KR101926350B1 - System and method for bidding automatically based on score of keyword group in keyword advertisement, and bidding management server - Google Patents

Abstract

The present invention relates to a method of automatic bidding in a keyword advertisement based on a keyword score, the method comprising the steps of: determining whether a set automatic bidding cycle arrives; Calculating a score for each keyword according to a preset method for at least one keyword when the automatic bidding cycle arrives; Calculating, in the agent server, a group score for each keyword cluster by combining the plurality of keywords; Selecting, in the agent server, a keyword combination that minimizes the calculated group score for each keyword cluster; Determining, at the agent server, a bid for each keyword in the selected keyword combination based on information received from the media server and the user terminal; And performing automatic bidding on the calculated real-time bid price.

Description

TECHNICAL FIELD The present invention relates to an automatic bid system, a method, and a bid management server for a keyword advertisement based on a keyword score group by keyword cluster,

More particularly, the present invention relates to an automatic bidding system for a keyword advertisement that automatically adjusts a bid in an Internet keyword advertisement of a competitive bidding method in which an advertisement is displayed or ranked according to a bid, And a bid management server.

In the modern information society, computers and the Internet have emerged as essential tools for new communication as a medium for transmitting various information. In addition, the Internet does not mean a computer and a computer are connected to each other, but rather serves as a new medium for realizing mutual communication between a company, an individual, and an individual. Accordingly, interest in various advertisements on the Internet is increasing day by day.

A keyword advertisement capable of keyword-based targeting among Internet advertisements is a method of allowing a customer (called an 'advertiser') sites to be listed at a predetermined location as an advertisement area in each portal site when a specific keyword is searched. For example, the keyword advertisement is highly effective because it can be exposed at a time point when a keyword is searched, that is, when a prospective customer who is likely to use the service of the advertiser wants information, .

1 is a diagram showing an advertisement method by keyword search in a general search site (e.g., a neighbor). Referring to FIG. 1, a search window 110 may be provided in a web page 100 of a general portal site. When a user inputs a search keyword (for example, an air conditioner) through a search window 110 included in the web page 100 and then performs a search, an advertisement for a specific company 120) will be given priority.

At this time, there are P4P (Pay for Performance) modeling method, CPM (Cost Per Millennium) modeling method, and CPA (Cost Per Action) modeling method according to a method of calculating the ad unit price for advertisement. The P4P model technique is a scheme in which an advertiser pays advertisement costs according to the number of times a PR for advertisement is clicked. The CPM model scheme is a scheme that spends advertisement expenses according to the number of times the PR is exposed. Meanwhile, the CPA model technique is an advertisement technique that is determined according to a result of a user's action, and an advertiser pays a fee in units of transactions such as purchase or membership of an actual user without involvement in impression, that is, advertisement exposure.

Meanwhile, in the P4P method, the amount determined according to the bid is deducted for each keyword every time a click is made, and the 'determined amount' is generally referred to as a 'cost per click' ). In the P4P method, the method of determining the unit price per click may be an important issue, and many companies providing the P4P advertisement solve this problem with a competitive bidding method. That is, if the minimum unit price (for example, 70 won) is set and each advertiser bids a larger amount of money in a specific amount unit (for example, 10 won), the advertisement is exposed to the upper part. Therefore, if there is no advertiser advertising with the desired keyword, it can occupy the first place even if it bids only with the minimum unit price (ie, 70 won). However, if another advertiser is already advertising for the keyword, it must be bid more than the minimum unit price (i.e., 10 won) than the advertiser so that the user can occupy the spot.

Accordingly, the competitive bidding method P4P advertisement method is performed as follows. First, an account (i.e., an ID and a password) for the advertiser is created, the amount of the advertisement is deposited, the information about the advertisement is registered, and the advertisement is reviewed by a media company (for example, Approval is determined. At this time, the information about the advertisement may include a keyword to be exposed, a page to be linked after the click, and a bid amount for each keyword. The approved advertisement may be simultaneously exposed to affiliate media linked to the media company.

On the other hand, only the exposure of the advertisement according to the P4P scheme does not reduce the cost. That is, when the customer clicks the corresponding advertisement, the amount corresponding to the set price per click is deducted from the amount paid for the advertisement. The advertiser can check the purchase conversion rate by the number of clicks and the number of purchases, and analyze the advertisement conversion rate by analyzing the conversion rate. Accordingly, if there is no advertising effect on the keyword, the user may adjust the bid or suspend the advertisement.

As shown in FIG. 1, in a keyword advertisement provided by a portal site or the like, an advertisement position or an advertisement may be determined according to a bid amount of another advertiser for the same search term. For example, the top five to fifteen ads may be displayed in order according to the bid amount at a predetermined position of a specific search engine (e.g., the first screen of the integrated search result page).

At this time, the fluctuation of the bid amount of other advertisers may also occur from time to time, so the ranking according to the bid amount inputted by the corresponding advertiser continuously changes. Accordingly, the advertiser must continuously observe the tendency of the bid price of other advertisers.

As described above, since the competitive bidding method is used in order to determine the advertisement ranking and the unit price per click in the P4P scheme, separate management is required for each advertiser. In order to maximize the advertising effect in the P4P system, since more similar keywords including representative keywords are registered and used, the bids must be continuously monitored and managed for each keyword.

On the other hand, when the keyword information is not provided to the advertiser in real time on the portal site or the like providing the keyword advertisement, ranking information about the keyword or minimum bid information for the target ranking is difficult to be effectively determined.

Therefore, there is a need for an effective method for determining the bids in various ways in accordance with the advertiser's advertising strategy, taking into consideration the advertising effectiveness against the cost to be invested in the keyword advertisement.

Korean Patent Registration No. 10-0754525 (Title of the Invention: Method of Setting Bid Rank in Internet Keyword Advertising of Automatic Bidding Method, Registered Date: Aug. 27, 2007)

The object of the present invention is to provide an automatic bidding system, a method and a bid for a keyword advertisement based on a group score per keyword group, which determines a real time bid for each keyword based on bidding history data collected for each account in a keyword advertisement over the Internet And a management server.

It is also an object of the present invention to provide an automatic bidding system, method, and program for a keyword advertisement based on a group score per keyword group, which determines a real time bid for each keyword according to a predetermined rule for each keyword or group in a keyword advertisement over the Internet And a bid management server.

According to another aspect of the present invention, there is provided a method of automatically generating a keyword keyword, the method comprising: calculating a group score for each keyword group based on bid-related information according to a predetermined method for each keyword group in a keyword advertisement over the Internet; And a bid management server in a keyword advertisement based on a group score of each keyword cluster.

In order to achieve the above-described object of the present invention and to achieve the specific effects of the present invention described below, the characteristic structure of the present invention is as follows.

According to an aspect of the present invention, there is provided an automatic bidding method for a keyword advertisement based on a group score per keyword cluster, the method comprising: calculating a score for each keyword based on bidding related information for at least one keyword; Calculating, in the agent server, a group score for each keyword cluster by combining the plurality of keywords; Selecting, in the agent server, a keyword combination that minimizes the calculated group score for each keyword cluster; Determining, at the agent server, a bid for each keyword in the selected keyword combination based on information received from the media server and the user terminal; And performing automatic bidding on the calculated real-time bid price.

Preferably, the method further comprises: determining whether a set automatic bidding cycle has arrived; And calculating a score for each keyword when the automatic bidding cycle arrives.

Preferably, the bid related information is real-time performance data on the same day collected from the media server.

Preferably, the keyword score may be calculated based on a click per action (CPA) and a number of conversions for each keyword.

Preferably, the keyword score may be calculated as a ratio of the CPA to the number of conversions.

Preferably, the keyword score can adjust the weight of the number of conversions by adjusting the exponent value of the number of conversions.

The calculating the group score for each keyword cluster by combining the plurality of keywords may include calculating a group score for each keyword by a keyword for each combination of the plurality of selected keywords; Adding the keywords that are not included in the combination of the selected keywords to the combination of the selected keywords in sequence to ship the group score for each cluster; And adding the re-computed keyword group score to the selected keyword combination if the calculated group score does not exceed a preset score threshold value.

According to another aspect of the present invention, in a bid management server in a keyword advertisement, an information collection module for collecting bid related information from a media server or an advertiser server that provides the keyword advertisement; Calculating a score for each keyword based on information related to keyword bidding for a keyword advertisement collected through the information collection module for at least one keyword, calculating a group score for each keyword group by combining a plurality of the keywords, A bid determination module that selects a keyword combination that minimizes the group score for each keyword cluster and determines a bid for each keyword in the selected keyword combination based on the information received from the media server and the user terminal; And an automatic bid processing module for performing automatic bidding with the calculated real time bids.

Preferably, the bid determination module may determine whether a set automatic bidding cycle has arrived, and, if the automatic bidding cycle arrives, calculate a score for each keyword.

Preferably, the bid related information is real-time performance data on the same day collected from the media server.

Preferably, the keyword score may be calculated based on a click per action (CPA) and a number of conversions for each keyword.

Preferably, the keyword score may be calculated as a ratio of the CPA to the number of conversions.

Preferably, the keyword score can adjust the weight of the number of conversions by adjusting the exponent value of the number of conversions.

Preferably, the bid determination module may calculate a group score for each combination of a plurality of selected keywords, a score for each keyword by a keyword, and sequentially add keywords not included in the combination of the selected keywords to the combination of the selected keywords And if the re-calculated group score for each cluster does not exceed a predetermined score threshold value, a keyword corresponding to a minimum value among the calculated group scores for each cluster is added to the selected keyword combination can do.

Meanwhile, the information for performing the automatic bidding method in the keyword advertisement based on the group score for each keyword cluster may be stored in a recording medium readable by the server computer. Such a recording medium includes all kinds of recording media in which programs and data are stored so that they can be read by a computer system. Examples include ROMs (Read Only Memory), Random Access Memory, CD (Compact Disk), DVD (Digital Video Disk) -ROM, magnetic tape, floppy disk, optical data storage device, (For example, transmission over the Internet). Such a recording medium may also be distributed over a networked computer system so that computer readable code in a distributed manner can be stored and executed.

