Various objects and advantages and a more complete understanding of the invention are apparent and more readily appreciated by reference to the following Detailed Description and to the appended claims when taken in conjunction with the accompanying Drawings wherein:
The invention generally relates to a system and method for modeling and/or optimizing, in the context of a search engine marketing (“SEM”) campaign, the value of one or more referral modes through which an Internet user is referred to or otherwise enters a particular website. The SEM campaign may, for example, comprise a search engine optimization (“SEO”) initiative and/or a search engine advertising (“SEA”) campaign (e.g., a pay-per-click and paid inclusion campaign). Embodiments of the invention permit advertising entities to assess the value of specific referral modes based on reconfigurable metrics and flexible, relative weightings of each metric.
As used herein, “value” pertains to any measurable commercial value pertaining to one or more referral modes.
As used herein, “referral mode(s),” “mode(s) of referral” or any variation thereof pertain, directly or indirectly, to the mode(s) or process(es) through which an Internet user enters or uses a website or webpage of interest. For example, a referral mode may comprise a particular keyword entered by an Internet user into a search engine. Upon entry of the keyword, the search engine displays organic search results and/or a paid search results that may list the webpage of interest. The user may then click on a web link associated with the webpage to enter or use the webpage. Thus, since the keyword is at least indirectly associated with the user's entry into the webpage, the value of the keyword (as a referral mode) can be determined.
In addition to a keyword, referral modes may comprise inbound links from other websites (other than search engines) and/or Internet-based advertisements (“ads”), including, e.g., text, image, video, and audio ads. In relation to an Internet-based ad, a user clicks on the ad, the user is connected to the website of interest and subsequently takes actions that result in measurable value. Thus, the Internet-based ad or the inbound link is at least one reason explaining why the user enters the webpage of interest.
Alternatively, referral modes may be described as actions taken by one or more Internet users in association with content offered at the webpage. For example, the action may include downloading or viewing content (e.g., text, image, video or audio). One of skill in the art will appreciate that a certain actions taken in association with content may directly or indirectly correspond to modes through which a webpage is entered and can thus be valued as referral modes.
Referral modes may also be described as a media ad viewings by Internet users prior to entering the webpage of interest. For example, the media ad may include text, image, video or audio ads available via the Internet, print media, and/or broadcast media, among others. The existence of a media ad viewing by a user may be determined via any number of methods within both the scope and spirit of the invention, including, e.g., an online survey-style entry by the user at the webpage of interest.
Referral modes may also be described as geographic, demographic, and/or temporal targeting of users prior to the users entering the webpage. Geographic, demographic and temporal targeting may be accomplished via any number of methods (e.g., delivering or making available particular media ads to particular geographic locations or particular demographics at particular times, delivering web links associated with the webpage of interest via email or screen pops, etc.). Geographic targeting may be based on a geographic area associated with the users. For example, the geographic area may be determined by a zip code, a city, a state, or a county associated with the users. Demographic targeting may be based on any number of categories, including, e.g., age, gender, race, or shopping history/preferences of users. Temporal targeting may be accomplished during a particular time period (e.g., during particular hours, days, weeks, months, years, etc.). By way of example, the existence of geographic, demographic or temporal targeting may be determined via any number of methods within both the scope and spirit of the invention, including, e.g., an online survey-style entry by the user at the webpage of interest.
Alternatively, by way of another example, the existence of geographic, demographic or temporal targeting may be determined in relation to a user clicking on an Internet-based ad. In one embodiment, data associated with the Internet-based ad may be stored, including data relating to the day the user clicked on the ad, the type of ad that was selected by the user, a keyword associated with the ad (if applicable), a geographical area to which the ad was targeted, and demographic information about the user that is available via any application capable of collecting information about the user.
For sake of clarity or presentation, embodiments of the invention described herein are directed to the valuation of referral modes in the form of keywords; however, one of skill in the art will appreciate alternative embodiments may be concerned with valuing referral modes other than keywords.
