KR20030069128A - The method and system of forecasting Reach of internet advertising on internet - Google Patents

Abstract

PURPOSE: A system capable of forecasting a reach(rate of target receivers who are exposed to an advertisement at least one time) and a method thereof are provided to forecast a reach result corresponded to an impression with respect to an individual medium or plural media combination based on advertisement exposure data with respect to an actual target receiver being recorded in a server by inserting a cookie having a unique ID into a client web browser. CONSTITUTION: A client web browser(101) connects to a medium having an advertisement, analyzes an HTML code of the medium, and requests an advertisement transmission to an advertisement management server(103). An advertisement servicing server(104) checks whether a cookie exists in the client web browser(101). If a cookie exists in the client web browser(101), a count impression data with respect to a unique ID of the corresponding cookie are increased in a database(106). If a cookie does not exist in the client web browser(101), a cookie having a unique ID is inserted into the client web browser(101), impression data of new unique ID in the database(106) are '1', an advertisement transmission requested medium name is stored, and the advertisement servicing server(104) instructs an advertisement transmission. The database(106) accumulates impression numbers count data according to unique IDs by an instruction of the advertisement management server(103) continuously. An analysis server(105) checks the number of unique IDs of an individual medium or plural media in impression data with respect to an advertisement executed for a predetermined term, calculates the number of unique viewers, calculates and stores a reach.

Description

The method and system of forecasting Reach of internet advertising on internet}

Usually, in the offline advertising industry, the advertising goal of media planning is set by the Reach of advertisement exposure. Setting advertising goals through the Reach of advertising impressions means that when you run an ad on a particular medium, you can predict how many people will be exposed to that ad, depending on how much the ad is served.

In the offline advertising industry, such as TV, based on the ratings data of individual programs surveyed by a certain number of panels (viewed as the Reach of individual programs), statistical probabilistic calculations can be used to calculate various programs on TV media. Reach results are predicted by eliminating the number of overlapping target audiences for a single advertisement run by.

Looking at how the traditional internet advertising industry predicts Reach, basically, like the offline advertising industry, the individual media surveyed on a certain number of panels (web media that can be paid paid. Web media is collectively referred to as 'medium', which means that the term 'medium' in Internet advertising means 'web medium'. First, we survey the panel's number of visits to a particular medium, assuming that it is the ratings of the entire advertising space of a particular medium, and attempt to predict the reach of one advertisement executed in multiple media by a statistical method. However, this method of statistically utilizing the visit data of the panel is a good way to know the target audience of the entire media where the advertisement is run.However, in the advertisement space, several advertisements are randomly rolled (in one advertisement space). Because multiple advertisements are posted in random order, and the advertisements that are posted change every time a user makes a request to the server for that advertisement page.) It is impossible to know the exact ratings of a single advertisement itself. Therefore, it is impossible to predict the results of Reach when advertising is carried out in multiple media. In other words, the ratings obtained by using panel data in the Internet advertising industry, like the offline advertising industry, are not about the actual advertisement but about the web page where the advertisement is loaded. Therefore, the conventional technology can predict the reach required in the original media planning. It's not a technology.

In order to solve this problem, the present invention does not use the data of the panel, but uses the data recorded in the server according to the actual transmission request of all target recipients. That is, in response to a request for transmitting an advertisement of a client web browser, a cookie having a unique ID is inserted into the browser, and each client web browser having a cookie with a unique ID is assumed to be a unique viewer. The server side of the transmission accurately measures the ad exposure of the target audience, calculates the actual ratings of one advertisement, stores it in the data storage of the server, accumulates it, and based on this, When advertising is executed, the method of predicting the Reach versus impression is used.

Here, predicting Reach means that the most basic Reach 1+ (Reach of target audience exposed to one or more advertisements, usually Reach 1+, unless Remarks is specified separately). Not only predicts how many times the target audience has been exposed to an ad on average, but also predicts all the detailed information about each frequency of Reach (Reach n +: n target recipients exposed to more than n ads). It means.

Information that includes such details of Reach is commonly referred to as Reach's Frequency distribution. The method of estimating the change in Reach's frequency distribution against exposure is based on empirical data. There are two ways to fit and a statistical probability distribution.