As described above, according to the present invention, unlike the existing automatic bidding method in which bids are adjusted based on real-time ranking provided by a media server such as a portal site, You can maximize your advertising effectiveness and increase your profits by automatically adjusting your bids to predict your conversion performance and reach your target performance.

According to the present invention, when the number of keywords to be managed in one account is large in the keyword bidding method, the automatic bidding can be efficiently managed by selectively bidding the efficiency of each keyword.

1 is a view showing a web page displaying a keyword advertisement according to a result of a keyword search provided by a media server.
2 is a diagram showing a configuration of a system according to an embodiment of the present invention.
3 is a signal flow diagram illustrating a process procedure between respective configurations of a system according to an embodiment of the present invention.
4 is a flowchart illustrating a connection processing procedure in a user terminal according to an embodiment of the present invention.
5 is a flowchart illustrating a switching event processing procedure in a user terminal according to an embodiment of the present invention.
6 is a block diagram illustrating a detailed configuration of an advertisement information management server according to an embodiment of the present invention.
7 is a block diagram illustrating a detailed configuration of a bid management server according to an embodiment of the present invention.
8 is a diagram illustrating a concept of grouping keywords in an account according to an embodiment of the present invention.
9 is a flowchart illustrating a procedure for processing information received by the advertisement information management server according to an embodiment of the present invention.
10 is a flowchart showing a processing procedure in the bid management server according to the embodiment of the present invention.
11 is a diagram illustrating information processed in databases according to an embodiment of the present invention.
12 is a diagram illustrating information processing procedures in a system according to an embodiment of the present invention.
13 is a flowchart illustrating an account setting procedure in a bid management server according to an embodiment of the present invention.
14A, 14B and 14C are flowcharts illustrating various setting procedures in the bid management server according to the embodiment of the present invention.
15 is a graph illustrating a correlation between click and conversion for conversion performance prediction according to an embodiment of the present invention.
FIG. 16 is a diagram illustrating an optimal daily advertisement cost allocation concept for achieving target performance according to an embodiment of the present invention.
17 is a flowchart illustrating an optimal daily advertisement cost allocation procedure for achieving target performance according to an embodiment of the present invention.
18 is a flowchart illustrating a procedure for calculating a monthly goal according to an embodiment of the present invention.
19 is a flowchart showing a bidding procedure by a machine bidding mode according to an embodiment of the present invention.
20 is a flowchart illustrating a detailed bidding procedure by a machine bidding mode according to an embodiment of the present invention.
21A is a flowchart illustrating an automatic bid processing procedure based on keyword scores according to an embodiment of the present invention.
FIG. 21B is a flowchart illustrating an automatic bid processing procedure based on a group score per keyword cluster according to an exemplary embodiment of the present invention.
22A to 22D are graphs showing the relationship between the conversion rate and the conversion unit price in consideration of the multiplier of the conversion according to the embodiment of the present invention.
23 is a table showing scores calculated for each keyword according to various embodiments of the present invention.
24 is a diagram illustrating a screen for providing a performance prediction function according to an embodiment of the present invention.
FIG. 25 is a diagram illustrating a screen for providing a target achievement rate per group according to an embodiment of the present invention. FIG.
FIG. 26 is a diagram illustrating a screen for providing target achievement rates by time of each group according to an embodiment of the present invention. FIG.
27A and 27B are views illustrating a screen for providing an account setting function according to an embodiment of the present invention.
28 is a diagram illustrating a screen for providing a group setting function according to an embodiment of the present invention.
29 is a diagram illustrating a screen for providing a function of setting a machine auction mode according to an embodiment of the present invention.
30 is a diagram illustrating a screen for providing a rule-based bid mode setting function according to an embodiment of the present invention.
31 is a diagram illustrating a screen for providing a rule-based bid mode setting function according to an embodiment of the present invention.

The following detailed description of the invention refers to the accompanying drawings, which illustrate, by way of illustration, specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention. It should be understood that the various embodiments of the present invention are different, but need not be mutually exclusive. For example, certain features, structures, and characteristics described herein may be implemented in other embodiments without departing from the spirit and scope of the invention in connection with one embodiment. It is also to be understood that the position or arrangement of the individual components within each disclosed embodiment may be varied without departing from the spirit and scope of the invention. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is to be limited only by the appended claims, along with the full scope of equivalents to which the claims are entitled, if properly explained. In the drawings, like reference numerals refer to the same or similar functions throughout the several views.

Various embodiments of the present invention provide a method for automatically bidding on keyword advertisements based on bid-related history data for determining real-time bids for each keyword based on bid-related history data collected for each account corresponding to an advertiser in a keyword advertisement over the Internet System, method, and bid management server.

In addition, according to various embodiments of the present invention, the determination of the real-time bid may be performed by analyzing the correlation between the bid and the conversion-related goal based on the bid-related history data (e.g., history data) Can be determined.

In addition, in various embodiments of the present invention, the bidder algorithm can be applied to each account based on the bid-related history data collected for each account to determine the real-time bid for each keyword.

Also, in various embodiments of the present invention, at least one keyword registered for each account may be grouped into at least one group, and a bid for each keyword may be determined according to a preset automatic bidding method for each group.

In addition, in various embodiments of the present invention, a keyword score is calculated based on bid-related information according to a method preset for each keyword for a plurality of keywords in each account, and a keyword to be automatically bidded is selected .

For convenience of explanation, various embodiments of the present invention will be described using specific terms (e.g., exposure, click, conversion, CPC, CPA, bid, etc.) related to a keyword advertisement or keyword type bidding. However, it should be understood that the present invention is not limited to the above-described specific terms, and other terms used in the keyword advertisement field having the same function as the above-mentioned terms are also included in the present invention.

Before describing the various embodiments of the present invention, the bidding method of keyword advertisement will be briefly described below.

In the case of keyword advertising on the Internet, advertisers should set appropriate bids for each keyword in consideration of the advertising effectiveness against the advertising costs. That is, the advertiser sets the most optimal bid by referring to various data (e.g., the number of clicks, the click rate, the average cost, the total advertising cost, the total number of purchase conversions, the purchase conversion rate, .

The order in which the advertisements are exposed on the search list as a keyword search result is referred to as a 'rank', and the rank may be determined by a bid price between search listings exposed in the search results for the keyword.

At this time, the bid price set by each advertiser becomes the maximum bid price, and the actual price paid by the advertiser is equal to the maximum bid price of the rear ranked bidder, which is one level lower than itself, plus the unit bid (for example, 10 won). For example, as shown in Table 1 below, the first bid is 1,500 won, but the actual billing amount may be charged to 910 won, which is +10 of the second order bid.

ranking Bid (won) 1st 1,500 2nd place 900 3rd place 750

The ranking can be determined according to the bid index and the quality index between the search listings exposed in the search result for the keyword as shown in Table 2 below.

Item A Advertiser B Advertiser C Advertiser Bid 500 450 600 Quality index 4 5 3 Final bid decision maker 2,000 2,250 1,800

For example, the exposure order of the search advertisement area may be determined in the order of the amount by which the bid and the quality rating index are multiplied. The quality index means a quality evaluation index of a search listing by keyword. Therefore, if the quality index of the keyword is high, it can be exposed to a high rank with a low bid. However, the embodiments of the present invention are not limited to the above-described ranking method.

Meanwhile, when the media server (for example, the search site server) providing the keyword advertisement provides the real-time minimum bid information for a desired rank, the user can bid in a desired ranking with reference to the information. However, if the above-described information is not provided in real time, it is difficult for the user to determine a bid for an advertisement to be displayed at a desired location, and it may be difficult to efficiently execute an advertisement for each account in consideration of a desired target.

In the various embodiments of the present invention described below, even if the minimum bid and ranking information for the target ranking of each keyword is not provided in real time in the media server, it is possible to maximize the number of conversions, You can provide optimal bids for each keyword to help you reach your goals.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, so that those skilled in the art can easily carry out the present invention.

Referring to FIG. 2, the structure of a system according to an embodiment of the present invention will be described. Next, a procedure according to an embodiment of the present invention will be described in detail with reference to FIG.

2 is a diagram showing a configuration of a system according to an embodiment of the present invention. 2, a system according to an embodiment of the present invention includes an agent server 200, at least one media server 210, a communication network 220, at least one advertiser server 230, The terminal 240 may include at least one of the terminals.

The agent server 200 may be a server that performs an automatic bidding to the media server 210 on behalf of the advertiser according to various embodiments of the present invention. The agent server 200 may include a bid management server 201 and an advertisement information management server 202. Also, the agent server 200 may include a plurality of databases. For example, the agent server 200 may include at least one of a bid management information database 203, a collection information database 204, a medium information database 205, and an account information database 206.

2, the agent server 200 is divided into a bid management server 201 and an advertisement information management server 202. However, the agent server 200 may be implemented as one physical server or as a separate physical server It is possible. The bid management server 201 or the advertisement information management server 202 may be implemented as one physical server or a plurality of physical servers interlocked with each other. Also, according to various embodiments of the present invention, the agent server 200 may be managed by one provider, and a plurality of providers may manage the bid management server 201 and the advertisement information management server 202, respectively have.

The agent server 200 receives or collects information from the media server 210 or the advertiser server 230 according to various embodiments of the present invention and transmits the collected information to the advertiser server 230, The media server 210 may perform an automatic bid for at least one keyword. The detailed operation of the agent server 200 will be described later.

The media server 210 may provide a web page including a search window for keyword search to at least one user terminal 240 connected through the communication network 220 as illustrated in FIG. . At this time, when a user inputs a search word through the search window using the user terminal 240 and makes a search request, at least one advertisement registered as a keyword on the search result screen is set on a web page Can be displayed in the keyword advertisement area.

