Identification and Management of Cannibalistic Ads to Improve Internet Advertising Efficiency

The system calculates cannibalism scores to identify and avoid ads that divert clicks from free listings, optimizing advertising campaigns by reducing unnecessary spending and enhancing profitability.

KR103045289B1Active Publication Date: 2026-11-06TASKMASTER TECHNOLOGIES INC
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
KR1020237003895
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Priority Date
2020-09-03
Filing Date
2021-04-23
Publication Date
2026-11-06
Estimated Expiration
2041-04-23

AI Technical Summary

Technical Problem

Cannibalistic ads in online advertising campaigns divert visitors from free listings to paid ads, increasing advertising costs without generating additional revenue, necessitating a method to identify and manage these ads to optimize campaign efficiency.

Method used

A system and method to calculate a cannibalism score for ads based on proximity, category, and revenue impact, allowing advertisers to determine the likelihood of cannibalism and estimate potential savings by avoiding such ads.

Benefits of technology

Enables advertisers to reduce unnecessary spending on cannibalistic ads, improving the efficiency and profitability of their campaigns by identifying and avoiding ads that divert clicks from free listings.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112023012385985-PCT00001_ABST
    Figure 112023012385985-PCT00001_ABST
Patent Text Reader

Abstract

Keywords related to an advertiser are collected, and rules are defined to calculate a cannibalism score for an advertisement in relation to a corresponding free listing—the cannibalism score estimates the decrease in revenue for the advertiser resulting from the advertisement appearing in the same SERP as the corresponding listing—the keywords are provided to a search engine, a SERP is received from the search engine, the position of the first advertisement placed by the advertiser is determined among one or more advertisements within the SERP, and the position of a corresponding free listing is determined among multiple free listings within the SERP; and by applying the rules to the advertisement and the corresponding free listing to obtain a cannibalism score for the advertisement, a cannibalism score for a paid advertisement within the search engine results page (SERP) is generated.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to online advertising technology, in particular to the ability to identify and manage the purchase of cannibalistic advertisements within web pages. Background Technology

[0002] Internet search engines return one or more web pages to a user's browser in response to a keyword search performed by the user. The returned web pages, known as "search engine results pages" or SERPs, generally include unpaid search listings, referred to as "organic" search results, and paid advertisements, also referred to as ads. Each listing contains a URL or a link to a worldwide webpage related to the search query entered by the user. The web page corresponding to the URL returned in a paid advertisement or free listing within a SERP is often referred to as a landing page.

[0003] The goal of online advertising is to induce users to click on paid ads and subsequently visit the advertiser's website. The effectiveness of an online advertising campaign is generally measured as a function of the number of clicks on paid listings, which is equivalent to the number of website visitors resulting from the campaign and the revenue generated by those visitors. The effectiveness of an advertising campaign can be measured as the ratio of revenue generated to advertising costs or the ratio of visitors to advertising costs.

[0004] However, the effectiveness of advertising campaigns can be severely limited by cannibalistic ads, which incur additional costs by diverting visitors who would otherwise have visited to paid ads as a result of clicking on free listings. In other words, cannibalistic ads increase the advertiser's costs by diverting visitors from free listings to cannibalistic ads.

[0005] In its most basic form, a cannibalistic ad is an ad purchased by an advertiser that appears right next to a free listing on a web page promoting the same service or product as a paid ad. Users sometimes click on the paid ad instead of the free listing. This results in unnecessary advertising costs for the advertiser, based on the assumption that in some percentages, the user would have clicked on the free listing. In other cases, there may be one or more ads between the advertiser's ad and the corresponding free listing. In such cases, the paid ad can still be considered cannibalistic. Therefore, it would be advantageous to be able to identify cannibalistic ads so that advertisers can decide whether or not to purchase them.

[0006] Since cannibalistic ad costs are a new metric for determining the efficiency of advertising campaigns, it is advisable to determine the amount advertisers spend on purchasing cannibalistic ads in addition to detecting them. Through this value, advertisers can determine the scale of the problem and reduce spending on cannibalistic ads without sacrificing the effectiveness of their advertising campaigns.

[0007] A method, system, and device for measuring an amount spent by an advertiser purchasing cannibalistic advertisements (ads) and an amount that can be reclaimed by not purchasing cannibalistic advertisements (referred to as reclaimed advertising costs) are disclosed.

[0008] In a specific embodiment, the method calculates a cannibalism score for an advertisement within a search engine results page (SERP), which estimates the likelihood that a paid advertisement will be cannibalistic, that is, that a paid advertisement will appear in close proximity to a corresponding free listing. In another embodiment, the cannibalism score estimates the loss in revenue for an advertiser resulting from the appearance of a paid cannibalistic advertisement in the same SERP as a corresponding free listing.

[0009] A specific embodiment relates to collecting keywords related to an advertiser, defining rules for calculating a cannibalism score for an ad in relation to a corresponding free listing—wherein the cannibalism score estimates the decrease in revenue for the advertiser due to the ad appearing in the same SERP as the corresponding listing—providing the keywords to a search engine, receiving a SERP from the search engine, determining the position of the first ad placed by the advertiser among one or more ads within the SERP, determining the position of a corresponding free listing among multiple free listings within the SERP, and generating a cannibalism score for a paid ad on a search engine result page (SERP) by applying the rules to the ad and the corresponding free listing to obtain a cannibalism score for the ad.

[0010] A specific embodiment relates to receiving a set of keywords—each keyword corresponding to a paid advertisement supplied to a search engine—collecting a cannibalism score for each paid advertisement—wherein the cannibalism score indicates that the presence of a specified paid advertisement within a SERP reduces the chance of a user clicking on a corresponding free listing—and generating an estimate of the cost of recovery for an internet advertising campaign by estimating the cost of recovery as the difference between actual revenue reported over a period and an estimate of the expected cost of recovery over a comparable period in which decannibalizing actions were taken; and reporting the estimate of the cost of recovery.