In general, however, in advertising media, it is known that the change in the medium's Reach as time or exposure increases shows a convex modified exponential model with an upper limit without exception. . Because of this mathematical characteristic, the actual offline advertising industry derives a mathematical formula that can be obtained by minimizing the error of Reach's frequency distribution versus exposure based on the data for a long period of time, and is mainly planning media based on this.

In the present invention, in obtaining the change change function compared to impression based on the advertisement exposure data of the target audience, a method of fitting the mathematical formula based on the results of the execution of a plurality of advertisement campaigns as described above is used.

The present invention is not a statistical method using a conventional panel, but by inserting a cookie with a unique ID into a client web browser, the individual media or multiple media combinations are based on advertisement exposure data for actual target audiences recorded on the server. As it predicts Reach results compared to impressions, Internet advertising planners can execute Internet advertising campaigns through specific media selection criteria and media planning goals, such as offline advertising campaigns.

In the present invention, in order to find out how many times a person has been exposed to an advertisement, a cookie having a unique ID is inserted into a user's browser and assuming that the user's browser is a unique target viewer. Analyzes the request to the ad server for the advertisement of the browser, stores the media name and the number of impressions for which the advertisement was executed for each unique ID in the data storage, and repeats the above process for the later executed advertising campaign and stores the accumulated data. Based on the mathematical formula, the Reach change function of the medium compared to the impression must be processed through a mathematical formula.

However, in actual advertisement execution, in general, a plurality of media are selected to execute an advertisement, and the total Reach obtained when the plurality of media is combined is the Reach data for each medium obtained through the above process, and the multiple media without duplicates. It can be obtained through the entire Unique Viewer data.

Finally, the present invention provides a system that allows Internet advertising planners to predict Reach results compared to impressions for an advertising campaign that is subsequently executed through a web browser, using the data of the advertisement server side obtained through the above process. Should be.

1 is a system configuration diagram that can predict the reach of the present invention Internet advertising.

2 is a diagram showing the overall processing executed by the present system.

FIG. 3 illustrates how an advertisement management server creates a unique ID of a cookie inserted into a client web browser in the process of FIG.

First, the system includes a client web browser that accesses the Internet, analyzes an HTML code of a medium (site) containing an advertisement, requests an advertisement to be transmitted to an advertisement management server, and permits the use of cookies by the advertisement management server;

In response to the advertisement transmission request, a client web browser checks whether a cookie previously inserted by an advertisement management server exists. If there is no cookie, a unique ID is inserted into a home appliance cookie in the client web browser. Command to save the media name that made the advertisement transmission request with the number of impresseion counts as 1, and in case of re-request, check the data storage unit with the unique ID of the client web browser and the media name that made the advertisement transmission request. An advertisement management server that issues an advertisement transmission instruction to an advertisement transmission server after giving a command to increase the number of impression counts for the unique ID;

An advertisement transmission server that transmits an advertisement for a request of a client web browser according to an instruction of an advertisement management server;

A data storage unit for counting the number of impressions for each unique ID of the advertisement by the command of the advertisement management server, and storing a medium name for which an advertisement transmission request is made; The number of unique IDs for each medium stored during the advertisement execution period is counted from the impression count data for each unique ID stored in the storage unit, that is, the number of unique viewers for each medium is calculated to calculate the reach for each medium on which the advertisement is executed. A processing unit calculating a reach of all the plurality of media on which the same advertisement is executed based on the data;

A data storage unit for continuously storing Reach data for each of the media and the plurality of media calculated by the processing process at each subsequent advertisement execution;

A processing unit for obtaining a change change function relative to the impression for each medium and the entire plurality of media based on the data stored in the data storage unit;

A data storage unit for storing a change change function compared to an impression obtained through the process;

Access the data store via a client web browser on the Internet, display a list of media on which advertisements have been executed, select a single or multiple media from the list, and enter the total Impresseion value to be executed. Program that outputs on the; characterized in that it comprises a.