When one of the displayed keyword advertisements is selected through the user terminal 240, a web page provided by the advertiser server 230 corresponding to the selected advertisement may be provided. That is, if one of the advertisements is selected through the user terminal 240, the user can access the advertiser server 230 linked to the advertisement through the communication network 220 (that is, hyperlinked). The corresponding advertiser server 230 may provide a predetermined web page (e.g., a web page of a corresponding company) according to the connection of the user terminal 240. The advertiser server 230 may provide information related to the connection through the user terminal 230 of the user to the agent server 200 through the communication network 220.

In addition, when a user performs a predetermined operation (e.g., an operation related to switching (e.g., membership registration, shopping cart purchase, payment, etc.)) on the web page of the advertiser displayed through the user terminal 240, (230) may provide information related to the operation of the user to the agent server (200) through the communication network (220). The information provided by each advertiser server 220 may be stored in the collection information database 204 of the agent server 200.

For example, the advertiser server 230 may execute a predetermined tracking script to track connection or operation of the user terminal 240, and provide the result to the agent server 200. The detailed embodiments thereof will be described later in the description of Figs. 4 and 5.

Meanwhile, the media server 210 sets information (e.g., ranking information, exposure information, click information, approval information, etc.) related to bids for at least one keyword set in the account corresponding to each advertiser server 230 To the agent server 200 in a periodical manner. The information transmitted from each media server 210 may be stored in the media information database 205 of the agent server 200.

The advertisement information management server 202 of the agent server 200 transmits the information received from the advertiser server 230 or the information received from the media server 210 to the collection information database 204 or the medium information database 205 Or may be provided to the bid management server 201. [ According to various embodiments of the present invention, the information received from the advertiser server 230 or the information received from the media server 210 directly from the bidding management server 201 may be stored in the collection information database 204 or May be stored in the medium information database 205 or may be provided to the advertisement information management server 202.

The advertisement information management server 202 may store the tracking data received by the tracking script operation in each advertiser server 230 in the collection information database 204. [ The advertisement information management server 202 may further include processing the received tracking data and storing the processed tracking data in the collection information database 204 or the bid management server 201 ). The detailed functions of the advertisement information management server 202 will be described later.

The bid management server 201 may set information related to an account for each advertiser and store the information in the account information database 206. Also, according to various embodiments of the present invention, at least one group may be set by grouping at least one keyword set for each account. According to various embodiments of the present invention, the same automatic bid algorithm may be applied to at least one keyword included in the set specific group. The detailed embodiments will be described later.

The bid management server 201 may calculate the expected performance based on the history data accumulated in the collection information database 204 or the medium information database 205 for each account according to various embodiments of the present invention. The bid management server 201 calculates an optimal daily or weekly differential budget that can achieve a conversion-related goal (for example, the number of conversions, CPA, etc.) according to the current month's budget based on the history data, You can set conversion goals based on your budget.

In addition, the bid management server 201 may store history information stored in at least one of the collection information database 204 and the medium information database 205 for each group set in each account according to various embodiments of the present invention The bid amount for each keyword can be calculated according to a preset automatic bid algorithm. For example, the bidding management server 201 can automatically bid on the media server 210 based on bids for each keyword calculated according to the automatic bid algorithm when a predetermined automatic bidding cycle arrives. Various kinds of information related to the automatic bidding performed in the bid management server 201 may be stored in the bid management information database 203.

The user terminal 240 may include the various types of electronic devices. For example, user terminal 240 in accordance with various embodiments of the present invention may be a smartphone, a tablet personal computer, a mobile phone, a videophone, an e-book reader, Such as a desktop personal computer, a laptop personal computer, a netbook computer, a workstation, a server, a personal digital assistant (PDA), a portable multimedia player (PMP) Devices, cameras or wearable devices such as smart glasses, head-mounted-devices (HMDs), electronic apparel, electronic bracelets, electronic necklaces, electronic apps, , An electronic tattoo, a smart mirror, or a smart watch).

In addition, the communication network 220 may include at least one wired communication network or wireless communication network. For example, the wireless communication network cellular communication protocol may use at least one of LTE, LTE-A, CDMA, WCDMA, UMTS, WiBro, or GSM. Also, for example, the wired communication network may include at least one of a universal serial bus (USB), a high definition multimedia interface (HDMI), a recommended standard 232 (RS-232), a plain old telephone service (POTS) . The communication network 220 may also include at least one of, for example, a computer network (e.g., a LAN or WAN), the Internet, or a telephone network.

3 is a signal flow diagram illustrating a process procedure between respective configurations of a system according to an embodiment of the present invention. Referring to FIG. 3, the bid management server 201 in the agent server 200 can set an account corresponding to each advertiser (301). For example, the account setting may include an account type, a power source, or an update method (for example, automatic or manual) of data of the previous year, a method of setting a group keyword, a period of data to be reflected in average operational performance, Goals, conversion items, daily goals, and so on.

Next, at least one keyword to be included in the account can be set (302). For example, a plurality of keywords may be registered in one advertiser account. Also, according to various embodiments of the present invention, the registered at least one keyword may be grouped into at least one group according to a preset method. The same automatic bidding method (or automatic bid algorithm) may be set 303 for each of the groups according to various embodiments of the present invention. A detailed description thereof will be described later.

A user connects (304) to the media server 210 (e.g., portal site) through the user terminal 240 and the web page provided by the media server 210 is displayed on the screen of the user terminal 240 . When a keyword is input to a search window included in the web page through the input means of the user terminal 240 and a search 306 of the input keyword is input to the media server 210, (307) an exposure ranking for a predetermined number of keyword advertisements in consideration of a bid corresponding to the keyword.

The media server 210 may transmit 308 the search result web page including the determined keyword advertisements to the user terminal 240 as the search result. The user terminal 240 may display the search result web page 309 on the screen.

When the user selects (310) a specific keyword advertisement on the search result web page displayed on the user terminal 240, the user can access (311) the corresponding advertiser server 230 linked to the selected keyword advertisement. The advertiser web page provided by the advertiser server 230 may be displayed on the screen of the user terminal 240 according to the connection.

For example, a tracking script may be included in a web page provided to the user terminal 240 by the advertiser server 230. When browsing the web page in the browser of the user terminal 240, can do.

When the tracking script is executed in the user terminal, the connection of the user terminal 240 is detected and it is determined whether the connection of the user terminal 240 is a normal connection (313). For example, the determination of the normal connection may include determining whether the user terminal 240 is connected through the media server 210 previously set. The access path of the user terminal 240 may include a referer included in an HTTP (Hypertext Transfer Protocol) header transmitted for requesting a web page when the user terminal 240 accesses the advertiser server 230 ) URL (uniform resource locator). A detailed description thereof will be described later.

If it is determined that the connection of the user terminal 240 is a normal connection through the media server 210 as a result of the execution of the tracking script, the agent server 200 (for example, Server 202 or bid management server 201). The advertisement information management server 202 may process the inflow tracking data related information received from the user terminal 240 or the advertiser server 230 and store the inflow tracking data related information in the database 315.

The user terminal 240 may perform various operations through the advertiser web page displayed on the screen. (E.g., an operation related to subscription, shopping cart purchase, payment, etc.) 316, which is an operation performed through the user terminal 240, according to various embodiments of the present invention, The advertiser server 230 may process the conversion-related data (317). At this time, the agent server 200 (for example, the advertisement information management server 202 or the bid management server 201) generates conversion tracking data as the conversion related information according to a tracking script operated in the user terminal 240, (318). The advertisement information management server 202 may process the conversion tracking data related information received from the user terminal 240 or the advertiser server 230 and store the conversion tracking data related information in the database 319.

The admission tracking data or conversion tracking data received by the advertisement information management server 202 may be analyzed by the advertisement information management server 202 and the analysis result information may be transmitted to the bid management server 320.

In accordance with various embodiments of the present invention, media server 210 may also provide media information (e.g., keyword bidding) to agent server 200 (e.g., ad information management server 202 or bid management server 201) (321). ≪ / RTI > The transmitted medium information may be stored in the medium information database 205. [

In accordance with various embodiments of the present invention, the bidding management server 201 may control the bidding server 201 based on information received from at least one of the media server 210 and the advertiser server 230 according to a preset automatic bidding method for each group The bids for each keyword in the group may be determined 322.

For example, if the predetermined bidding condition is satisfied, or a preset bidding cycle arrives, the media server 210 may perform automatic bidding for the keyword according to the determined bid amount (step 323).

The media server 210 can determine the exposure or the exposure ranking according to the keyword search in consideration of the bid price in accordance with the automatic bid from the bid management server 201.

According to various embodiments of the present invention, an advertiser server may set a tracking script in advance to collect information related to a keyword advertisement. Accordingly, when a user terminal accesses the advertiser server through a specific media server or performs a predetermined operation (e.g., an operation related to switching) through a web page provided by the advertiser server, Alternatively, the advertiser server 230 may collect information related to the connection or switching by the operation of the tracking script and provide the collected information to the agent server.

4 is a flowchart illustrating a connection processing procedure in a user terminal according to an embodiment of the present invention. Referring to FIG. 4, when a user terminal accesses an advertiser server (or an advertiser website) (401), a predetermined tracking script may be operated (402).

At least one piece of data related to the connection of the user terminal may be collected 403 according to the operation of the tracking script. For example, the collected data may include at least one of a visit time, a session ID, a tracking code, a referrer URL, a landing page URL, a script version, and a user OS.

The user terminal may generate 404 the collected tracking code and session ID as cookie information.

According to various embodiments of the present invention, the user terminal can determine whether the connection by the user terminal is a normal inflow (405) by referring to the collected data. For example, it is possible to check the referrer URL among the collected data, and judge that the referral URL is a normal inflow when the referrer URL corresponds to a predetermined URL of the media server.

As a result of the determination, if the normal inflow is confirmed, the user terminal can transmit (406) inflow tracking data related to the connection to the agent server.