[0011] The embodiment relates to a computer-implemented method for estimating the efficiency of an internet advertising campaign, the method comprising: receiving a plurality of keywords—each keyword corresponds to a paid advertisement supplied to a search engine, the paid advertisement includes a link to a web page, and in response to receiving keywords from a web browser, the search engine returns a search engine result page (SERP) containing (1) a corresponding paid advertisement and (2) at least one free listing, the free listing includes a link to a web page, and the web page is within a domain—; (1) a cannibalism score for a specified paid advertisement—the cannibalism score indicates that the presence of a specified paid advertisement within the SERP reduces the opportunity for a user to click on a corresponding free listing, and the corresponding free listing has a web page linked to the same domain as the web page linked to the specified paid advertisement—and (2) collecting actual revenue from clicks on a specified paid advertisement and any corresponding free listing within the SERP during a period when no decannibalization action is taken—the decannibalization action occurs when the paid advertisement is not supplied to the search engine due to its cannibalization score—; The method includes the step of estimating the recoupment cost for a designated paid advertisement as the difference between the estimated cost of purchasing the designated advertisement during one period and the actual cost reported during a comparable period in which decannibalization measures were taken; and the step of reporting the estimate of the recoupment cost. Brief explanation of the drawing

[0012] The present invention will be more fully understood and recognized from the following detailed description taken together with the drawings. Figure 1 is a simplified block diagram of a cannibalistic ad system (CAS) that automatically identifies cannibalistic ads and calculates potential improvements that can be achieved by not purchasing such ads. Figure 2 is an example of a cannibalistic advertisement. Figure 3 is a simplified flowchart of the entire method for identifying cannibalistic advertisements. Figure 4 is an overall method for determining the measure of efficiency for an advertising campaign based on the estimation of potential savings resulting from reduced spending on cannibalistic advertising. Figure 5 is a block diagram depicting a software module included in a cannibalism advertising system. The drawings depict embodiments of the invention for illustrative purposes only. Those skilled in the art will readily recognize from the following discussion that alternative embodiments of the structures and methods illustrated herein, as well as combinations of the embodiments, may be used without departing from the principles of the invention described herein. Specific details for implementing the invention

[0013] The present invention will now be described more fully below with reference to the accompanying drawings, which show specific exemplary embodiments that form part of the invention and, as examples, show how the invention can be practiced. However, the invention may be embodied in many different forms and should not be interpreted as being limited to the embodiments described herein; rather, such embodiments are provided to ensure that the disclosure is thorough and complete and sufficiently conveys the scope of the invention to those skilled in the art. Above all, the invention may be embodied as a method, process, system, business method, or device. Accordingly, the invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Accordingly, the following detailed description should not be taken in a limiting sense.

[0014] The following terms used here have the meanings given below:

[0015] User - refers to an individual who accesses the service provided by the present invention via a network using a mobile device, PC, or other electronic device.

[0016] Advertiser - Refers to an individual, company, or other organization that places online advertisements through search engines for goods or services being advertised, sold, or promoted, or causes online advertisements to be placed.

[0017] Keyword or search term – also referred to as a keyword query – refers to a word, words, phrase, or sentence entered by a user into a search field on a web page and then transmitted to a search engine that performs the requested search and returns results. Advertisers may purchase or bid on advertisements corresponding to keywords; in this case, the SERP returned by the search engine in response to the user entering the keyword includes paid advertisements placed by advertisers that correspond to the keyword.

[0018] A search engine or web search engine—generally refers to a computer server or internet service that receives keywords as a result of a keyword query, uses the keywords to search for web pages corresponding to the keywords, and returns one or more search engine result pages (SERPs) including paid advertisements and free or organic listings.

[0019] Listings are results from keyword searches that appear in the SERP. Each listing includes a link to a corresponding web page. Listings can be paid advertisements, i.e., paid listings, or free or organic listings generated by search engines. Listings in the SERP are ranked; each listing has a numerical position starting from the top, or first, or highest position. Unless otherwise specified, a listing position within the SERP refers to the numerical position of the listing relative to the top of the free search results. Thus, the first position is the highest, the second position is the next highest, and so on. Paid listings have paid listing positions, and free listings have free listing positions.

[0020] Search Engine Results Page (SERP) – refers to a list of web pages returned by a search engine in response to a keyword query. Each element of this list, or each listing, typically includes a title, a URL or link to the webpage, and a brief description showing where the keyword matched the content within the page. A SERP can refer to a single webpage containing a sequence of paid and free listings, or possibly a set of all links returned for a search query spanning multiple web pages.

[0021] Landing page – Refers to a web page whose URL corresponds to a listing within the SERP. When a user clicks on a listing within the SERP, the web browser requests and displays the corresponding landing page.

[0022] Cannibalistic ads refer to paid advertisements provided by search engines on web pages in response to user searches, diverting clicks from nearby free listings. Generally, compared to cases where cannibalistic ads appear in the SERP and there are no corresponding free listings, cannibalistic ads have reduced value when they appear adjacent to or close to corresponding free listings. In this context, value is typically measured in terms of visitors, sales revenue of the advertised product, or similar metrics. Additionally, corresponding free listings are free listings that refer to the same product or service advertised by the paid ads. As described below, links within paid ads and free listings may refer to the same landing page or different landing pages.

[0023] Generalized behavior

[0024] The operation of a specific aspect of the present invention is described below in relation to FIGS. 1 to 4.

[0025] FIG. 1 is a simplified block diagram of a cannibalistic ad system (CAS) (100) that automatically identifies cannibalistic ads and calculates potential improvements that can be achieved by not purchasing such ads.