Aspects of the invention are designed to operate on computer systems, servers, and/or other like devices. While the details of the embodiments of the invention may vary and still be within the scope of the claimed invention,
Aspects of the invention may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer or server. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform particular tasks or implement particular abstract data types. The invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
As is shown, the network system 100 includes a communications network 110, such as the Internet or a private network, capable of providing communication between devices at search engine(s) 120, advertiser/client(s) 130, value modeling system 140, and third party user(s) 150 described hereinafter. The devices of
As shown in
As those skilled in the art will appreciate, various intermediary network routing and other elements between the communication network 110 and the devices depicted in
Referring again to
Computing devices at each of the third party users 150 may execute the web browser through which search terms may be entered via a search page representation provided by a search engine 120. Upon receiving the search terms from the third party user 150, the search engine 120 typically returns a plurality of search results to the third party user 150. The returned search results generally include links to web pages hosted by the websites of various business entities (e.g., the clients 130), thereby enabling the third party user 150 to view information from these web pages through the web browser executing on the third party user device 150.
In the case of the third party user 150 that clicks on a web link listed at the search engine 120, the database 123 stores information pertaining to the click such as the date and time of the click, the cost of the click, and the client 130 with which the link is associated. Information pertaining to subsequent clicks, by other third party users 150, of the client's web link is added to the database 123, and is then typically available to the client 130 and/or the value modeling system 140 in a report downloadable from the search engine 120.
After a third party user 150 clicks on a web link associated with a client 130, the third party user 150 is connected to the client's website associated with the web link. Upon this connection, one or more web analytics tools operating on a website server 131 track the website activity (e.g., usage and behavior) associated with the third party user 150. For example, the web analytics tool may track the number of page views, registrations, e-commerce sales, telephone sales, downloaded documents, multimedia views, and other activities associated with the third party user 150. Information associated with the website activity of the third party user 150 may be stored in a database 133, and is typically available as a report to the client 130 and/or to the value modeling system 140.
One aspect of the invention pertains to analyzing the effectiveness of a keyword purchase by a client 130 from a search engine 120. The measurement of effectiveness of a keyword purchase can, for example, be derived from any one of: a report from the search engine that includes, among other things, a listing of the purchased keyword and the number of “clicks” pertaining to the keyword for a given time period; a report from a web analytics tool that includes, among other things, a listing of the website activities associated with a third party user 150; and a combination of the search engine and web analytics tool reports. With respect to deriving the effectiveness of a keyword from a combination of the search engine and web analytics tool reports, the invention, as will be shown in the description of
Attention is now drawn to
Referring to
After identifying the tracking code in step 205, the client 130 and/or the value modeling system 140 appends the identified tracking code to the URL associated with a the keyword purchase transacted between the search engine 120 and the client 130 (step 210). The tracking code is added to the URL provided by the client 130 to the search engine 120 at the time of, or after, the keyword transaction.
In one embodiment, the tracking code includes information pertaining to the search engine 120 from which the keyword was purchased, as well as an indication of the keyword. Additional information may include indications of an advertisement (“ad”) group to which the keyword belongs, the type of advertising network, and the section of the website of the client 130 to which the URL pertains. One aspect of the invention enables the client 130 to utilize this invention with no additional website code or HTML tagging beyond that which is already present as a result of the requirements of any web analytics tools operating on the website of the client 130.
As is also shown in
The weight is a value assumption placed on a given performance indicator by the client 130 to represent the value of that performance indicator with respect to the commercial operations of the client 130. The weight, for example, may be measured in currency (e.g., the US dollar), a rating system, and/or other measurement parameters. As will be described in further detail hereinafter, the performance indicators and their assigned weights may be used in conjunction with the search engine and web analytics tool reports to build a formula for assessing the value of a keyword.
Attention is now drawn to
In an exemplary embodiment, the interface 300 is provided to the computing device 137 of the client 130 by the value modeling system 140 via the communication network 110. In another embodiment, the interface is generated locally at the client 130.
Attention is now turned to the main process 202 shown in
In steps 235 and 240, data is gathered from each of the search engine and web analytics tool reports, and then combined into a normalized master data set in step 245. Specifically, at step 245 the tracking codes identified and appended in steps 205-210 are used to match search engine data with associated web analytics data for SEA campaigns. For example, the tracking code “ppc_gg∥group|1|s” may represent the keyword “hairstyle” purchased from the search engine Google.
Data may be collected during a configurable instance of time, during a configurable period of time, or during configurable intervals of time. Additionally, collected data may be stored as historical data (e.g., in the database 143) and subsequently retrieved for comparison to collected data.
Once the master data set has been formed, it is stored at step 250 (e.g., in the database 143 at the value modeling system 140, or in the database 133 at the client 130). One of skill in the art will appreciate that more than one of the above steps may be omitted while staying within both the scope and spirit of the invention. For example, step 245 may not be required.