5 is a flowchart illustrating a switching event processing procedure in a user terminal according to an embodiment of the present invention. Referring to FIG. 5, a user terminal may access an advertiser server (or an advertiser web site) and perform various operations as described above with reference to FIG. When an event related to a predetermined transition among operations performed by the user terminal occurs (501), a predetermined tracking script may be operated (502). The conversion means a meaningful operation of a user who visited a web site of an advertiser to be tracked, and may include an operation such as membership registration, shopping cart purchase, purchase, payment, and the like.

For example, when an event related to the switching occurs, the session ID and tracking code may be extracted 503 from the cookie information previously generated in FIG. Also, at least one piece of data related to the switching of the user terminal may be collected 504 through the extracted session ID and tracking code according to the operation of the tracking script. For example, the collected data may include at least one of a conversion occurrence time, a session ID, a tracking code, a page URL where a conversion is made, a script version, and a user OS.

According to various embodiments of the present invention, the user terminal may generate conversion tracking data based on the collected data and transmit (505) the conversion tracking data associated with the conversion to the agent server.

At least one of the operations shown in FIGS. 4 and 5 may be omitted and at least one other operation may be added between the operations. 4 and 5 may be processed in the order shown, and the order of execution of at least one operation may be changed and processed in accordance with the order of execution of other operations. 4 and 5 may be performed in one electronic device or server, and in accordance with another embodiment, at least one of the operations shown in FIGS. 4 and 5 may be performed in the first And the remaining operations may be implemented to be performed in the second device.

Hereinafter, the detailed configuration of the advertisement information management server and the bid management server will be described with reference to FIG. 6 and FIG.

6 is a block diagram illustrating a detailed configuration of an advertisement information management server according to an embodiment of the present invention. 6, the advertisement information management server 201 according to various embodiments of the present invention includes at least one of a log information collection module 610, a log information analysis module 620, and an analysis information transmission module 630 . According to various embodiments of the present invention, the log information analysis module 620 may include at least one of a normal data determination unit 621, a redundant data removal unit 622, and a tracking log update unit 623 .

The log information collection module 610 may collect the tracking log information collected by the operation of the tracking script of the advertiser server as described above. The tracking log information to be collected may include at least one of a visited time, a user IP, a session ID, a tracking code, a keyword code, a referrer URL, a landing page URL, a script version user OS, and a browser version of the accessed user terminal.

The log information analysis module 620 may analyze the log information collected by the log information collection module 610 and process the log information. For example, the normal data determination unit 621 can confirm whether the visit of the advertiser site of the user IP is a site visit through the normal media server through the referrer URL among the collected information. As a result of the determination by the normal data determination unit 621, the tracking log information can be removed if the visit is based on an abnormal path.

The redundant data removing unit 622 can check and remove the collected tracking log if it is duplicated with the previously collected tracking log. For example, the redundant data removing unit 622 can check whether the collected tracking logs are duplicated data through the user IP and the session ID of the tracking log.

The tracking log updating unit 623 may update the log information such as the media code, the campaign, the group, and the keyword through the keyword code for the tracking log. In addition, the tracking log updating unit 623 can update media data and perform other necessary data processing operations.

The analysis information transmission module 630 may report the processed tracking log to the bid management server 021 or the like as a result of analysis through the log information analysis module 620 according to various embodiments of the present invention. For example, the processed tracking log may include information such as date, advertiser, ID, media code, campaign name, group name, keyword name, keyword code, advertisement expense, click, conversion classification, . The processed tracking log may include, for example, at least one performance data.

7 is a block diagram illustrating a detailed configuration of a bid management server according to an embodiment of the present invention. 7, the bid management server 201 according to various embodiments of the present invention includes an account setting module 710, a group setting module 720, a budget and goal setting module 730, an information collection module 740, A bid determination module 750, an automatic bid processing module 760, and a performance data management module 770.

The account setting module 710 may set an account corresponding to each advertiser. For example, the account setting module 710 may set the account type, the power or the previous year's data update method (e.g., automatic or manual), the group keyword setting method, the period of data to be reflected in the average operational performance, Set up budget plans, goal items, conversion items, and daily goals. In addition, the account setting module 710 may set a bid weight for each time period, and may set a weight value for each type of business.

The group setting module 720 registers at least one keyword to be included in the account and groups the registered at least one keyword into at least one group according to a predetermined method according to various embodiments of the present invention Can be performed. In addition, an automatic bidding method to be applied for calculating a bid for each of the set groups can be set. For example, the same automatic bidding method (or automatic bid algorithm) may be set for each of the groups according to various embodiments of the present invention.

The budget and goal setting module 730 may perform a function of setting budgets and goals for the entire account and may be configured to set budgets and goals for at least one group in the account according to various embodiments of the present invention . ≪ / RTI > In addition, the budget and goal setting module 730 may set a budget measurement method, derive a budget for each day of the week according to the set budget measurement method, and output the result in the form of monthly data.

The information collection module 740 may receive media data directly from the media server 210 as described above, or may directly receive performance data related to incoming or outgoing from the advertiser server. According to another embodiment, the information collection module 740 may collect the media data collected from the media server 210 or the performance data collected from the advertiser server in the advertisement information management server 202, As shown in FIG.

The bid determination module 750 determines whether or not the group setting module 720 has received the bid information based on the history information (e.g., media data, performance data, etc.) collected by the information collection module 740 for each group, The bidding amount for each keyword can be determined by applying an automatic bidding algorithm for each group set by the user.

The automatic bid processing module 760 may bid on the corresponding media information with the bid determined by the bid determination module 750 for each keyword.

The performance data management module 770 can perform a function of updating the performance of each group in real time. For example, the performance data management module 770 can calculate and store a budget amount, a conversion target achievement, and a market saturation according to a bid change.

In the meantime, the respective components of the apparatus are separately shown in the drawings to show that they can be functionally and logically separated, and do not necessarily mean physically separate components or separate codes.

In this specification, each functional unit (or module) may mean a functional and structural combination of hardware for carrying out the technical idea of the present invention and software for driving the hardware. For example, each functional unit may refer to a logical unit of a predetermined code and a hardware resource for executing the predetermined code, and may be a code physically connected to the functional unit, But can be easily deduced to the average expert in the field of the invention.

8 is a diagram illustrating a concept of grouping keywords in an account according to an embodiment of the present invention. Referring to FIG. 8, at least one keyword may be set for the account 810 corresponding to each advertiser. The at least one keyword set in the account 810 may be grouped into a plurality of groups 821, 822, 823, 831, 832, 833 according to various embodiments of the present invention. Also, according to various embodiments of the present invention, each of the groups may be a basic unit for setting an automatic bid algorithm, and may be a basic unit for budget allocation or determination of goal achievement.

Between the account 810 and the plurality of groups, campaigns 820 and 830 can be classified. Each of the campaigns 820 and 830 may be a larger unit of automatic bid algorithm configuration. For example, it can be classified into two campaigns 820 and 830 of one account 810 as shown in FIG. According to various embodiments of the present invention, a machine bidding mode may be set for the groups 821, 822, and 823 corresponding to the first campaign 820, and groups 831 corresponding to the second campaign 830 , 832, and 833, a rule-based bid mode may be set. A detailed description of the machine bidding mode or the rule-based bidding mode will be described later.

According to various embodiments of the present invention, among the plurality of automatic bid algorithm corresponding to the machine auction mode in each of the plurality of groups 821, 822, and 823 corresponding to the first campaign 820, You can make detailed settings. Similarly, in each of the plurality of groups 831, 832, and 833 corresponding to the second campaign 830, detailed settings can be made for each group among a plurality of automatic bid algorithms corresponding to the rule-based bidding mode have.

Meanwhile, according to the various embodiments of the present invention, since the setting of various automatic bidding modes is performed for each group as described above, it may be effective to group similar keywords into the same group.

Also, according to various embodiments of the present invention, a small number of high importance keywords may be placed in the groups 821, 822, 823 corresponding to the machine bidding mode set in the first campaign 820, A plurality of keywords having a relatively small number of conversions may be arranged in one group in the groups 831, 832, and 833 corresponding to the rule-based bid mode set in the rule-based bidding mode.

In addition, according to various embodiments of the present invention, as shown, by classifying at least one of the groups into a plurality of campaigns, the bid mode can be effectively managed according to the characteristics of each keyword. For example, the keywords corresponding to the first campaign 820 may be modeled through correlation analysis between the conversion and the bid according to various embodiments of the present invention, as the keywords having a relatively large number of conversions. On the other hand, Based on the rule-based bidding mode designated by the person in charge of the advertisement, by arranging the keywords in the second campaign 830 for the keywords whose correlation is difficult to analyze and the number of which is small.

8, various settings can be made for each account 810, campaigns 820 and 830, and groups 821, 822, 823, 831, 832, and 833 in FIG. For example, the business type can be selected for the account 810, and the monthly budget (for example, total budget or daily budget), goal (conversion or CPA), and the like can be set. In addition, a bid weight for each account time unit can be set for the account 810, and a weight for each type of business can be set. In addition, it is possible to perform data preprocessing on the account 810, and to perform weekly budgeting and finalization on week days. Also, it may turn off or on the keyword bidding for the entire account 810 according to the user's selection.

Also, as described above, a machine bidding mode may be set for the first campaign 820, and a rule-based bidding mode may be set for the second campaign 830. [

According to various embodiments of the present invention, a plurality of groups 821, 822, and 823 corresponding to the first campaign 820 may include a specific bid algorithm corresponding to the machine auction mode for each group A specific bid algorithm corresponding to the rule-based bid mode may be set for each of a plurality of groups 831, 832, and 833 corresponding to the second campaign 830, as described above have.

In the machine bidding mode, data collected in real time according to the target of each group is analyzed, and automatic bidding can be performed by a machine bidding algorithm corresponding to the predetermined machine bidding mode. In the rule-based bidding mode, the automatic bidding can be performed according to the bidding rule and cycle set in advance for each group.