[0026] A user (110) visits a website that enables them to perform a keyword search using a web browser, such as Google Chrome or Mozilla Firefox, or other client application, referred to herein as a browser (118). The browser (118) transmits the keyword to a search engine (120) that performs the requested search and returns a SERP displayed by the browser (118). The SERP generally includes one or more paid listings or advertisements and one or more free listings. Each advertisement or free listing corresponds to a web page, also referred to as a landing page, which is determined by the search engine (120) to match the keyword. In the case of a paid advertisement, the advertiser "purchases" the keyword and provides the corresponding advertisement to an ad server (130) that communicates with the search engine (120). Thus, when the user (110) enters the keyword into the search box, the search engine includes the corresponding paid advertisement in the SERP returned to the browser (118).

[0027] A landing page belongs to a domain or website (145) hosted by a web server or web service, simply referred to as a web server (140). The web server (140) may host multiple domains. The web page may be static, that is, exist as a computer file in HTML format or another format, or may be dynamically generated. Additionally, the web server (140) may provide e-commerce, enabling the user (110) to purchase items or otherwise perform a transaction that generates revenue from the website (145).

[0028] The ad server (130) provides paid advertisements to be included in the SERP to the search engine (120). The cannibalistic ad (CA) analyzer (135) analyzes the SERP and identifies cannibalistic ads. Generally, the CA analyzer (135) generates a list of keywords where cannibalistic ads can be placed when purchased. The operation of the CA analyzer (135) is described in more detail below with reference to FIGS. 2 through 4.

[0029] It may be recognized that the CA analyzer (135) may operate on a server or computer system different from the advertising server (130). Additionally, the advertising server (130) may be implemented as more than one physical server computer or, for example, by a cloud service such as Amazon AWS. Additionally, the CA analyzer (135) may be implemented as more than one physical server computer or, for example, by a cloud service such as Amazon AWS.

[0030] The network (150) enables various computers, servers, and services identified in the CAS (100) to exchange data. The network (150) generally refers to the public internet, but may refer to a private network or any combination of private and public networks.

[0031] FIG. 2 is an example of a search engine result page (SERP) (200) containing cannibalism advertisements. In response to a user (110) entering the search term (215) "surebuy grocery" into a browser (118), the search engine (120) returns a SERP (200). The SERP (200) includes paid listings (210) and free listings (220), which are advertisements for a grocery store or a chain of grocery stores named "surebuy grocery store".

[0032] Listing (210) is the first and only paid advertisement included in the SERP (200). Listing (220) is a free advertisement (220). The free advertisement (220) is in the first free listing position within the SERP (200).

[0033] The ad (210) is a cannibalistic ad because it is located directly above the free listing (220). If the ad (210) does not appear above the free listing (220), a larger percentage of users will click on the free listing (220). Therefore, the advertiser paid for the ad (210) even in cases where users clicked on the free listing (220) when the ad (210) was not in the SERP (200). Thus, the placement of the ad (210) in a position directly above the free listing (220) has correspondingly increased the advertising costs.

[0034] In the case of cannibalistic advertising

[0035] The case of FIG. 2, where a free listing appears immediately below a paid advertisement for the same product or service and the paid advertisement is in the first position, is considered a primary case of cannibalistic advertising. In this case, there is only one paid advertisement, that is, a cannibalistic advertisement. However, there are other cases where an advertiser may consider an advertisement to be cannibalistic.

[0036] Several factors may be considered to determine whether paid advertising is cannibalistic. These factors include (1) location, (2) distance, (3) appearance rate or frequency, (4) whether the landing page is the same or different, (5) whether it is a "friendly" ad or a "competitive" ad (or neither), and (6) incremental or net revenue generated from the ad placed by the advertiser compared to revenue generated from natural search results. Factors 1 through 5 are discussed first, and a rule-based approach to evaluating whether an ad is cannibalistic based on these factors is presented.

[0037] Position refers to the location of a paid advertisement or a corresponding free listing in the search results. As mentioned, the primary case is when the paid advertisement and the free listing are in the first position, that is, when the paid advertisement is in the first position among the paid advertisements and the free listing is in the first position among the free listings. However, a paid advertisement in the second or third position and a free listing in the first position may be considered cannibalistic in some cases.

[0038] Distance refers to the number of paid and free listings between paid ads and free listings. Distance can be determined from the location of paid ads, the total number of paid ads, and the location of free listings as follows:

[0039] Distance = (# PA - PA Location) + UL Location Equation 1

[0040] Here, # PA is the number of paid advertisements, the PA location is the location of the paid advertisement being analyzed among all paid advertisements, and the UL location is the location of the corresponding free listing among free listings. In this embodiment, it is understood that the paid advertisements appear sequentially on top of the free listings. However, if the paid advertisements appear on the side of the web page or at another geometric location, a similar distance measurement can be formulated.

[0041] The occurrence rate or frequency refers to the fact that the position of advertisements and free listings may change with each search. Accordingly, in some embodiments, search terms may be "sampled" over a period of time or over a number of repetitions to determine the average position of paid advertisements or free listings in the SERP provided in response to receiving specific keywords. For example, to obtain the position of paid advertisements or free listings in the received SERP, searches may be repeated once every minute or once every hour for a day or a week. Alternatively, searches may be repeated 100 times a day. Of course, other sampling methods may also be used.

[0042] In some cases, the landing page of a cannibalistic ad differs from the landing page of its corresponding free listing. In some cases, these ads are considered cannibalistic; in other cases, the advertiser may be testing the landing page or simply prefers to have users see their ads rather than the free listings generated by search engines.

[0043] "Friendly" advertising and "competitive" advertising

[0044] Advertisements can be classified as "friendly" or "competitive" advertisements for specific ads placed by an advertiser. Then, the category assigned to the advertisement can be used as part of determining whether the advertiser's advertisement is cannibalistic.

[0045] Example 1: In the first example, if the first company is a business partner of the second company, the first company may agree to treat advertisements by the second company as "friendly" and not advertise competitively with advertisements placed by the second company. Therefore, in this example, a friendly advertisement placed at any position among paid advertisements implies that the advertiser's own advertisement is cannibalistic. Several other examples are given below.