In addition, the method of the present invention comprises the steps of the ad management server accepts the request to send the advertisement of the client web browser, and checks the presence of the cookie inserted previously;

If a cookie does not exist in the client web browser, inserting a cookie having a unique ID into the client web browser, and if the cookie exists, retrieving the unique ID of the cookie and the name of the medium that requested the advertisement from the data storage;

The advertisement management server, recording the number of requests for advertisement transmission, that is, impression and advertisement transmission medium name, in the data storage unit according to the unique ID inserted in the cookie, according to whether the cookie exists in the client web browser;

Based on the impression data counted for each unique ID of the cookie, the number of unique IDs is searched for each medium where the advertisement is executed and for each of the plurality of media where the same advertisement is executed. 3 steps of converting the Reach of the entire plurality of media to the data storage unit;

Based on the data stored in the above step, the advertisement is subsequently executed on the same medium again, and if additional data is accumulated by repeating steps 1 to 3, the fourth step of obtaining a Reach change function relative to the impression for each medium and the entire plurality of media;

A five step of outputting the result of the above four steps through a web browser according to a client's request on the Internet;

Characterized in that it comprises a.

Hereinafter, the objects and features and advantages of the present invention described above will be described in more detail with reference to the accompanying drawings and preferred embodiments.

1 is a system configuration diagram that can predict the reach of the present invention Internet advertising.

As shown in the figure, first, a client web browser 101 connected to the Internet 102 accesses a medium containing an advertisement, and then analyzes the HTML code of the medium to request advertisement transmission to the advertisement management server 103. .

The ad management server 104 checks whether a cookie exists in the client web browser 101, and if the cookie exists, increments the count of Impression data for the unique ID of the cookie in the database 106, and if the cookie does not exist Insert a cookie with a unique ID into the client web browser 101, set the Impression data of the newly created Unique ID to 1 in the database 106, store a media name for which an advertisement transmission request is made, and advertise the server 104. ) To send the ad.

The database 106 continuously accumulates data counting the number of impressions for each Unique ID by an instruction from the advertisement management server 103. The analysis server 105 checks the total number of unique IDs in the individual media and the entire plurality of media in the impression data of the advertisement executed for a certain period of time to calculate the number of unique viewers, and then calculates and stores the reach. The analysis server 105 calculates the Reach change function compared to the impression and stores it in the database 106 when the advertisement is executed on individual media and plural media through repetition of the above process.

The advertisement planner accesses the client web application 107 through the client web browser 101, selects a list of media having pre-execution results, inputs an arbitrary impression, and the client web application 107 performs the above process. By substituting the value into the Reach change function of Impression vs. Impression of the stored database 106, a Reach prediction value of Impression of individual media or a plurality of media is obtained.

3 illustrates how the advertisement management server 103 creates a unique ID of a cookie inserted into the client web browser 01 in the process of FIG.

The ad management server 103 expresses the server's time in 13 digits in units of 1/1000 seconds, extracts a random 5 digits, and adds a total of 18 digits to the previous 13 digits representing the time. Generate ID

2 is a process performed by the present system, and the method of the present invention will be described in detail with reference to the embodiment at each step.

First, when the client requests the advertisement management server 103 to transmit the banner advertisement through the web browser 101 (step S100), the client accesses the advertisement management server 103 (step S110), and the advertisement management server 103. In step S120, the client web browser 101 determines whether a cookie exists.

If the cookie inserted by the advertisement management server 103 does not exist, the cookie having a unique ID is inserted into the client web browser 101 (step S131), and the advertisement management server 103 advertises the advertisement transmission server 104. The request is sent to the advertisement (step S132). When the advertisement transmitting server 104 transmits an advertisement to the client web browser 101 (step S133), as shown in [Table 1], the Impression data having the new Unique ID as the key value is generated in the database 106, and the Impresseion column is generated. Is to count 1 (step S134).

However, if there is already a cookie inserted into the advertisement management server 103 by the client web browser 101 through the above process, the advertisement management server 103 sends an advertisement transmission request to the advertisement transmission server 104 after step S120. In operation S141, the advertisement transmission server 104 may transmit an advertisement (operation S142). The advertisement management server 103 retrieves data from the database 106 using the unique ID of the cookie of the client web browser 101 which made the advertisement transmission request and the name of the medium that made the advertisement transmission request as a kit value. The data is updated to increase the data as shown in [Table 2] (step S143).