In addition, for each group, a group-based budget may be set, and a goal related to the conversion (e.g., conversion, CPA, etc.) may be set. According to various embodiments of the present invention, time-based goals (e.g., conversion, CPA, etc.) can be calculated for each of the groups. Accordingly, it is possible to determine whether or not the target group is accomplished for each group, and determine whether or not to update the group's bid for the group.

The priority for the budget in the classification may be account, campaign, and group, and the priority for the conversion target may be set in the order of group, campaign, and account.

When the automatic bidding function is executed, it is checked whether the conversion target is achieved in the order of each account, campaign, and group according to the preset automatic bidding cycle. In accordance with the result of checking, each keyword Can be automatically calculated.

More specifically, in the machine bidding mode, the bids can be calculated by calculating unattained numerical values for the goals set for each group and analyzing the correlation between the CPC and the conversion. In addition, if the explanatory power of the result according to each condition is low, a notification may be output to the manager. In the rule-based bidding mode, performance-based bidding such as raising or lowering bids according to the previous day's ranking based target ranking can be applied, or performance based bidding can be applied such as raising or lowering bids based on performance for each set index have. The application of the above performance may be weighted and applied in the order of conversion, click, exposure, and the like. In addition, criteria such as maximum bid, bid cycle, and bid change amount may be applied to each group.

In addition, according to various embodiments of the present invention, when a bid is determined for keywords in the group for each group, automatic bidding is performed at a determined bid price, and the result is updated. For example, after the automatic bidding is performed at the calculated bids, the performance of each group may be updated in real time. For example, the amount of budget used, the achievement of the conversion target, and the degree of market saturation due to bid changes can be calculated for each group.

Hereinafter, procedures to be processed in the respective servers will be described with reference to FIGS. 9 and 10. FIG.

9 is a flowchart illustrating a procedure for processing information received by the advertisement information management server according to an embodiment of the present invention. Referring to FIG. 9, as described above, the advertisement information management server can receive (901) tracking data (e.g., incoming tracking data or conversion tracking data) from each advertiser server. The received tracking data may be stored 902 in a database.

In accordance with various embodiments of the present invention, abnormal data in the received tracking data may be removed 903, and redundant data may also be removed 904. The tracking log may then be updated 905 via the keyword code and the media data may be updated 906 as described above. The processed tracking log may be reported (907) to the bid management server at predetermined time intervals or at the occurrence of a predetermined event.

10 is a flowchart showing a processing procedure in the bid management server according to the embodiment of the present invention. Referring to FIG. 10, in the bid management server 1001, an account corresponding to each advertiser can be set (1001), at least one group is set by grouping at least one keyword set in each account (1002) . In addition, according to various embodiments of the present invention, the bidding method may be set 1004 for each of the set groups.

Thereafter, the information may be received directly from the media server or the advertiser server, or the information processed through the advertisement information management server may be received and stored in the database. (1005) various history information collected directly or indirectly from the media server or the advertiser server, and determines a bid for each keyword in the group according to the bidding method set for each group based on the collected information (1006).

When the bids for the respective keywords are determined as described above, the automatic bidding for the keywords may be performed 1007 by the media server at the determined bids.

11 is a diagram illustrating information processed in databases according to an embodiment of the present invention. Referring to FIG. 11, the agent server may include a bid management information database 203, a collection information database 204, a medium information database 205, and an account information database 206.

The account information database 206 may store account information set for each advertiser as described above. Data for each account-specific sub structure as shown in FIG. 8 (for example, grouping for campaigns or groups Data) can be stored.

The medium information database 205 may store media data provided from each medium that provides keyword advertisement according to a search word input through a web page. The media data stored in the medium information database 205 may store real-time media data for each time period received from the media server through the set API (application programming interface) and daily media data received on a daily basis.

The collection information database 204 may store performance data or conversion item information collected from each advertiser server. For example, the collection information database 204 may store real-time performance data for each time period received from each advertiser server in a time zone and daily performance data received in a day.

The media data stored in the medium information database 205 and the performance data stored in the collection information database 204 may be stored in the bid management information database 203 as backup data or performance backup data, respectively.

The bid management information database 203 may store media data, account setting data, bid log information, group setting data, daily performance data backup, budget / goal setting data, and the like.

12 is a diagram illustrating information processing procedures in a system according to an embodiment of the present invention. Referring to FIG. 12, the daily performance data stored in the collection information database and the real-time performance data per time period may be provided 1201 to the bid management information database.

For example, the daily performance data may be transmitted to the bid management information database of the bid management server when requested by the bid management server as daily data for a period of one month. The daily performance data may include at least one of a date, a keyword name, an echo key, a conversion item, a conversion, and a group sales by date. Also, for example, the real-time performance data for each time period may be transmitted to the bid management information database at predetermined intervals (for example, every 30 minutes) as time-based data for the previous 24 hours. The time-based real-time performance data may include at least one of time, keyword name, EchoKey, conversion item, conversion, and time-based group sales.

The daily media data stored in the medium information database may be transmitted to the bid management information database of the bid management server upon request from the bid management server. The daily media data may include at least one of daily date, month, month, month, month, day, month, month, day, month,

The data of the past 24 hours among the performance data and the media data stored in the bid management information database may be transmitted 1203 to the bid management server or the bid management module (for example, the bid management program) in the bid management server by keyword matching. For example, performance data or media data stored in the bid management information database may be read out and used for real-time automatic bidding when the bid management module determines that the bid cycle has arrived, or reads the performance data or the media data stored in the bid management information database every predetermined cycle. In addition, previously set values stored in the bid management information database may be transmitted to the bid management module.

In addition, information related to the account structure of keywords to be billed out among the account structure data stored in the account information database may be transmitted 1204 to the bid management module. In addition, the previous year's data YOY may be provided to the bid management module as data for budget, goal, and forecast.

The bid management module may receive the log data, the auction result log, and the like and store it in the bid management information database (1205).

In addition, according to various embodiments of the present invention, when the bidding management module requests (1206) a keyword status or a bid to a media server, the media server transmits the requested keyword status or bid information to the bid management module through a media API (1207).

In the medium report module, media data for the time period of the previous 24 hours period can be requested (1209) to the media server (1209). In the media server, the media report module can respond to the time report by the medium report module (1210) through API. In the medium report module, the media data for each time period received from the media server is provided to the bid management server at a predetermined time period (for example, at intervals of one hour), and the provided media data for each time period can be stored in the bid management information database . For example, the media data for each time period may include at least one of time, keyword name, keyword ID, conversion item, conversion, sales of each group by time, and keyword ID.

Hereinafter, operations of the bid management server according to various embodiments of the present invention will be described in detail with reference to FIG. 13 to FIG.

13 is a flowchart illustrating an account setting procedure in a bid management server according to an embodiment of the present invention. Referring to FIG. 13, an account setting screen may be displayed for an account corresponding to each advertiser, and various account settings may be provided through the account setting screen. For example, business type can be selected 1301 through the account setting screen, and daily keyword data 1302 can be uploaded. In addition, data to be reflected in the average operation may be selected 1303, a monthly budget may be written 1304, and a budget measurement method may be selected 1305.

When the budget measurement method is automatically set 1306, the weekly difference day budget may be derived 1307 based on the past data, and the resultant value may be output 1308 in the form of month data. When the determined data is determined as the output data 1309, the derived budget may be stored 1311 in the database.

On the other hand, if the budget measurement method is manually set 1306 or the output data is not determined 1309, the user can input the monthly budget 1310 every day.

When the budget is stored, a target item can be selected 1312 and a conversion item can be selected 1313. In accordance with the selection, the weekly difference day goal can be calculated 1314 based on the past data.

When the determined target is determined (1315), the derived target may be stored (1317) in the database. On the other hand, if the derived goal is not determined (1315), the user may input the monthly goal (1316) on a daily basis.

14A, 14B and 14C are flowcharts showing various setting procedures in the bid management server according to the embodiment of the present invention. Referring to FIG. 14A, after the account setting is completed as shown in FIG. 13, at least one group may be created by grouping at least one keyword set in the account according to an embodiment of the present invention.

In order to create the group, a group name is input 1401, and a keyword corresponding to the group is uploaded 1402. The method for grouping the keywords by each group may be performed by a user or may be automatically grouped into keywords having similarity according to a predetermined algorithm.

If a group including at least one keyword is set as described above, the bidding method may be selected 1403 for each group. Then, the weekly difference day budget of each keyword included in the corresponding group may be derived 1405, and the derived result may be output 1405 as a month data type.

If the user decides 1406 for the output data, the determined data may be stored in the database as a budget 1408. [ If the output data is not determined (1406), the user may input (1407) the monthly budget on a daily basis.

When the budget for each group is determined as described above, a bid increase / decrease interval, a maximum bid, a group on / off, and a bid on / off may be set 1409. In addition, a black keyword condition can be set 1409. In addition, according to whether the bidding method for the group is a machine bidding method or a rule-based bidding method, steps 1410, 14b, or 14c may be performed.

For example, referring to FIG. 14B, when a bidding method of the set group is a machine bidding method, a target item and a conversion item can be set (1411, 1412).

Subsequently, a weekly discrimination day-goal can be derived (1413) for each of the generated groups. When the user determines (1414) the derived data, the determined target may be stored in the database (1416). If the derived data is not determined (1414), the user can input the monthly goal (1415) on a daily basis.

In addition, according to various embodiments of the present invention, it is also possible to additionally set 1417, 1418, and 1419 a time-weighted weight, a bid cycle, a bid change interval, and the like.

Referring to FIG. 14C, when the bidding method of the set group is the rule-based bidding method, a rule name of an automatic bidding method to be applied to the group can be set 1420. The bidding cycle, The bidding method, and the maximum bid amount (1421). Finally, rules may be additionally set 1422 according to media / performance data conditions.