[0046] Example 2: Car dealer B sells cars manufactured by car manufacturer A. Manufacturer A considers the advertisement by dealer B for products from Manufacturer A to be favorable and may decide not to advertise when the advertisement by dealer B appears.

[0047] Example 3: Alternatively, Manufacturer A may consider the advertisement placed by Dealer B for its product (i.e., Manufacturer A's product) to be competitive and want to advertise directly against such advertisement, that is, when the advertisement placed by Dealer B is statistically likely to appear in the SERP.

[0048] Example 4: An advertiser considers an ad to be cannibalistic if there are no competitive ads, regardless of the location of the corresponding free listing.

[0049] More generally, advertisers may consider advertisements placed by a particular company or organization to be friendly or competitive and may implement advertising rules based on such determination.

[0050] Additionally, advertisers can easily determine the landing page domains for advertisements within the SERP by scraping and analyzing the SERP. Thus, a favorable advertisement can be considered as an advertisement having a landing page on a favorable domain, and a competitive advertisement can be considered as an advertisement having a landing page on a competitive domain. Accordingly, in certain embodiments, favorable advertisements and competitive advertisements can be determined based on a list of favorable domains and a list of competitive domains. In other embodiments, company names, organization names, or even product names may be used to determine whether an advertisement is favorable or competitive.

[0051] Therefore, from the advertiser's perspective, each advertisement within the SERP can be classified as follows: 1) self-advertising, 2) favorable advertising, 3) competitive advertising, and 4) others, that is, advertisements from companies, organizations, or domains other than the advertiser itself that are neither favorable nor competitive.

[0052] In a specific embodiment, the CA analyzer (135) analyzes only the first SERP returned by a search for a search term. Generally, there is a maximum number of paid ads in a SERP; for example, the Google search engine returns a maximum of 4 ads in a SERP. Thus, the most frequently returned ad for each ad position can be determined. In the following discussion, it is assumed that the most frequently returned ad for each SERP position is determined by the CA analyzer (135) and that the category of the ad is similarly determined.

[0053] In another embodiment, each advertisement may be categorized or classified in a more general manner, that is, any number of categories other than friendly and competitive categories may be used.

[0054] Rules for determining cannibalistic ads

[0055] Paid advertisements can be evaluated using rules that evaluate cannibalistic scores or measurements. Rules can be formulated based on previously discussed factors, namely: (1) categories assigned to paid advertisements within the SERP, (2) the number of paid advertisements within the SERP, (3) the average distance between an advertiser's ad and a corresponding free listing, and (4) the average position of the corresponding free listing.

[0056] Table 1 provides an example of an approach to formulating rules that can be used to evaluate whether an ad is cannibalistic using the previously described factors of ad location, ad category, distance, and the location of a corresponding free listing.

[0057] In Table 1, each row represents a rule, and the columns are as follows: A. is the rule number, B. is the category of ads appearing in the first position in the first SERP, C. is the category of ads appearing in the second position in the first SERP, D. is the average position of corresponding free listings for ads placed by advertisers, E. is the average distance between advertisers' ads and corresponding free listings, F. indicates whether advertisers' ads are considered cannibalistic (yes) or not (no) when the rule is satisfied, and column G provides a brief descriptive description of the rule. Additionally, the categories that can be assigned to ads in this example are A-Advertiser, F-Friendly, C-Competitive, and O-Other.

[0058] In the example in Table 1, columns D and E are shown as having integer values, but in certain embodiments, they may be real or decimal values ​​based on sampling values ​​obtained for the D-average position of the corresponding free listing and the E-average distance from the advertiser's ad to the corresponding free listing. Additionally, columns B, C, and F may alternatively have percentage values, decimal values, or fractional values. For example, the category values ​​for the first-place listing in column B may be A(.75), F(.1), C(.1), and O(.05), representing the proportion or percentage of time the first-place ad is in category A, F, C, or O.

[0059] In certain embodiments, a favorable ad is considered to be an ad located on a domain where the landing page is considered favorable, whereas in other embodiments, it may be further recognized that there may be specific rules for specific domains. For example, Rule 5 is evaluated based on whether an ad having a link to a specified domain (Dom A) is the first or second paid ad location and whether the corresponding free ad is less than 3 distance from the advertiser's ad.

[0060] Finally, when more than one rule is applied to an ad, the cannibalism score represents the sum, average, or weighted average of the results of all rules applied to the ad. For example, if one rule is evaluated at 75% and another at 25%, in the simplest case, the average value of 50% is the cannibalism score for the ad.

[0061] Rules based on incremental ad value

[0062] There may be a difference between the revenue generated for advertisers from ads and the revenue generated from corresponding free listings. This can occur when ads and free listings link to different landing pages, as different landing pages may have different levels of effectiveness. Consequently, if ads perform worse on average than corresponding free listings, revenue is reduced by cannibalizing clicks on free listings from the average ad clicks. Using data sources provided by search engines, such as Google Analytics, Google AdWords, and Google Search Console—as explained in more detail in Table 2 below—you can determine the average value or revenue from ad clicks and the average value or revenue from clicks on corresponding free listings. Average click value refers to the expected revenue per click or visit by a visitor to the linked landing page, i.e., the average revenue.

[0063] One rule that can be defined by the advertiser is based on the incremental or comparative value of ad clicks versus free listing clicks as defined in Equation 2 below:

[0064] [(Value of an ad click - Value of a free click)] / Cost of an ad click Equation 2

[0065] Here, if the incremental value defined by Equation 2 is greater than 1, there is a net positive revenue from ad purchases after considering the cost of ad purchases. Therefore, a simple rule based on Equation 2 is that if the incremental value is less than 1, the ad is cannibalistic. However, other rules may also be considered, and the rule based on the incremental value (Equation 2) may be combined with rules such as those exemplified in Table 1. Additionally, other formulas for defining the incremental value may be defined without departing from the scope of the present invention.