Thus, when impression data for each medium is accumulated in the database 106 as shown in [Table 1] and [Table 2] during the advertisement execution period, after the advertisement execution period is over, the analysis server 105 checks the [table] in the database 106. 3] Unique Viewer data is calculated and stored for each medium as shown in step S151, and as shown in [Table 4], Unique Viewer data is calculated by removing the unique viewer overlap between each medium for a plurality of mediums for which one advertisement is executed. In step 106, the storage is performed at step S161.

After the above steps S151 and S162, data such as [Table 3] and [Table 4] are stored in the database 106, and if another advertisement is subsequently executed on the same medium, the same steps at S100 and S162 are performed. In addition to [Table 3], UniqueViewer results are accumulated and stored as shown in [Table 5].

If the result of the execution of the advertisement in one medium more than two times is accumulated, the advertisement management server 103 automatically based on the advertisement execution period and the number of times the advertisement is exposed for each medium based on the data in [Table 5] above. Accordingly, the Reach change function is calculated and stored compared with the impression satisfying the following Equation 1 (step S152).

[Equation 1]

(注: y is Reach, x is Impression, a is the actual maximum reach rate (media-specific) that can be reached in the medium, and b is the constant for that medium that specifies the slope of the curve represented by the function above)

For example, if you want to calculate the curve of reach exposed to the advertisement more than once from the result data of the advertisement execution for one month in the media A of [Table 5], you know the x, y value first, If there are at least two results executed by different imps, the constant a, b of the media can be calculated. Currently, Media A has a total of three executions for a month, of which Reach is exposed at least one ad in two campaigns with 1000 imp. . Since uv (+1) is 170 when executed as the second 2000imp, the value x1 = 1000, x2 = 2000, y1 = 110, y2 = 170 is substituted into the above formula. Since the value of the slope b can be obtained, it is possible to find the y (Reach) value for the x value (impression).

Afterwards, if the result of ad execution is continuously accumulated in the media A and the data in [Table 5] is large, the x and y values are input through the above formula to obtain more constant values of a and b of the media. . In this case, by averaging the a and b values according to each execution result, more accurate a and b values can be obtained. As a result, a closer y (Reach) value can be obtained.

However, in order to know the Reach of a plurality of media, as shown in the data in [Table 4], the value obtained by removing the duplicate of the unique viewer for each media is calculated by processing the data, and the following [Equation 2] is used. Then, the constant value (k) for knowing the degree of redundancy for each medium should be calculated.

[Equation 2]

If the medium is two of A and B:

Unique viewer of A, B overall = Unique viewer of A + Unique viewer of B-k (Unique viewer of A * Unique viewer of B)

If the medium is three of A, B and C:

Unique viewer of A, B, C = Unique viewer of A + Unique viewer of B + Unique viewer of C-k1 (Unique viewer of A * Unique viewer of B)-k2 (Unique viewer of B * Unique viewer of C) -K3 (Unique viewer of C * Unique viewer of A) + (Unique viewer overlaid on both A, B, C)

(注: k is a constant that specifies the degree of redundancy between two media. The data shown in Table 4 shows the left-hand side of the formula, and the data shown in Table 3 identifies the uniqueness of the two media. Since the value of the viewer can be obtained, a part of the right term can also be obtained, so that the value of k can be obtained, but when there are three media, Unique Viewer, which is repeatedly exposed to the same advertisement in the entire media of A, B, and C, is almost 0 according to the measured value. Converges to, the number of Unique viewers exposed to both A, B, and C is 0.

The redundancy of each medium calculated through Equation 2 above is corrected by a constant value (k) which determines the redundancy when the results of Table 4 are further accumulated in another advertising campaign. As a result, the degree of redundancy for each medium close to the measured value is calculated.