15 is a graph showing a relationship between click and conversion for conversion performance prediction according to an embodiment of the present invention. According to various embodiments of the present invention, the conversion performance can be predicted by analyzing the correlation between each item index based on the history data collected from the media server or the advertiser server. For example, by analyzing the correlation between each of the item indicators, the expected goal of the account unit can be derived.

Mathematical regression analysis algorithms may be applied to the correlation analysis according to various embodiments of the present invention. The regression analysis algorithm is one of statistical techniques for finding trends and trends of future data by referring to information of past data. Therefore, the regression analysis algorithm can be used to analyze the relationship between exposure and click, the relationship between click and conversion, and the relationship between CPC (or bid) and conversion.

For example, daily data of at least three months can be uploaded, and regression analysis can be performed with the uploaded data. According to the regression analysis, the correlation between clicks and conversions, clicks and advertising costs can be analyzed, and predicted conversions or conversion steps according to the monthly budget can be predicted according to the analyzed results.

FIG. 15 is a graph showing the relationship between click and conversion by the regression analysis algorithm as described above. Referring to FIG. 15, it is possible to perform a regression analysis on the basis of accumulated conversion data and click data for a predetermined period of time, and to establish a regression equation.

For example, the regression equation according to the above analysis can be determined as follows.

If a target click value is substituted for x in Equation (1), a prediction value for conversion can be derived as a y value. For example, when the number of clicks is 10,000, substituting into Equation (1) yields y = 0.0283 * 10000 + 118.55, and a conversion value of 402 can be derived. That is, according to various embodiments of the present invention, by applying regression analysis using the historical data, it can be predicted that 402 occurrences occur when the number of clicks is 10,000.

In this manner, the relevance of the keyword advertisement can be determined by applying the regression analysis algorithm to the keyword advertisement through the accumulated history data. In addition, since the relationship between variables can be output as a number, intuitive and quick understanding is possible.

FIG. 16 is a diagram illustrating an optimal daily advertisement cost allocation concept for achieving target performance according to an embodiment of the present invention. Referring to FIG. 16, according to various embodiments of the present invention, an optimal daily advertisement cost allocation for achieving the monthly conversion target can be performed based on the historical data.

For example, considering the daily schedule data of the previous month, which is the operation history data, and the daily performance data of the same month of the previous year, You can allocate advertising costs. In this way, the automatic bid algorithm can be operated according to the achievement of the conversion target based on the target value by automatically performing the quantitative conversion target plan necessary for the advertisement operation based on the data so as to systematically achieve the monthly goal.

In other words, based on the daily performance data of the previous month, which is the operation history data, and the daily performance data of the previous year, it is possible to allocate the optimal daily advertisement cost for the monthly goal by considering the advertisement plan of the total advertisement cost of the corresponding month, And the target conversion value can be calculated by substituting the conversion efficiency index CPA according to the advertisement cost allocated to each day.

For example, referring to FIG. 16, the budget for the fourth month on the fourth month of the month (MOM) is 5% (1610) of the monthly budget, while the budget for the fourth day on the year (YOY) (1620), the average value considering the above two values may be 6%. Therefore, the budget for the 4th day of the month can be set to the monthly budget * 6% (1630). The monthly target conversion * 6% may be set as the 4-day target conversion (1640) in consideration of the set monthly budget * 6%.

17 is a flowchart illustrating an optimal daily advertisement cost allocation procedure for achieving target performance according to an embodiment of the present invention. Referring to FIG. 17, according to various embodiments of the present invention, it is possible to automatically allocate an optimal daily advertisement cost for achieving monthly goal conversion based on the operational history data.

First, an average value of data by day of week is calculated for the account / group (1701). If the day of the first day of the budget setting is earlier than the delivery date / month (1702), the same number of days of weekly average value arrangement can be input (1704). On the other hand, if the day of the first day of the budget setting is not earlier than the delivery / the same month, the day-of-week data array of the difference can be deleted (1703).

Next, if the day of the last date of the budget setting is earlier than the delivery / month (1705), the day of the week data array can be deleted (1706). On the other hand, if the day of the last day of the budget setting is not earlier than the delivery / month, an array of the weekday average value can be inputted (1707) as long as it is not enough.

Then, the day of the week is matched for each parking (1708). If the data exists only in the previous month (1709), the daily data ratio may be calculated (1710) based on the total data sum value. On the other hand, if the data does not exist only in the previous month (1709), it is possible to derive 1711 the parking, day of the week, and daily budget based on the MOM / YOY data. The resultant value of the derived budget may be output 1712 in the form of monthly data.

18 is a flowchart illustrating a procedure for calculating a monthly goal according to an embodiment of the present invention. Referring to FIG. 18, the daily budget and target data of the last month (transfer) account / group is read (1801) from the database. And analyzes the correlation between the advertising cost and the target using the read data. At this time, the target prediction equation can be derived 1802 by analyzing the correlation between the advertisement cost and the target using the simple regression analysis according to one embodiment of the present invention. Based on the daily budget set in the previous step, the target value can be calculated 1803 using the regression analysis result. The resultant value may be output 1804 in the form of monthly data.

19 is a flowchart showing a bidding procedure by a machine bidding mode according to an embodiment of the present invention. Referring to FIG. 19, when a bid cycle comes (1901), a score for at least one keyword included in the group is calculated, and at least one keyword to be changed based on the calculated score is selected (1902) have. Specific examples of the score calculation method will be described later in the description of FIG. 21 to FIG. 23. FIG.

If the time-based strategy application is active for the group and the current time zone corresponds to the power application (1903), the competitive-condition-based bid price may be calculated 1904. The calculated bid price may be updated 1909, and the log and the setting value may be stored 1910. In addition, automatic bidding may be performed at the calculated bid price.

On the other hand, if the time-based strategy application is disabled (1903) for the corresponding group, the bid amount may be calculated 1905 based on the bid-related history data. Accordingly, a preset bid determination algorithm can be operated 1906.

If the bidding calculation exception condition is satisfied in accordance with the bid determination algorithm (1907), the difference between the calculated bid amount and the current bid amount may be summed (1908) at the current bid amount. On the other hand, if the bidding calculation exception condition is not satisfied according to the bid determination algorithm, the bidding may be updated 1909 with the determined bid price, and the log and the setting value may be stored 1910.

20 is a flowchart illustrating a detailed bidding procedure by a machine bidding mode according to an embodiment of the present invention. Referring to FIG. 20, when a bidding cycle arrives (2001), real-time data may be loaded (2002) from the database. A bid determination algorithm may be performed (2003) based on the loaded real time data.

According to the operation of the bid determination algorithm, an alarm is output (2011) to the user for budget / goal adjustment (2010), and the budget / goal is reset (2012).

Then, it may be determined 2020 to increase or decrease the bid according to the operation of the bid determination algorithm. More specifically, a keyword score may be calculated for a plurality of keywords included in the group, and a keyword to be changed may be selected (2021) based on the calculated keyword score. Specific examples of the score calculation method will be described later in the description of FIG. 21 to FIG. 23. FIG.

If it is set to compare the accomplishments by time zone 2022, a bid price may be calculated 2023 based on the competitive situation and then updated 2024 to the calculated bid price. On the other hand, if the time zone achievement is not set to be compared (2022), the bid price may be calculated based on the bid history data (2025) and the bid price may be updated (2024). The calculated bids may be updated and log and set values may be stored (2030).

Hereinafter, detailed embodiments for performing automatic bidding in consideration of keyword-specific scores according to an embodiment of the present invention will be described with reference to FIGS. 21 to 23. FIG.

21A is a flowchart illustrating an automatic bid processing procedure based on keyword scores according to an embodiment of the present invention. Referring to FIG. 21A, when an automatic bidding cycle arrives, a score can be calculated based on bid related information for each keyword.

When the conversion data is used as the bid related information for the score calculation, the conversion weight may be set 2101 according to various embodiments of the present invention. The conversion weight can be used as a weight for adjusting the weight of conversion, for example, the exponential power of conversion can be used as a weight in accordance with an embodiment described later. In addition, according to various embodiments of the present invention, the ratio of the upper and lower scores of the score or the upper and lower scores may be set 2102 for determining the number of keywords according to the score result.

When the real-time bidding cycle arrives and the user desires to select a keyword to be bidded in real time, the same day real-time performance data is confirmed (2103). Then, the score for each keyword may be calculated 2104 using the bid-related information (e.g., CPA or conversion) for each keyword. The keyword score calculation method can be set by various mathematical expressions. For example, a keyword score can be determined as a ratio of CPA to conversion as shown in Equation (2) below.

In Equation (2), the exponent value n of the conversion can be used as a conversion weight. For example, the larger the value of n, the greater the proportion of conversion. The score according to Equation (2) may mean conversion efficiency. The lower the CPA or the greater the number of conversions, the smaller the score, and the smaller the score, the more efficient the keyword is. Therefore, the weight of the conversion can be adjusted by adjusting the value of n in Equation (3). For example, as the value of n increases, the weight of conversion becomes higher, and the order of conversion for each keyword becomes similar.

If the score for each keyword is calculated as shown in Equation (2), the calculated keywords are sorted according to the ranking, and keywords in the predetermined upper or lower section are determined (2105). As a more specific example, it is possible to select a keyword in an upper section corresponding to a predetermined ratio or a predetermined number according to the score for each keyword, and to select a keyword in a lower section of a predetermined ratio or number.

As the automatic bidding cycle arrives, automatic bidding 2106 can be performed on the keywords of the selected upper or lower section among all the keywords.

The period of the keyword score calculation may be the same as or different from the automatic bidding period. The bid-related information on which the keyword score calculation is based may use the same day real-time performance data as the bid history information collected from the media server or the advertiser server.

FIG. 21B is a flowchart illustrating an automatic bid processing procedure based on a group score per keyword cluster according to an exemplary embodiment of the present invention. Referring to FIG. 21B, when an automatic bidding cycle arrives, the group score for each group can be calculated based on the bid-related information for each keyword.

