, and applies the precomputed parameters at serve time for providing the most appropriate ad count for each individual search event.

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 30% of the total text.

Method and System to Resolve Sponsored Search Advertisement Footprint

at Query Level

Abstract

A method and system is disclosed that provides a higher fidelity resolution for maximum number of advertisements (ads) to include in prominent positions within a search results page, specifically in north section of the search results page.  The method and system computes offline parameters using historic logs, and applies the precomputed parameters at serve time for providing the most appropriate ad count for each individual search event.

Description

Disclosed is a method and system that provides a higher fidelity resolution for maximum number of advertisements (ads) to include in prominent positions within a search results page, specifically in north section of the search results page.  The method and system computes offline parameters using historic logs, and applies the precomputed parameters at serve time for providing the most appropriate ad count for each individual search event.

The method and system resolves maximum north ad count dynamically at serve time using search event specific features and marketplace conditions.  Further, the method and system resolves maximum north ad count for a search advertising results page employing query specific information, trading off the impact on user experience and ad monetization, using an offline and online

policy.

Further, the method and system allows for the tradeoff to be made at serve time, considering the following aspects:

For example, the method and system takes into account the following factors:

 

  • Each ad’s performance at its intended rank in the ad unit and the overall depth of the resulting ad unit.
  • A marginal return of each additional ad in the intended ad unit constructed for the current search query in terms of an overall objective that consists of a monetization goal and user impact.
  • A determination to not show an additional ad unit in the particular event that it has less than significant marginal return relative to some tradeoff level.
  • The tuning of such trade off parameters dynamically based on real time data, some queries that may deliver a trimmer, more positive ad experience paying off in long term user satisfaction and their subsequent search volume and
  • The same query that may exhibit different footprint for different users, and in combination with other factors such as, but not limited to, location, time of day and the user’s behavioral interests.

The above mentioned aspects are in contrast to a fairly static and offline configuration of a predefined ad footprint size that may at best be refined by a section of the page or recently to loyal or subscribing users.

The figure below illustrates an implementation of the overall search advertising technique of the method and system in accordance with an embodiment.

Figure

As illustrated in the figure, the solid lines denote the existing system whereas the dotted lines denote the modifications introduced to implement the method and...