In an exemplary embodiment, steps 225-250 are performed by the value modeling system 140 via the communication network 110. In another embodiment, steps 225-250 are performed by the client 130. In yet another embodiment, steps 225-250 are performed by both the client 130 and the value modeling system 140.
Valuation of Keyword(s)
One aspect of the invention enables the client 130 to maximize return on investment (ROI) with respect to one or more keywords purchased from one or more search engines and/or one or more keywords pertaining to organic search results. Additional aspects may enable the client 130 to value keywords based on any number of metrics, including a cost per value point, a number of value points per visitor, a number of page views per visitor, a cost per page view, a cost per registration, a cost per download, a cost per video view, total cost, total revenue, total margin, a return on advertising spent (ROAS), margin per visitor, revenue per visitor, a cost per customer acquisition, a cost per click, and a click-through-rate.
Certain aspects of the invention allows the client 130 to effectively value its investment (i.e., a keyword purchase or a cost of optimizing a website to obtain a higher ranking in an Organic listing) based on parameters selected and weighted by the client 130. Another aspect of the invention enables the client 130 to identify unused or inefficient marketing strategies of which the client 130 may not be aware. Such strategies may be based on, for example, historical data, competitor bidding data and/or other data pertinent to identifying such strategies.
As shown in step 255 of
As shown in
In one embodiment, calculation of the gross keyword value is performed by multiplying (i) the weights of each of the performance indicators selected in steps 215-220 and (ii) respective web analytics data pertaining to those selected performance indicators. For example, if the client 130, during steps 215-220, selected ‘registrations’ to have a weight of $0.50, then the total number of registrations associated with the keyword of interest, as determined by the web analytics data, is multiplied by $0.50. The result is the gross value of the keyword with respect to registrations. During step 430, calculations similar to the one described in the example above are performed with respect to every performance indicator that was selected in step 215. Additionally, calculations may be performed on a per-third-party-user-basis or a per-visit-basis. Each gross value of these calculations is then summed and the resulting value corresponds to the gross total value of the keyword with respect to the performance indicators of importance to the client 130. As a result, total revenue associated with a visit to a website by third party user 150 occurring as a consequence of clicking a keyword advertisement at a search engine can be calculated as the sum of individual revenues associated with individual performance indicators selected by the client 130.
Additional revenue streams may also be calculated at step 430. For example, the client 130 may be content-focused rather than commerce-focused. A content-focused client 130 generates revenue by selling advertising on its website. Many content-focused clients 130 use the ‘revenue per 1000’ model, where advertisers on the client's website pay a set fee for every 1000 views of a webpage that includes their advertisement. The total revenue for each page view associated with the advertisement is calculated by dividing 1000 into the fees paid by a specific advertiser for a specific advertisement.
Either before, after or during steps 410-430, the value modeling system 140 accesses search engine data pertaining to the keyword of interest (step 440). The accessed search engine data may include, among other data, the cost of the keyword of interest for the time period in which the keyword value is being analyzed. At optional step 450, other cost data associated with the keyword is accessed. For example, the other cost data may include various business expenses associated with billed hours, resources used, transaction costs, and research and development costs attributable to the keyword. At step 460, the overall cost is calculated by adding the costs determined in steps 440-450.
Once the keyword value and keyword cost are determined, the value modeling system 140 determines the net/margin value of the keyword of interest (step 470). In one embodiment, the net keyword value is determined by subtracting the keyword cost from the gross keyword value. The result may then be used to create one or more static and/or interactive media displays (step 480) that may be charted for the client 130 as a function of time, search engine, and other discriminators, to provide a variety of actionable views for the client 130 to pursue optimizations of their search engine marketing strategy.
One aspect of the invention enables trending and graphing of individual keywords, search engines, campaigns, or other grouping techniques to compare relative performance and identify areas of optimization and performance improvement. For example, as shown in
At step 260, the client 130 may re-weigh the performance indicators selected in step 215 in order to analyze the value of a keyword using different weight parameters. The client 130 may also select and weigh a different group of performance indicators than those that were selected and weighed in steps 215-220. One advantage of step 260 is that it allows the client 130 to value the keyword based on different commercial metrics. The client is then enabled to compare and contrast different approaches to search engine marketing campaigns.