[0066] It will be understood that each block of the flowchart exemplified in FIG. 3 and the combination of blocks in the flowchart example may be implemented by computer program instructions. Such program instructions may be provided to a processor to produce a machine such that instructions executed on the processor create means for implementing the actions specified in the flowchart blocks or blocks. Computer program instructions may be executed by a processor to cause a series of action steps that implement the actions specified in the flowchart blocks or blocks to be performed by the processor to create a computer-implementable process or method. Computer program instructions may also cause at least some of the action steps indicated in the blocks of the flowchart to be performed in parallel. Additionally, some steps may be performed across two or more processors, as may occur, for example, in a multiprocessor computer system. Furthermore, one or more blocks or combinations of blocks in the flowchart example may be performed simultaneously with other blocks or combinations of blocks, or even in a different order than exemplified without departing from the scope or spirit of the present invention.

[0067] Accordingly, the blocks of the flowchart example support combinations of means for performing a specified action, combinations of steps for performing a specified action, and means of program instructions for performing a specified action. Furthermore, it will also be understood that each block of the flowchart example and combinations of blocks within the flowchart example may be implemented by a special-purpose hardware-based system for performing a specified action or step, or by a combination of special-purpose hardware and computer instructions.

[0068] FIG. 3 provides a complete method (300) for identifying cannibalistic advertisements and generating a cannibalism score for an advertisement. The purpose of the method (300) is to evaluate search engine advertisements being purchased by advertisers to determine which ones are cannibalistic.

[0069] In step (305), the CA analyzer (135) collects a set of search terms to be evaluated. These may be advertisements currently being purchased by advertisers or potential advertisements. Generally, each advertisement corresponds to a search term or keyword. This step can be performed in various ways. For example, if a search engine marketing (SEM) program of a specific advertiser, e.g., SureBuy (see FIG. 2), is being evaluated, the advertiser may provide a list of search terms that they typically purchase and corresponding paid advertisements appearing in the result SERP.

[0070] Alternatively, search engines can provide keyword sets. For example, Google Search Console, provided by Google, can provide search term listings and the average search position of ads placed according to each search term. Thus, for example, if you are interested only when paid ads appear in the first and second positions and free listings are in the first or second position, only search terms with an average position of less than 2 may be evaluated.

[0071] Additional data can be obtained from search engines. For example, Google AdWords, provided by Google, offers information including ad placement, ad cost, click-through rate, ad impression share, and similar metrics.

[0072] In step (310), rules for evaluating and scoring advertisements are generated. These rules may be similar to the rules given in Table 1 or Equation 2.

[0073] A keyword search is performed on a search engine for one of the search terms collected in step (315).

[0074] In step (320), the SERP is received from the search engine. Note that while the method (300) is applied to a single search engine, the method can be performed on additional search engines of interest. Thus, the method (300) is applied equally to all search engines.

[0075] In step (325), listings (paid and free) within the received SERP are analyzed to determine the listing positions of paid advertisements by advertisers and the listing positions of corresponding free listings (if any). In a specific embodiment, the search is performed iteratively, that is, sampling is performed to obtain the average listing positions. In this case, the average listing positions for paid advertisements and corresponding free listings are updated at this step.

[0076] In step (330), if sampling of the average listing position is performed, a decision is made as to whether more sampling should be performed. If so, the process returns to step (315); otherwise, the process proceeds to step (335).

[0077] In step (335), the rule formulated in step (310) is applied to generate a cannibalism score for the paid advertisement. The cannibalism score may have various meanings. For example, in a specific embodiment, the cannibalism score estimates the percentage or amount of sales or revenue lost due to the proximity of the paid advertisement to the corresponding listing within the SERP.

[0078] In another embodiment, the cannibalism score indicates the likelihood that the advertisement is actually cannibalistic and will reduce the revenue generated by the free listing. In such an embodiment, a cannibalism score of 75 indicates that there is a 75% likelihood that the advertisement is cannibalistic.

[0079] In another embodiment, the score may be a Boolean value that simply indicates whether the advertisement is considered cannibalistic.

[0080] If not all keywords are processed in step (340), processing returns to step (315). If all keywords are processed, processing continues in step (345).

[0081] In step (345), a report on the cannibalism score for each keyword-paid ad combination is generated. This report may include all collected keywords or only keywords determined to be cannibalistic. For example, only keyword-paid ad combinations with a score higher than a given threshold may be included in the report. In some cases, the method is terminated at this step and the report is provided to the advertiser or a service company designated by the advertiser, such as an online advertising company.

[0082] In another embodiment, in step (350), the advertiser or their web service company may modify their search engine ad purchases based on the report generated in step (345).

[0083] Efficiency Measurement: Reclaimed Ad Spend

[0084] Improving the efficiency of advertising campaigns by automatically analyzing the concept of cannibalistic ads and whether paid ads are likely to be cannibalistic is novel. Using this measurement can significantly reduce campaign costs and improve efficiency. Therefore, it is important to be able to measure the reduced costs and increased efficiency achieved by not purchasing cannibalistic ads. This efficiency measurement can be provided to advertisers to verify the effectiveness of using automated methods to identify cannibalistic ads and take appropriate measures known as de-cannibalizing actions.

[0085] FIG. 4 provides a method (400) for determining an efficiency measure for an advertising campaign based on an estimate of potential savings resulting from not purchasing some or all of the cannibalistic ads identified by the method (300). The method (400) calculates a measure referred herein as “advertising cost recovery,” which estimates savings resulting from not purchasing one or more paid ads determined to be cannibalistic. The method (400) uses the cannibalism score determined by the method (300) to determine whether or not to purchase individual ads.

