Browse Prior Art Database

SETTING ALLOCATIONS AND PRICES IN POSITION AUCTIONS WITH LONG-TERM VALUES

IP.com Disclosure Number: IPCOM000245947D
Publication Date: 2016-Apr-19
Document File: 6 page(s) / 160K

Publishing Venue

The IP.com Prior Art Database

Abstract

Allocations and prices in position auctions are set with long-term values. For each class of advertisers, a cumulative distribution function corresponding to the distribution of eCPM bids is determined. A probability density function corresponding to the cumulative distribution is determined for each advertiser. An amount is determined that an ad distributer would value not having to show an ad for advertiser on the page due to the user experience cost that will decrease user propensity to click on ads in the future. A virtual valuation is determined for each advertiser. Ads are ranked in order of their virtual valuations by assigning the highest slot to the advertiser with the highest virtual valuation. Ads are provided for display. A cost-per-click is completed that each advertiser must pay for their ads.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 37% of the total text.

Page 01 of 6

SETTING ALLOCATIONS AND PRICES IN POSITION AUCTIONS WITH LONG-TERM VALUES

                      TECHNICAL FIELD
[0001] This disclosure generally relates to providing content and charging associated content providers.

                       BACKGROUND
[0002] Content providers, such as advertisers, provide bids for the selection of their content (e.g., ads) in auctions. Different classes or types of content (e.g., ads) may have different costs, and the presentation of content (e.g., ads) can affect how a user may react to similar or different content (e.g., ads) in the future.

                        SUMMARY
[0003] In general, allocations and prices in position auctions can be set with long- term values. For each class of advertisers, a cumulative distribution function corresponding to the distribution of eCPM bids is determined. A probability density function corresponding to the cumulative distribution is determined for each advertiser. An amount is determined that an ad distributer would value not having to show an ad for advertiser on the page due to the user experience cost that will decrease user propensity to click on ads in the future. A virtual valuation is determined for each advertiser. Ads are ranked in order of their virtual valuations by assigning the highest slot to the advertiser with the highest virtual valuation. Ads are provided for display. A cost-per-click is completed that each advertiser must pay for their ads.

                  DESCRIPTION OF DRAWINGS
[0004] FIG. 1 is a flow diagram of an example method for setting allocations and prices in position auctions with long-term values.

1


Page 02 of 6

                   DETAILED DESCRIPTION
[0005] FIG. 1 is a flow diagram of an example method 100 for setting allocations and prices in position auctions with long-term values.

100

For each class of advertisers, determine a cumulative distribution function (CDF) corresponding to the distribution of eCPM bids

102

Determine a probability density function corresponding to the cumulative distribution for each advertiser

104

Determine an amount that an ad provider would value not having to show an advertiser's ad on a page due to a user experience cost that will decrease user propensity to click on ads in the future

106

Determine a virtual valuation for each advertiser

108

Rank ads in order of their virtual valuations by assigning the highest slot to the advertiser with the highest virtual valuation

110

Provide the ads for display on the page

112

Compute a cost-per-click that each advertiser must pay for their ads 114

FIG. 1

2


Page 03 of 6


[0006] At 102, for each class of advertisers, a cumulative distribution function (CDF) corresponding to the distribution of eCPM bids is determined. For example, for each class of advertisers, the cumulative distribution function (CDF) can be measured that corresponds to the distribution of eCPM bids that are made by this particular class of advertisers (e.g. advertisers bidding on the keyword "hotels"). Let Fi(v) denote the cumulative distribution function corresponding to the distribution of eC...