At any time the value modeling system 140 or the client 130 may take action based on the generated value models (step 265). For example, the value modeling system 140 may alert the client 130 (e.g., via email, a user interface, etc.) when the value of a keyword does or does not meet predetermined standards. The client 130 might choose to optimize its marketing campaign to reflect the assessed value of a keyword. A multitude of optimizations at the keyword and search engine level can be performed using the value of the keyword, such as lowering of a bid to increase keyword profitability, raising of a bid to capture additional clicks of the third party user 150, eliminating a keyword from a search engine to re-allocate budget to higher value keywords, or targeting a specific profit per keyword or search engine. Many variations, modifications and alternative optimizations can be performed using insight gained from the value model. Additionally, the value model system 140 may be configured to automatically adjust bids without requiring any manual input from the client 130.
For example, if the keyword value is negative or below a threshold value, or if a particular performance indicator is below a threshold value, the value modeling system 140 may recommend or automatically execute removal or lowering of a bid associated with the keyword at a particular search engine. Under some circumstances, the value modeling system 140 may recommend or automatically execute changing of the landing page associated with the URL of the web link at the search engine 120. Alternatively, if the keyword value is positive or above a threshold value, or if a specific performance indicator is above a threshold value, the value modeling system 140 may recommend or automatically execute increasing of a bid or the budget associated with the keyword. In some embodiments the value modeling system 140 may identify similar keywords and rotate them into the pay-per-click program of the client 130.
During a bid optimization process, the value modeling system 140 may compare a computed value of a particular keyword with values of that keyword for competitors of the client 130. In order to do so, the value modeling system 140 downloads bid landscape data from search engine application programming interfaces (APIs), including bid data pertaining to the competitors. The value modeling system 140 may also compare a computed value of a particular keyword with computed values of the same keyword based on higher or lower bid levels. Alternatively, the value modeling system 140 may compare a computed value of a particular keyword with historical values of the same keyword.
One aspect of the invention enables modeling and optimization based on frequently changing weights of multiple performance indicators in order to ensure such indications remain aligned with changing commercial needs. Any subset of these changing performance indicators can be used to establish the value of a keyword and build an appropriate value model for a specific time period. For example, cost rates for keyword advertisements, profit margins for items sold based on seasonal sales, lifetime value of customer or customer segments, and click-fraud rates at the various search engines or advertising networks may all change frequently. Embodiments of the invention are configured to enable these value assessments to be adjusted so as to reflect these dynamic changes.
In one embodiment of the invention, the value modeling system 140 performs fraud analysis to determine whether abuse exists within a sponsored search. For example, the value modeling system 140 may detect a spidering program that automatically selects (i.e., “clicks”) a website without visiting the website. In such a case, data pertaining to a number of visits to a website may be compared to the number of clicks associated with that website, and any disproportionate volumes of clicks when compared to number of visits may indicate fraud (e.g., 5000 clicks compared to 2500 visits). Alternatively, by way of example, the fraud analysis may use historical data (e.g., data collected in steps 235-240 of
At step 265, for example, the value modeling system 140 or the client 130 may turn off, lower or increase bids with respect to keywords and/or search engines having performance levels below or above predetermined thresholds. For example, a keyword at a poor performance level (e.g., a reported value in the bottom 20% of all keywords, or a reported value below a desired value) may be turned off or its bid may be drastically lowered. By way of another example, the bid level of a keyword with a good performance level may be adjusted to an optimal level, which may include setting the bid so as to obtain a maximum value (e.g., margin) with respect to the keyword. As the cost-per-click for a keyword increases, the reported value of the keyword decreases unless the additional cost-per-click is offset by increased revenue (or another type of value-based metric) generated via additional clicks.
Alternatively at step 265, the value modeling system 140 or the client 130 may examine advertisements and/or landing pages associated with keywords and/or search engines to perform a similar measurement of value for the keyword-advertising pair or the keyword-landing page pair.
One aspect of the invention pertains to predicting future value of a referral mode (e.g., a keyword). In accordance with one embodiment, a predictive future value of a keyword may be determined by analyzing historical values of the keyword (and in some cases, similar keywords). For example, a future value of the keyword may be achieved by trending the historical values (e.g., over time) and then assigning a future value in accordance with the trend (e.g., if the value of the keyword has a historical growth rate of 1%, the future value would be determined based on that growth rate).