[0086] In a specific embodiment, the method (400) calculates a reclaimed ad spend (RAS) for a set of keywords for one day (or another suitable period), wherein the reclaimed ad spend for a search term for one day may be defined as in Equation 3 below:

[0087] RAS = Estimated Ad Cost - Optimized Ad Cost Equation 3

[0088] In step (430), the estimated advertising cost (EAS) for one day can be calculated as follows:

[0089] EAS = Average Cost-Per-Click (CPC) * Average Click-Through Rate (CTR) * Daily Keyword Impressions (i.e., Number of times users searched for the keyword) Equation 4

[0090] It is understood that while a period of one day is used in certain embodiments, other periods may be used in other embodiments of the present invention.

[0091] Optimized ad spend (OAS) is the actual measured ad spend or cost generally reported by the search engine during the period when decannibalizing measures are being taken, that is, during the period when the cannibalization method (300) is in operation and the decision on whether to purchase keywords to run paid ads is based at least partially on the cannibalization score for the keywords. The operation of the method (300), which includes deciding not to purchase cannibalistic ads, is also referred to as taking decannibalizing measures. An example of a decannibalizing measure is not purchasing paid ads when the cannibalization score is higher than a threshold value.

[0092] For example, when the method (400) is performed on an advertisement purchased from the Google search engine, a measurement of the advertising cost is obtained from a service called Google Ads provided by Google.

[0093] For example, regarding the estimated ad cost per search term per day (Equation 4), when the CPC is $1, the CTR is 10%, and the estimated impressions for the day are 2,000, the estimated ad cost for the day (EAS) is = $1 * 0.1 * 2,000 = $200.

[0094] Therefore, if the actual optimized ad cost reported by the search engine is $50, the ad cost recovered is $200 - $50 = $150.

[0095] In a specific embodiment, the method (400) operates using only a single search engine, e.g., the Google search engine. In such an embodiment, the method (400) may operate using the data presented in Table 2 below. In another embodiment, comparable data is obtained from other search engines or from additional search engines to a single principal search engine. Then, results can be aggregated across multiple search engines.

[0096] Data Sources and Inputs Data source specific input Google Ads API It provides metrics for keywords, including impressions, clicks, average click-through rate (CTR), and cost per click (CPC). Google Search Console API Provides search traffic data for keywords over a specified period, including organic clicks, organic impressions, and landing page URLs. Google Analytics API Provides revenue and traffic sources for the date. Decannivalization system (i.e., result from method (300)) Provides cannibalization scores for paid advertisements for a date or period.

[0097] Some terms used in Table 2 are defined as follows:

[0098] Traffic sources refer to how visitors arrived at the site, for example, by clicking on a natural listing or by clicking on a paid advertisement.

[0099] An ad click refers to the number of times a user clicks an ad corresponding to the search term being evaluated, which is displayed in the SERP returned as a keyword search result using the Google search engine.

[0100] Advertising Cost: Dollars spent by advertisers for paid advertisements appearing on the SERP.

[0101] Organic clicks refer to the number of times a user clicks on a free, "natural" Google search result listing.

[0102] Organic impressions refer to the number of times a free listing appears in the SERP in response to a Google search performed by a user.

[0103] The baseline period refers to the period during which no decannibalization action is taken for a search term, that is, the purchase decision does not take the cannibalization score for the search term into account. The baseline period is typically measured in days. This can be several days within a larger time interval, for example, four days out of the last eight, or several consecutive days.

[0104] Baseline data refers to data captured during a reference period.

[0105] Additional measurement of efficiency

[0106] In addition to recovered ad costs, the efficiency of spending reduction on cannibalistic ads can be indicated using additional or alternative measurements. Each of the following measurements can be compared between a period when decannibalization measures were not taken and a similar period when decannibalization measures were performed: ad clicks, ad costs, organic clicks, impressions, and revenue. Data can be aggregated across all search terms and over different time intervals (days, weeks, months, etc.).

[0107] Returning to method (400), keywords to be evaluated in step (405) are received or collected. Next, in step (410), the corresponding cannibalism scores generated by method (300) and the performance data discussed above in Table 2 are collected for each received keyword.

[0108] In step (420), keywords are selected for processing. In step (430), the estimated advertising cost for the keywords is calculated according to Equation 4. Then, in step (440), the optimized advertising cost is retrieved from the data collected in step (410). As previously mentioned, the optimized advertising cost is generally collected from search engines.

[0109] In step (450), the recovered advertising cost is calculated according to Equation 3. Additionally, in step (450), other measurements as discussed above may be calculated, particularly including ad clicks, ad costs, organic clicks, impressions, and revenue. As a basis for comparison, these measurements may be calculated or collected over a period of comparable length between when the decannibalizing measure was taken and when it was not.

[0110] In step (460), a decision is made as to whether all keywords have been processed. If not, processing returns to step (420), and if so, processing proceeds to step (470).

[0111] In step (470), optionally, results including recovered advertising costs are aggregated across all keywords and potentially over several time intervals.

[0112] In step (480), results are provided to the customer or client. These may be provided in the form of a file, such as a Microsoft Excel table, for example, or as a presentation. Additionally, these data may constitute intermediate results and may be further analyzed and used for reporting or decision-making purposes.

[0113] FIG. 5 is a block diagram depicting software modules of a cannibalistic ad system (CAS) (100). FIG. 5 illustrates related software elements of the CAS (100), including a client computer (115), a search engine (120), a web server (140), and an ad server (130).

[0114] The client computer (115) interacts with the user (110) and enables the user (110) to perform a web search using a web browser (118).

[0115] A browser (118) is typically a commercially available standard browser such as Mozilla Firefox or Microsoft Internet Explorer. Alternatively, it may be a client application configured to receive and display graphics, text, multimedia, etc. over a network.

[0116] A browser (118) issues an HTTP request to an internet-connected computer, such as a search engine (120), a web server (140), and a client computer (115), and receives an HTTP response. An application server (420) receives the HTTP request and calls an appropriate ad server (130) software module to process the request. The application server (520) may be a commercially available application server that includes a web server that accepts and processes the HTTP request and sends back the HTTP response along with optional data content, and the optional data content may be a web page, such as an HTML document and a linked object (image, etc.).