For example, if the advertisement is executed through three media A, B, and C, the same data as [Table 3] is calculated for each of the media A, B, and C (step S151), and A as shown in [Table 4]. Unique Viewer data without duplicates is calculated for each combination of W, B, C, C, A, and A, B, and C (step S161). The value of k, which is a constant specifying the degree of redundancy, is obtained. Afterwards, the advertisement is executed again in three plural media such as A, B, and C. When the result is accumulated, the value of k is calculated again, and the initial k value and the mean are corrected. By applying the process of calculating the individual media until the step S152, using the equation (1) to obtain the change function of the change compared to the impression of the particular combination of multiple media (step S162).

When the Reach change function data, which is finally obtained through the above process, is stored in the database 106 in relation to the impressions of the individual media and the multiple media, the advertisement planner outputs the advertisement execution results to the database 106 through the client web browser 101. Check the list of media that are present, select the media or media to which you want to run the advertisement, enter an arbitrary Impression value, request a prediction, and the client web application 107 accesses the database 106 to compare the stored Impression. By using the Reach change function, the Reach prediction is output through the client web browser 101 (step S170).

Looking at the configuration of the invention described in detail through the embodiments of the present invention, those skilled in the art to which the present invention belongs, various substitutions, modifications within the scope without departing from the spirit of the present invention It will be apparent that the present invention may be modified. Accordingly, modifications to future embodiments of the present invention will not depart from the technology of the present invention.

As described above, according to the present invention, the Internet advertising planners can set specific and clear advertising goals as the existing offline advertising planners because it can predict the Reach compared to the impression of individual media and multiple media in internet advertising. In the case of media planning, advertisements can be executed by selecting media through rational standards. In this way, it has the effect of dramatically improving the work efficiency and methods of the representatives such as the Internet advertising planner, but in the second place, the Internet media is the same as the four offline media (TV, newspaper, radio, magazine). Advertisement goals can be set and the results can be evaluated, which will attract advertisers who have run advertisements only in the offline media, and will have a dramatic effect on the overall size of the Internet advertising market. .

Claims (2)

In the method for predicting the reach of the Internet advertisement (ad delivery rate) by the data transmission request data of the web browser recorded on the ad management server, The first time the client sends an ad delivery request to the ad management server, the ad management server checks the client web browser for a cookie previously inserted by the ad management server; If the cookie does not exist, the advertisement management server determines that it is the first advertisement request, generates an arbitrary unique ID (Unique ID) through the server system's time, and sends a cookie including the unique ID to the advertisement transmission request. Insert into one client web browser; Count Impression (number of advertisement exposures) to 1 for the Unique ID of the inserted cookie, and store the media (site) name to which the client visited and sent an advertisement transmission request to the data storage; If there is a cookie inserted by the ad management server when the client sends an advertisement, the data storage unit searches for the unique ID of the cookie and the name of the medium that requested the advertisement in the key storage, and retrieves the Impression data of the corresponding unique ID. Increment to count to save; The number of unique IDs is calculated for each individual medium or multiple media for which an advertisement is executed, that is, for calculating a unique viewer (target audience) exposed to the advertisement, and the above steps are repeated, and the individual media for which the advertisement execution result is performed two or more times. And a mathematical formula for a Reach change function versus an Impression capable of predicting Reach of a plurality of media combinations; Methods including methods. In the system that can predict the Reach (ad delivery rate) of the Internet advertising by the data transmission request data of the web browser recorded on the ad management server, An access client web browser requesting the transmission of advertisements over the Internet; An advertisement transmission server for transmitting advertisements in response to a client request; An advertisement management server that inserts a cookie having a unique ID into a client web browser when the client first requests an advertisement, and collects impression data about advertisements transmitted for each unique ID and a name of a medium for which the client web browser requests an advertisement; A data storage unit for storing the impression data for each Unique ID obtained through the above process and the name of the medium for which an advertisement transmission request is requested by the client web browser; An analysis server extracting unique viewer data from a data storage unit storing impression data of the advertisement, and finally calculating Reach data of impressions of individual media or a combination of media on which the advertisement is executed using a specific mathematical formula; When the user (advertiser) checks the list of media that can predict the Reach according to the impression by the execution result of the execution through the web browser, selects the media to execute the advertisement in the future and inputs an arbitrary Impresseion value. A client web application that outputs a Reach prediction of an advertisement on a browser; Internet advertising Reach prediction system, characterized in that.