When the conversion data is used as the bid-related information for calculating the group score for each cluster, the conversion weight may be set 2111 according to various embodiments of the present invention. The conversion weight can be used as a weight for adjusting the weight of conversion, for example, the exponential power of conversion can be used as a weight in accordance with an embodiment described later. Also, in accordance with various embodiments of the present invention, a keyword combination that minimizes the group score per keyword cluster for all possible keyword combinations may be selected.

When the real-time bidding cycle arrives and the user wishes to select a keyword to be bombarded in real time, the same day real-time performance data is confirmed (2112). Then, the score for each keyword may be calculated 2113 using the bid-related information (e.g., CPA or conversion) for each keyword. The keyword score calculation method can be set by various mathematical expressions. For example, the keyword score can be determined as a ratio of the CPA to the conversion, as shown in Equation (2).

In Equation (2), the exponent value n of the conversion can be used as a conversion weight. For example, the larger the value of n, the greater the proportion of conversion. The score according to Equation (2) may mean conversion efficiency. The lower the CPA or the greater the number of conversions, the smaller the score, and the smaller the score, the more efficient the keyword is. Therefore, by adjusting the value of n in Equation (2), it is possible to control the weight of conversion. For example, as the value of n increases, the weight of conversion becomes higher, and the order of conversion for each keyword becomes similar.

When a score for each keyword is calculated as in Equation (2), a group score for each keyword cluster is calculated (2114) by summing up the score for each keyword, and at least one group score for minimizing the group score for each keyword cluster (2115). As a more specific example, it is possible to calculate the group score of each keyword cluster for all possible keyword combinations, and to select a keyword combination that minimizes the group score for each calculated keyword cluster.

The group score calculation method for each cluster may be set by various mathematical expressions. The keywords are sorted according to the rank of the calculated score for each keyword, and the keyword of the upper section corresponding to the preset minimum number of group keywords is selected.

For each selected keyword combination, the score for each keyword is calculated as shown in Equation (3) below to calculate a group score for each reference cluster.

The keywords not included in the selected keyword combination are sequentially added to the selected keyword combination one by one, and the group score for each cluster is calculated again.

If the calculated group score for each cluster does not exceed a predetermined score threshold value, a keyword corresponding to a minimum value among the calculated group scores for each cluster is added to the selected keyword combination.

Table 3 below shows an example in which the index n of the conversion is 1, the minimum number of group keywords is 2, and the score threshold is set to 150. [

keyword transform cost CPA Score A 2 1000 1000/2 = 500 500/2 = 250 B 4 800 800/4 = 200 200/4 = 50 C 8 12800 12800/8 = 1600 1600/8 = 200 D 5 3000 3000/5 = 600 600/5 = 120

The keyword B and the keyword D are selected according to the preset minimum number of group keywords (= 2), and the group score per reference cluster is calculated as 800/9 = 88.89.

If the keyword A is added to the selected keyword combination, the group score per cluster is calculated as 1300/11 = 118.18. If the keyword C is added, the group score per cluster is calculated as 2400/17 = 141.18.

Therefore, the keyword A is added to the selected keyword combination, and the finally selected keyword combination can be A, B, and D.

As the automatic bidding cycle arrives, the keywords in the selected keyword combination among all the keywords may be checked 2116 and the automatic bidding 2117 may be performed on the confirmed keywords.

The period of the keyword score calculation may be the same as or different from the automatic bidding period. The bid-related information on which the keyword score calculation is based may use the same day real-time performance data as the bid history information collected from the media server or the advertiser server.

22A to 22D are graphs showing the relationship between the conversion rate and the conversion unit price in consideration of the multiplier of the conversion according to the embodiment of the present invention. Referring to FIGS. 22A to 22D, it can be seen that the score is changed according to the weight of conversion by adjusting the n value of the exponentiation power of the conversion in Equation (2). For example, it can be seen that there are many differences in the score particularly in the sections 2201, 2202, and 2203 shown in FIGS. 22A to 22C.

As described above, the efficiency of each keyword can be expressed by a numerical score. For example, the score can be set to a value obtained by dividing the CPA by the number of conversions.

When the score is defined as such, the conversion efficiency of each keyword is better as the CPA is lower, and the more the number of conversions, the better. That is, in Equation (2), the denominator becomes large, and when the molecule becomes small, the value becomes small.

More specifically, assuming that the CPA of the keyword A is 17,369 won and the conversion is five, the score of the keyword A becomes 3,473 points. Assuming that the CPA of the keyword B is 85,155 and the conversion is 80, the score of the keyword B becomes 1,064 points. Therefore, it can be determined that the keyword B having a low score is a keyword having a higher efficiency.

On the other hand, a weight for the conversion can be given to the score of the keyword as described above, and the weight for each conversion can be set differently for each group. For example, when it is desired to increase the weight of conversion for keywords in a specific group, the value of n may be increased in the score calculation expression of Equation (2) for the group.

When the score is calculated for each keyword, as shown in FIG. 23, automatic bidding can be performed by selecting keywords in a specific section.

23 is a table showing scores calculated for each keyword according to various embodiments of the present invention. Referring to FIG. 23, scores can be calculated and sorted for each keyword according to the above-described method in FIGS. 21 and 22.

In FIG. 23, an upper section or a lower section may be selected according to the score order for a plurality of keywords. The upper section or the lower section may be set to a number or a ratio, and a keyword combination that minimizes the group score for each possible keyword combination may be selected.

Accordingly, the automatic bid can be selectively performed for the keywords corresponding to the upper section and / or the lower section. The keywords of the upper section may be judged to be a group of keywords having higher efficiency and the keywords of the lower section may be judged to be group of keywords having lower efficiency.

On the other hand, the more the keywords whose bids are changed in the lower section, the more passive the bidding becomes, and the fewer the keywords whose bids are changed, the more active the bidding can be.

Hereinafter, an example of a user interface screen provided by the bid management server according to the embodiment of the present invention will be described with reference to FIG. 24 to FIG.

24 is a diagram illustrating a screen for providing a performance prediction function according to an embodiment of the present invention. According to various embodiments of the present invention, conversion performance can be predicted based on history information collected from a media server, an advertiser server, or the like.

 Referring to FIG. 24, when daily data upload to be used for performance prediction is requested, daily data may be uploaded and used for performance prediction. For example, as shown in FIG. 15, it is possible to predict the conversion performance based on the accumulated history information so far. Therefore, as shown in FIG. 5, prediction, exposure, click, click through ratio (CTR), and CPC according to budget or conversion can be proposed based on existing operating history data. In addition, by analyzing the correlation between the budget and the conversion, it is possible to predict the budget when the conversion is selected from the operational history data. If the budget is selected, the conversion can be predicted and proposed. Various mathematical algorithms such as regression analysis can be applied as a method of predicting various data through the correlation analysis.

In addition, the correlation analysis can analyze not only the correlation between the budget and the conversion, but also the relationship between the conversion and the click, the relationship between the conversion and the CPC, and the relationship between the click and the exposure. It is possible to predict exposure, click, CTR, CPC, conversion, conversion rate, conversion cost, etc. according to the correlation between such indicators.

FIG. 25 is a diagram illustrating a screen for providing a target achievement rate per group according to an embodiment of the present invention. FIG. Referring to FIG. 25, the achievement rate of each keyword in the corresponding group can be graphically displayed at the request of the user so that the goal (KPI) achievement rate for each group can be seen at a glance.

FIG. 26 is a diagram illustrating a screen for providing target achievement rates by time of each group according to an embodiment of the present invention. FIG. Referring to FIG. 26, the at least one keyword corresponding to each group can be displayed through a table or a graph as shown in FIG.

25A and 25B are views showing a screen for providing an account setting function according to an embodiment of the present invention. Referring to FIG. 25A, when the account setting function is executed according to various embodiments of the present invention, various items in the account can be set as shown in FIG.

For example, on the account setup screen, you can select the business type for the account. On the other hand, the criterion for classifying the industry may be determined in consideration of the weight and the number of sales keywords. Also, it is possible to request data uploading in the account setting screen, and it is possible to calculate and propose the total monthly budget to be operated in the account through the data preprocessing process. In addition, through the data preprocessing process, it is possible to confirm the keyword, the performance and the weight of each account of the corresponding account, and provide the functions of recommending an account, a daily budget per group, and a daily goal based on the corresponding performance. Thus, in various embodiments of the present invention, it is possible to provide a goal or plan for each account based on the bid-related history data through the data preprocessing process.

As a more specific example, the user may select data for the previous month or last year as data to be reflected in the average operational performance. In addition, as shown in the figure, the total monthly budget to be operated in the account can be automatically calculated for each day of the week, week, and day, and can be proposed to the advertisement manager, and the proposed budget can be normally reflected have. The data update may be automatically updated for a preset period (for example, the previous day, Jeonju, previous month, etc.), or the manager may manually request the update manually. Also, according to various embodiments of the present invention, at least one keyword may be mapped to each of the bidding groups through the account setting screen.

Referring to FIG. 25B, according to various embodiments of the present invention, a target for the account can be set through the account setting screen. For example, the target CPA, the target ROAS (Return On Ad Spend), and the target ROI (Return On Investment) can be set through the account setting screen. Also, as shown in the figure, the monthly goal of the account can be automatically calculated for each day of the week, week, and day, and can be proposed to the advertiser, and the proposed monthly goal can be normally reflected by the advertiser's decision.

In addition, a daily achievement plan for the current month's performance may be proposed on the basis of the set monthly budget and the target, and it may be provided in a graph form.

26 is a diagram illustrating a screen for providing a group setting function according to an embodiment of the present invention. Referring to FIG. 26, according to various embodiments of the present invention, various settings can be made for each group through a group setting screen.

For example, you can enter a group name and select the type of bidding algorithm to apply to that group. For example, a machine bidding mode may be selected or a rule based bidding mode may be selected. The machine bidding mode is a mode for analyzing the hysteresis data in real time in accordance with the target for each group and bidding by the preset bidding algorithm. The rule-based bidding mode is a mode for bidding according to the bidding rule or period set by the administrator.