[0117] The application server (520) establishes and manages sessions with the search engine (120) and the web server (140). Additionally, the application server (520) can interact with the client computer (115).

[0118] Software modules of the search engine (120) are generally outside the scope of the present invention. However, as discussed above and described in detail in Table 2, the search engine is expected to be purchased by advertisers and provide various result data related to advertisements provided by the SERP in response to searches. The web server (140) manages one or more websites (145), each of which includes one or more domains.

[0119] The ad server (130) includes a keyword collector (530), a rule definer (532), a cannibalistic ad (CA) analyzer (135), a cannibalistic ad (CA) report generator (534), an ad buyer (536) in a specific embodiment, a reclaimed ad analyzer, a keyword database (550), a rule database (552), and an ad database (554). It may be recognized that each of the databases mentioned above may be implemented as one or more computer files distributed across one or more physical storage mechanisms. In one embodiment, each of the aforementioned databases is implemented as one or more relational databases and accessed using a structured query language (SQL).

[0120] The keyword collector (530) obtains keywords from the search engine (120) and potentially from other sources. The keyword collector (530) may also obtain keywords from advertisers, for example, from computer files supplied by advertisers. The keyword collector (530) stores the keywords in the keyword database (450). The keyword collector (530) performs processing associated with step (305) of the method (300).

[0121] Additionally, the keyword collector (530) collects data required by the method (400) to calculate the return on advertising costs, including revenue from keywords, average cost per click (CPC), average click-through rate (CTR), and impressions from the search engine (120).

[0122] The rule definer (532) defines rules for determining scores for advertisements in the SERP. The rule definer (532) stores rules in the rule database (552). The rule definer (532) implements step (310) of the method (300). The rule definer (532) may be implemented in various ways; for example, in a specific embodiment, the rule definer (532) simply receives a text file that defines rules; whereas in another embodiment, the rule definer (532) provides a graphical interface to a client computer (115) that allows a user to define rules interactively. Generally, the method for defining rules is outside the scope of the invention.

[0123] The CA analyzer (135) performs processing associated with steps (315 to 340) of the method (300). The CA analyzer (135) uses collected keywords stored in the keyword database (550) to obtain SERPs, determines the locations of paid advertisements and corresponding listings, and stores the results in the keyword database (550). Additionally, the CA analyzer (135) evaluates the SERPs to generate cannibalistic ad scores and stores the results in the ad database (554).

[0124] The CA report generator (534) generates a report providing a cannibalistic ad score for an ad. The CA report generator (534) performs step (345) of the method (300).

[0125] In a specific embodiment, an ad buyer (536) purchases ads from a search engine (120) by considering the results of a cannibalistic ad analysis represented by a cannibalistic ad report stored in an ad database (554). Specifically, the ad buyer (536) determines on a case-by-case basis whether to purchase ads based on a cannibalism score. In another embodiment, the ad buyer (536) is not part of the CAS (100). For example, the function of the ad buyer (536) may be performed by a third-party ad agency.

[0126] The reclaimed ad cost analyzer (RAS analyzer) (538) analyzes the results obtained by not purchasing ads considered cannibalistic, namely, efficiency. Generally, the RAS analyzer (538) performs the method (400) to generate reports or other results that may be provided to advertisers. The RAS analyzer (538) may also calculate other efficiency measures other than reclaimed ad costs, as previously discussed. These measures include, among others, ad clicks, ad costs, organic clicks, impressions, and revenue over a period of comparable length when decannibalization measures were taken and when they were not. The RAS analyzer (538) stores the results in the ad database (554).

[0127] The keyword database (550) stores collected keywords. The keyword database (550) also stores the location or average location of advertisements and corresponding free listings, where the advertisements are placed by the advertiser, by the ad server (130), or by another party acting on behalf of the advertiser, and the advertisements appear in the SERP returned as a result of the keyword search.

[0128] The rule database (552) stores the rules used to generate cannibalism scores for advertisements.

[0129] The advertising database (554) stores advertisements supplied by advertisers corresponding to collected keywords. Generally, each keyword of interest to an advertiser has a corresponding paid advertisement that can be placed using a search engine. The advertising database (554) also stores CA reports and recovered advertising costs generated by the CA report generator (534), and other result data generated by the RAS analyzer (538).

[0130] The foregoing specification, examples, and data provide a complete description of the manufacture and use of the composition of the present invention. Many embodiments of the present invention may be made without departing from the spirit and scope of the present invention.

[0131] Exemplary rules for determining whether an advertisement is cannibalistic A. Rule # B. Category Ad 1 C. Category Ads 2 D. Free Ad Locations E. Street F. Is it cannibalistic? G. Rule Explanation 1 A Not applicable 1 1 yes (Cannibalistic) A typical case of cannibalistic advertising. The advertiser's ad is in the first position, and free listings are generally in the first position. Typically, there is no ad in the second position. 2 A F 1 2 yes (Cannibalistic) Advertisers' ads are in the first position; second-position ads are friendly ads; and free listings are generally in the first position. 3 A C 1 2 yes (Cannibalistic) The advertiser's ad is in the first position; the second ad is a competitive ad; free listings are generally in the first position. 4 A C 2 3 no (Not cannibalistic) The advertiser's ad is in the first position; the second ad is competitive; the free listing is in the second position. 5 Domain A Domain A Not applicable <3 no (Non-cannibalistic) An ad that is not an advertiser's ad appears in the first or second paid ad position, and the advertiser's corresponding free listing is less than 3 distances from the advertiser's ad.