In accordance with another embodiment, a predictive value of a keyword may be determined using a variety of historical/actual and/or estimated data. As one of skill in the art will appreciate, the following approach may be used to arrive at an actual value of a keyword, as opposed to predicted/estimated value of a keyword. For example, a number of searches made in association with a particular keyword at one or more search engines may be downloaded from the one or more search engines or may be calculated using historical data related to a number of searches. When calculating a number of searches for a particular search engine, a known number of searches for a second search engine may be multiplied by a ratio of the particular search engine's market share over the second search engine's market share. If, for example, Company A has a market share of 40% and Company B has a market share of 60%, an estimated number or searches for Company A will be achieved by multiplying a known number of searches for Company B by 40/60. Additionally, an estimated number of searches for a particular country may be calculated by multiplying an estimated or known number of searches in a second country numbers by the percentage of Internet users in the particular country with respect to Internet users in the second country.
The number of searches may be multiplied by a click through rate to determine a number of clicks associated with the keyword. The number of clicks may then be multiplied by cost-per-click data to arrive at a media ad cost associated with the keyword. A number of conversions may be determined by multiplying the number of clicks associated with the keyword by a conversion rate. A conversion may include various things, including a lead, a sale, a purchase, a content view, a content download, and a membership registration, among others. The conversion rate pertains to a percentage of visitors to a particular website who take a desired action. A cost-per-conversion may then be determined by dividing the media ad cost by the number of conversions. A cost-per-conversion describes the cost of acquiring a customer, typically calculated by dividing the total cost of an ad campaign by the number of conversions. One of skill in the art will appreciate that any of the variables (e.g., a number of searches, a conversion rate, etc.) used in the above analysis may be actual numbers or estimated numbers. One of skill in the art will also appreciate that averages of historical data, or desired portions of the historical data, may be used as one or more of the variables or may be used to calculate one or more of the variables in the above analysis.
One of skill in the art will also appreciate alternative embodiments to those described above that achieve a predicted value of a referral mode (e.g., a keyword).
Client Architecture
Attention is now drawn to
The implementation depicted in
The storage device 639h is described herein in several implementations as a hard disk drive for convenience, but this is certainly not required, and one of ordinary skill in the art will recognize that other storage media may be utilized without departing from the scope of the invention. In addition, one of ordinary skill in the art will recognize that the storage device 639h, which is depicted for convenience as a single storage device, may be realized by multiple (e.g., distributed) storage devices.
As shown, a value modeling software application 641 includes a performance indicator weighing module 641a, a tracking code module 641b, a data set collection module 641c, a normalization module 641d, and a value model generation module 641e, which are implemented in software and are executed from the memory 639g by the processor 639a. The software 641 can be configured to operate on personal computers (e.g., handheld, notebook or desktop), servers or any device capable of processing instructions embodied in executable code. Moreover, one of ordinary skill in the art will recognize that alternative embodiments, which implement one or more components in hardware, are well within the scope of the invention.
Each module 641a-e is associated with one or more of the steps described above with respect to
Those skilled in the art can readily recognize that numerous variations and substitutions may be made in the invention, its use and its configuration to achieve substantially the same results as achieved by the embodiments described herein. Accordingly, there is no intention to limit the invention to the disclosed exemplary forms. Many variations, modifications and alternative constructions fall within the scope and spirit of the disclosed invention as expressed in the claims. For example, the exemplary systems and methods of the invention have been described above with respect to the value modeling system 140. One of skill in the art will appreciate alternative embodiments wherein the functions of the value modeling system 140 are performed on other devices in the networked system 100.
The present application claims priority under 35 U.S.C. 119(e) to U.S. provisional application No. 60/823,615 entitled “System and Method for Modeling Value of an On-Line Advertisement Campaign,” filed on Aug. 25, 2006. This application relates to and incorporates by reference Provisional Application No. 60/778,594, entitled “System and Method for Managing Network-Based Advertising Conducted by Channel Partners of an Enterprise,” filed on Mar. 1, 2006, Provisional Application No. 60/823,615, entitled, “System and Method for Aggregating Online Advertising Data and Providing Advertiser Services,” filed on Aug. 25, 2006, Provisional Application No. 60/868,705, entitled “System and Method for Measuring the Effectiveness of an Online Advertisement Campaign,” filed on Dec. 5, 2006, Provisional Application No. 60/868,702, entitled “Centralized Web-Based Software Solution for Search Engine Optimization,” filed on Dec. 5, 2006.
Number | Date | Country | |
---|---|---|---|
60823615 | Aug 2006 | US |