Management of cannibalistic ads to reduce internet advertising spending
AssigneeTaskmaster Technologies Inc.
InventorsMatt LeBaron, John Holsworth
Life of this patent
Estimating the term…
Abstract
Generating an estimate of reclaimed ad spend for an Internet advertising campaign, by receiving a set of keywords, where each keyword corresponds to a paid ad that is supplied to a search engine, gathering a cannibalism score for each paid ad, where a cannibalism score indicates that the presence of a designated paid ad in a SERP reduces the chance that a user will click on a corresponding unpaid listing, and estimating a reclaimed ad spend as the difference between the actual revenue reported for a period of time and an estimate of the expected ad spend for a comparable period of time when de-cannibalizing actions were taken; and reporting the estimate of the reclaimed ad spend.
Claims18
A computer-implemented method for estimating the efficiency of an Internet advertising campaign, comprising:
- receiving a plurality of keywords, wherein each keyword corresponds to a paid ad that is supplied to a search engine, wherein a paid ad includes a link to a web page, wherein each web page is in a domain, and wherein in response to receiving a keyword from a web browser the search engine returns a corresponding search engine results page (SERP) that includes (1) at least one paid advertisement, and (2) at least one unpaid listing, wherein an unpaid listing includes a link to a web page, and wherein a paid ad has an ad position within a web page in relation to other paid ads wherein the first ad position is the most valuable position and larger ad position values are less valuable and an unpaid listing has an unpaid listing position within a web page in relation to other unpaid listings wherein the first unpaid listing position is the most valuable position and larger unpaid listing position values are less valuable;
- determining an ad position for a designated paid ad and an unpaid listing position for each unpaid listing that corresponds to the designated unpaid ad, wherein a corresponding unpaid listing includes a link to the same domain as the link included in the designated paid ad, and wherein a position is an average position across multiple samples of the keyword that corresponds to the designated ad;
- generating (1) a cannibalism score for the designated paid ad, wherein a nonzero cannibalism score indicates that the designated paid ad appears in a first ad position in a SERP, there is no paid ad in a second ad position, and a corresponding unpaid listing appears in the first unpaid listing position in the SERP, and (2) an actual revenue due to clicks on the designated paid ad and on the corresponding unpaid listing for a time period where no de-cannibalizing actions were taken, wherein a de-cannibalizing action occurs when a paid ad is not supplied to the search engine due to having a nonzero cannibalism score;
- estimating a reclaimed ad spend for the designated paid ad as the difference between an estimate of the expected cost of purchasing the designated ad for a period of time and the actual cost reported for a comparable period of time during which de-cannibalizing actions were taken;
- and
- reporting the estimate of the reclaimed ad spend for an ad buy based on the generated cannibalism score.
A server computer implemented on a cannibalistic advertising system (CAS), comprising:
- a keyword gatherer for gathering:
- a plurality of keywords that are relevant to a designated advertiser, wherein each keyword corresponds to a paid ad that is supplied to a search engine, wherein a paid ad includes a link to a web page, wherein each web page is in a domain, and wherein in response to receiving a keyword from a web browser the search engine returns a corresponding search engine results page (SERP) that includes (i) at least one paid advertisement, and (ii) at least one unpaid listing, wherein an unpaid listing includes a link to a web page, and wherein a web page is within a domain and wherein a paid ad has an ad position within a web page in relation to other paid ads wherein the first ad position is the most valuable position and larger ad position values are less valuable and an unpaid listing has an unpaid listing position within a web page in relation to other unpaid listings wherein the first unpaid listing position is the most valuable position and larger unpaid listing position values are less valuable;
- an ad database for storing (1) a paid ad for each of the plurality of keywords, (2) a cannibalism score for each paid ad, wherein a nonzero cannibalism score for a paid ad indicates that the paid ad appears in a first ad position in a SERP, there is no paid ad in a second ad position, and a corresponding unpaid listing appears in the first unpaid listing position in the SERP, wherein a corresponding unpaid listing includes a link to the same domain as the link included in the designated paid ad;
- a cannibalistic ad analyzer for:
- determining an ad position for a designated paid ad and an unpaid listing position for each unpaid listing that corresponds to the designated unpaid ad, and wherein a position is an average position across multiple samples of the keyword that corresponds to the designated ad;
- and
- generating a cannibalistic ad score for the designated paid ad;
- wherein the keyword gatherer additionally gathers an actual revenue due to clicks on a paid ad and on any corresponding unpaid listings in a SERP for a time period where no de-cannibalizing actions were taken, wherein a de-cannibalizing action occurs when a paid ad is not supplied to the search engine due to having a nonzero cannibalism score;
- a reclaimed ad spend analyzer for estimating a reclaimed ad spend for a keyword as the difference between an estimate of the expected ad spend due to purchasing a paid ad corresponding to the keyword for a comparable period of time and the actual cost of purchasing the ad as reported for a comparable period of time during which de-cannibalizing actions were taken;
- and
- a cannibalistic ad report generator for reporting the estimate of the reclaimed ad spend for an ad buy that based on the generated cannibalism score.
Dependencies are inferred from the claim text.