Claims

Claim 1 A computer-implemented method for generating a cannibalism score for an advertisement, comprising the step of collecting a plurality of keywords associated with a specified advertiser, wherein in response to receiving the keywords, a search engine returns a search engine results page (SERP), and the SERP comprises (1) one or more advertisements—the advertisements are placed by the advertiser and include a link to a landing page, and the advertisements have ad positions within the SERP where higher positions are more valuable than lower positions—and (2) a plurality of unpaid listings—each unpaid listing includes a link to a landing page, and the unpaid listings have unpaid listing positions within the SERP where higher positions are more valuable than lower positions—and the step of maintaining at least one cannibalism rule for calculating a cannibalism score for an advertisement, wherein the cannibalism score for an advertisement is at least partially (1) an ad position of an advertisement within the SERP and (2) the same SERP A step of maintaining at least one cannibalism rule, wherein if an ad generates reduced value due to the free listing position of a corresponding free listing within, the ad is cannibalistic, and the corresponding free listing refers to the same product or service as that referred to by the ad; for each received keyword: a step of providing the keyword to a search engine; a step of receiving a SERP from the search engine - the SERP includes one or more ads and one or more free listings -; a step of determining the ad position of the first ad placed by the specified advertiser among the one or more ads; a step of determining the free listing position of the corresponding free listing among the one or more free listings;A computer-implemented method for generating a cannibalism score for an advertisement, comprising the step of generating a cannibalism score for the first advertisement by applying at least one cannibalism rule to the first advertisement and the corresponding free listing. Claim 2 A computer-implemented method for generating a cannibalism score for an advertisement, wherein the cannibalism score represents a measure selected from a group consisting of an estimate of the advertiser's revenue loss due to the advertisement appearing in the same SERP as a corresponding free listing, an estimate of the likelihood that the advertisement is cannibalistic, or a BOOLEAN value indicating whether the advertisement is cannibalistic. Claim 3 A computer-implemented method for generating a cannibalism score for an advertisement, wherein, in claim 1, the primary case of a cannibalistic advertisement is when only one paid ad exists and the corresponding free listing appears immediately below said paid ad. Claim 4 A computer-implemented method for generating a cannibalism score for an ad, wherein each rule is based on one or more factors selected from the group consisting of the ad position within the SERP, the number of ads within the SERP, the position of a corresponding free listing within the SERP, the average revenue per click for the ad, and the average revenue per click for the free listing. Claim 5 A computer-implemented method for generating a cannibalism score for an advertisement, wherein, in claim 1, each advertisement within the SERP may be assigned a category selected from a group consisting of advertisements placed by the designated advertiser, friendly ads, competitive ads, or other advertisements, and at least one rule is based on the category of the advertisement within the SERP. Claim 6 A computer-implemented method for generating a cannibalism score for an advertisement, wherein, in claim 1, the distance between an advertisement and a free listing within a SERP can be calculated, and at least one rule in the sequence of rules is based on the distance between an advertisement placed by the advertiser and the corresponding free listing. Claim 7 A computer-implemented method for generating a cannibalism score for an advertisement, further comprising the step of generating a report including a cannibalism score for each advertisement placed by the specified advertiser that appears in the received SERP. Claim 8 A computer-implemented method for generating a cannibalism score for an advertisement, wherein, in claim 7, the method further comprises the step of purchasing an advertisement from the search engine based at least partially on the report of the cannibalistic advertisement. Claim 9 As a network computing device, a keyword collector for collecting multiple keywords associated with a specified advertiser, wherein in response to receiving keywords, a search engine returns a search engine result page (SERP), and the SERP comprises (1) one or more advertisements—the advertisements are placed by the advertiser and include a link to a landing page, and the advertisements have ad positions within the SERP where higher positions are more valuable than lower positions—and (2) multiple free listings—each free listing includes a link to a landing page, and the free listings have free listing positions within the SERP where higher positions are more valuable than lower positions—the keyword collector; and a rule database for maintaining at least one cannibalism rule for calculating a cannibalism score for an advertisement, wherein the cannibalism score for an advertisement indicates that the advertisement is cannibalistic if it generates reduced value due to at least partially (1) the ad position of the advertisement within the SERP and (2) the free listing position of the corresponding free listing within the same SERP, and the corresponding free listing is a product identical to that referred to by the advertisement or A network computing device comprising: a service; and a CA analyzer for processing each of the received keywords; wherein the processing comprises: providing the keywords to a search engine; receiving a SERP from the search engine; determining the ad position of the first ad placed by the designated advertiser among one or more ads within the SERP; determining the free listing position of the corresponding free listing among a plurality of free listings within the SERP; and generating a cannibalism score for the first ad by applying at least one cannibalism rule to the first ad and the corresponding free listing. Claim 10 A network computing device, wherein, in paragraph 9, each rule is based on one or more factors selected from the group consisting of an ad position within the SERP, the number of ads within the SERP, the position of a corresponding free listing within the SERP, the average revenue per click for the ad, and the average revenue per click for the free listing. Claim 11 A network computing device according to claim 9, wherein each advertisement within the SERP may be assigned a category selected from a group consisting of advertisements placed by the designated advertiser, friendly advertisements, competitive advertisements, or other advertisements, and at least one rule is based on the category of the advertisement within the SERP. Claim 12 A network computing device according to claim 9, wherein the distance between an advertisement and a free listing within the SERP can be calculated, and at least one rule in the sequence of rules is based on the distance between an advertisement placed by the specified advertiser and the corresponding free listing. Claim 13 A network computing device according to claim 9, further comprising a cannibalistic ad report generator that generates a report including a cannibalism score for each ad placed by the specified advertiser appearing in the received SERP. Claim 14 In paragraph 13, a network computing device further comprising an ad buyer for purchasing ads from the search engine based at least partially on the above report of cannibalistic ads. Claim 15 delete Claim 16 delete Claim 17 delete Claim 18 delete Claim 19 delete Claim 20 delete Claim 21 delete Claim 22 delete Claim 23 delete Claim 24 delete Claim 25 delete Claim 26 delete Claim 27 delete Claim 28 delete Claim 29 delete

Citation Information

Patent Citations

  • System and method for optimizing paid search advertising campaigns based on natural search traffic

    US8396742B1