Referring to FIG. 27, the total monthly budget to be operated in the group can be suggested through the group setting screen. For example, as shown in the figure, the monthly budget of the group can be automatically calculated for each day of the week, week, and day, and can be proposed to the advertiser. The proposed monthly budget can be normally reflected by the advertiser. Through the data preprocessing process, the budget can be distributed by reflecting the YOY based on the previous month data.

On the other hand, if the bidding mode for the group is set to the machine bidding mode in the group setting screen of FIG. 26, a menu for providing the detailed setting function for the machine bidding mode shown in FIG. 27 can be displayed. The menu providing the detailed setting function for the rule-based mode shown in FIG. 29 can be displayed.

27 is a view showing a screen for providing a function of setting a machine bid mode according to an embodiment of the present invention. Referring to FIG. 27, if the machine bidding mode is set for the group, the target (for example, the number of conversions, the amount of sales, the CPA, the ROAS, etc.) of the group can be selected and set. In addition, weighting can be given to items such as conversion and CPA through the target engagement setting.

28 and 29 are views showing screens providing a rule-based bid mode setting function according to various embodiments of the present invention.

28 is a view showing a screen for providing a rule-based bid mode setting function according to an embodiment of the present invention. Referring to FIG. 28, a user can set a rule-based bidding mode for the group and select a performance based bidding mode in detail. If the performance-based bidding mode is selected, the ranking, exposure, click, and conversion values of the previous day, the previous week, the last one month, etc. for each keyword on the basis of the performance reach a preset section (for example, , You can bid automatically for the amount specified in the bid rule.

For example, in the performance-based bidding mode, bidding may be performed every predetermined period (for example, every hour) based on data updated in real time, and bidding may be performed by selecting a specific period.

29 is a diagram illustrating a screen for providing a rule-based bid mode setting function according to an embodiment of the present invention. Referring to FIG. 29, a user can set a rule-based bidding mode for the group and select a history-based bidding mode in detail. If the history based bidding mode is selected, the bid amount set by default can be increased or decreased according to the historical data based on the target ranking.

Meanwhile, the method according to an embodiment of the present invention may be implemented in the form of a program command which can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

The invention has been described above with the aim of method steps illustrating the performance of certain functions and their relationships. The boundaries and order of these functional components and method steps have been arbitrarily defined herein for convenience of description. Alternative boundaries and sequences may be defined as long as the specific functions and relationships are properly performed. Any such alternative boundaries and sequences are therefore within the scope and spirit of the claimed invention. In addition, the boundaries of these functional components have been arbitrarily defined for ease of illustration. Alternative boundaries can be defined as long as certain important functions are properly performed. Likewise, the flow diagram blocks may also be arbitrarily defined herein to represent any significant functionality. For extended use, the flowchart block boundaries and order may have been defined and still perform some important function. Alternative definitions of both functional components and flowchart blocks and sequences are therefore within the scope and spirit of the claimed invention.

The invention may also be described, at least in part, in the language of one or more embodiments. Embodiments of the invention are used herein to describe the invention, aspects thereof, features thereof, concepts thereof, and / or examples thereof. The physical embodiment of an apparatus, article of manufacture, machine, and / or process for implementing the invention may include one or more aspects, features, concepts, examples, etc., described with reference to one or more embodiments described herein . Moreover, in the entire drawings, embodiments may incorporate the same or similarly named functions, steps, modules, etc. that may use the same or different reference numerals, and so forth, Steps, modules, etc., may be the same or similar functions, steps, modules, etc., or the like.

As described above, the present invention has been described with reference to particular embodiments, such as specific constituent elements, and limited embodiments and drawings. However, it should be understood that the present invention is not limited to the above- And various modifications and changes may be made thereto by those skilled in the art to which the present invention pertains.

Accordingly, the spirit of the present invention should not be construed as being limited to the embodiments described, and all of the equivalents or equivalents of the claims, as well as the following claims, belong to the scope of the present invention .

100: Portal Site Web Page 110: Search Window
120: Keyword advertisement area 200: Agent server
201: bid management server 201: advertisement information management server
203: bid management information DB 204: collection information DB
205: Medium information DB 206: Account information DB
210: media server 220: communication network
230: Advertiser server 240: User terminal
610: log information collection module 620: log information analysis module
621: normal data determination unit 622: redundant data removal
623: tracking log update unit 630: analysis information transmission module
710: Account setting module 720: Group setting module
730: Budget and Goal Setting Module 740: Information Collection Module
750: bid determination module 760: automatic bid processing module
770: Performance data management module 810: Account
820: campaigns 821, 822, 823, 831, 832, 833: group

Claims (15)

A media server 210 for providing a keyword advertisement reflecting the bid ranking of the requested keyword in response to a keyword search request by the user terminal 240 to the user terminal 240, The agent server 200 performs a bid for a keyword advertisement with the keyword server 210 according to the selection of a specific keyword advertisement among a plurality of keyword advertisements provided to the user terminal 240 through the media server 210, And an advertiser server (230) for providing an advertiser's web page corresponding to the advertisement server (230) to the user terminal (240), the automatic bid method in a keyword advertisement operating in the agent server (200)
In the agent server 200, setting at least one keyword for keyword advertisement bidding for each account set for each advertiser;
In the agent server 200, according to execution of a tracking script included in a web page of the advertiser provided to the user terminal 240, inflow tracking data including information related to a connection of the user terminal 240 is transmitted to the user terminal 240 From the terminal (240);
In the agent server 200, according to execution of a tracking script included in a web page of the advertiser provided to the user terminal 240, conversion tracking data including information related to switching of the user terminal 240 is transmitted to the user terminal 240 From the terminal (240);
The user terminal 240 connected to the advertiser server 230 through the media server 210 which is preset in the user terminal 240 connected to the advertiser server 230 corresponding to the account, From the media server 210, information relating to connection or switching of the media server;
Receiving, at the agent server (200), the same day real-time performance data for at least one keyword set in an account corresponding to each advertiser server from the media server (210);
Calculating, by the agent server (200), a score for each keyword based on real-time performance data on the same day for the keyword advertisement collected from the media server (210);
Calculating a group score for each keyword cluster by combining the plurality of keywords in the agent server 200;
Selecting, by the agent server (200), a keyword combination that minimizes the calculated group score for each keyword cluster;
Determining, at the agent server (200), a bid for each keyword in the selected keyword combination based on information related to connection or conversion received from the media server (210) and the user terminal (240); And
And performing, at the agent server (200), a keyword advertisement bid for the selected keyword with the determined bid price to the media server (210)
The score for each keyword is calculated on the basis of the CPA and the number of conversions for each keyword and is calculated as a ratio of the exponent of the CPA to the number of conversions, Automated bidding in keyword ads that adjusts weights.
The method according to claim 1,
Determining, at the agent server, whether a set automatic bidding cycle has arrived; And
Further comprising the step of, in the agent server, calculating a keyword-specific score when the automatic bidding cycle arrives.
delete delete delete delete The method according to claim 1, wherein the step of calculating a group score for each keyword cluster by combining the plurality of keywords comprises:
Calculating a keyword-specific group score for each combination of the plurality of selected keywords;
Adding the keywords that are not included in the combination of the selected keywords to the combination of the selected keywords in sequence to ship the group score for each cluster; And
And adding to the selected keyword combination a keyword corresponding to a minimum value among the calculated group scores for each community when the recalculated group score does not exceed a predetermined score threshold value. Automatic bidding method for keyword ads.
A media server 210 for providing a keyword advertisement reflecting the bid ranking of the requested keyword in response to a keyword search request by the user terminal 240 to the user terminal 240, The agent server 200 performs a bid for a keyword advertisement with the keyword server 210 according to the selection of a specific keyword advertisement among a plurality of keyword advertisements provided to the user terminal 240 through the media server 210, And an advertiser server (230) for providing a web page of an advertiser corresponding to the advertiser to the user terminal (240), wherein in the bid management server (201) of the keyword advertisement,
An account setting module 710 for setting an automatic bidding method for at least one keyword set for each account corresponding to each advertiser;
Information related to connection and switching of the user terminal 240 is collected from the user terminal 240 in accordance with the execution of the tracking script included in the web page of the advertiser provided to the user terminal 240, Information related to connection or switching of the user terminal 240 connected to the advertiser server 230 through the media server 210 which is preset among the user terminals 240 connected to the advertiser server 230, An information collection module 740 for receiving and collecting real-time performance data on the same day for at least one keyword set in the account corresponding to the keyword 230 from the media server 210;
Calculating a score for each keyword based on real-time performance data on the same day for at least one keyword keyword advertisement collected through the information collection module, calculating a group score for each keyword group by combining a plurality of the keywords, Determining a bid for each keyword in the selected keyword combination based on information related to connection or conversion received from the media server and the user terminal, selecting a keyword combination that minimizes the group score for each keyword cluster, (750); And
And an automatic bid processing module (760) for performing automatic bid at the calculated bid price,
The score for each keyword is calculated on the basis of the CPA and the number of conversions for each keyword and is calculated as a ratio of the exponent of the CPA to the number of conversions, A bid management server in a keyword advertisement that adjusts weights.
The method according to claim 8, wherein the bid determination module comprises:
Wherein the bid management server determines whether a set automatic bidding cycle has arrived and calculates a score for each keyword when the automatic bidding cycle arrives.
delete delete delete delete The method according to claim 8, wherein the bid determination module comprises:
Calculating a group score for each of the plurality of selected keywords based on the score for each group and re-calculating the group score for each group by sequentially adding the keywords not included in the combination of the selected keywords to the combination of the selected keywords,
Wherein the re-computed keyword group score is added to the selected keyword combination if the re-computed group score does not exceed a preset score threshold.
A computer-readable recording medium storing a program for executing the method of claim 1.

Images