KR20150088348A - Targeting automation device and method - Google Patents

Abstract

A targeting automation apparatus and method are disclosed. A targeting automation device for automatically selecting a target for transmitting content for a marketing target product for advertisement and sales promotion includes a target customer setting section for setting a target customer; A step-by-step target selecting unit for selecting, as a step-by-step target, a customer who wishes to transmit the content step by step among the target customers; And a customer response receiver for confirming a response customer for the content transmitted to the target according to the step, wherein the target-specific target selection unit predicts a response probability of the target candidate customer according to a result of analyzing the characteristics of the response customer, And setting a step target range.

Description

[0001] TARGETING AUTOMATION DEVICE AND METHOD [0002]

The present invention relates to a targeting automation apparatus and method.

Recently, as the information and communication infrastructure such as the Internet and data communication has expanded, it is considered that business using this is the best investment item and has unlimited potential.

Since most of the customers are acquiring information on products through this information communication infrastructure, the online advertising field using the information communication infrastructure is expected to provide business operators with a new marketing field that can replace or supplement the off- .

The information communication infrastructure uses banner ads, e-mail ads, sales promotions, and sales of short messages (SMS), which launch simple images on the Internet site and link their advertisement pages.

However, the effectiveness of such marketing techniques is not very positive. In the case of e-mail and short messages, indiscriminate dispatch is perceived as spam by the customer, which causes the marketing effect to deteriorate.

In order to solve this problem, target advertisements have emerged in order to grasp the customer's tendency based on the customer information acquired in advance and to select and transmit the advertisement that the customer is interested in. Korean Unexamined Patent Application Publication No. 2002-0072016 discloses a web user target advertisement method in which an advertisement window of an associated advertisement corresponding to a field of interest of a web user is automatically implemented before being transferred to another web screen based on a search word or a selected category .

However, in this case, the advertisement is organized around the customer, and in the presence of a large number of advertisements, the advertisement suitable for a specific customer is selected. Therefore, There are different differences.

The above-described background technology is technical information that the inventor holds for the derivation of the present invention or acquired in the process of deriving the present invention, and can not necessarily be a known technology disclosed to the general public prior to the filing of the present invention.

Korean Patent Publication No. 2002-0072016

In the present invention, the following targets are selected step by step based on the response of some customers to marketing (advertisement and promotion promotion) by the employer, and the contents are sent out so that, even when the customer characteristic information such as age and sex is insufficient, The present invention provides a targeting automation apparatus and method that can achieve superior marketing effects by utilizing behavior information of a target.

The present invention analyzes customer responsiveness to a specific content and selects targets targeted to be highly responsive to the content, thereby enhancing the marketing effect and reducing the content transmission, which may be regarded as spam, Apparatus, and method.

According to the present invention, the reaction probability of a specific content is predicted according to the analysis result of the customer, and the target range of the next step is set. If the response is below a predetermined level, And to provide a targeting automation apparatus and method capable of acquiring the target.

An object of the present invention is to provide a targeting automation apparatus and method capable of increasing the overall reaction rate by optimally distributing a plurality of marketing contents according to the response probability of customers, thereby increasing the overall advertising revenue.

Other objects of the present invention will become readily apparent from the following description.

According to an aspect of the present invention, there is provided a targeting automation device for automatically selecting a target for transmitting a content for a marketing target product for promotion of advertisement and sales, comprising: a target customer setting unit for setting a target customer; A step-by-step target selecting unit for selecting, as a step-by-step target, a customer who wishes to transmit the content step by step among the target customers; And a customer response receiver for confirming a response customer for the content transmitted to the target according to the step, wherein the target-specific target selection unit predicts a response probability of the target candidate customer according to a result of analyzing the characteristics of the response customer, And a step target range is set.

The target candidate customer may be the remaining target of the target customer except the step-by-step target.

The stepwise target selecting unit may include a reaction customer analyzing unit for analyzing the characteristics of the reaction customer to calculate a degree of similarity between the target candidate customer and the reaction customer; A response probability predicting unit for predicting a response probability of the target candidate customer based on the similarity; And a target range setting unit that sets the similarity rank when the reaction probability itself or the amount of change is equal to or greater than a predetermined reference value to the lower limit of the rank of the next step target range.

The response customer analyzer compares the distribution function of the target selected in the previous step with a distribution parameter of a given variable parameter to calculate a reactivity score, applies a reactive score to the target candidate customer, The degree of similarity with the customer can be calculated.

The response customer analysis unit may normalize and compare the previous step selection target distribution function and the response customer distribution function.

And a content transmitting unit for transmitting the content to the target for each step selected by the target selecting unit for each step.

If the content is plural, the stepwise target selecting unit selects the target content. A reaction customer analysis unit for analyzing the characteristics of the responding customer for each of the plurality of contents to calculate the degree of similarity between the target candidate customer and the responding customer; A response probability predicting unit for predicting a response probability of the target candidate customer based on the similarity for each of the plurality of contents; And a target range setting unit configured to set a next target range for each of the plurality of contents so that a content having a highest reaction probability or a reaction probability of a predetermined reference value or more is transmitted to the target candidate customer.

The target range setting unit may first select a target candidate customer satisfying an advertiser's requirement for each of the plurality of contents among the target candidate customers, and then use the response probability to divide and set the next target range.

According to another aspect of the present invention, there is provided a targeting automation method for automatically selecting a target for transmitting marketing and marketing content for promotion to a marketing target in a targeting automation apparatus, and a recording medium on which a program for performing the targeting automation is recorded.

According to one embodiment, a method of automating a target includes: (a) establishing a target customer; (b) selecting one of the target customers as a first stage target; (c) confirming a response customer for the marketing content transmitted to the selected target; (d) analyzing characteristics of the reaction customer; (e) predicting a response probability of the target candidate customer according to the analysis result; (f) setting a next step target range based on the reaction probability, selecting a next step target, and returning to step (c).

The target candidate customer may be the remaining target of the target customer except the step-by-step target.

Wherein the step (d) calculates a degree of similarity of the target candidate customer with the reaction customer by analyzing the characteristics of the reaction customer, and the step (e) predicts a reaction probability of the target candidate customer based on the degree of similarity , The step (f) may set the similarity rank when the reaction probability itself or the amount of change is equal to or greater than a predetermined reference value to the lower limit of the rank of the next step target range.

The step (d) includes the steps of: calculating a reactivity score by comparing a distribution function of a target selected in a previous step with a distribution function of a response customer for an arbitrary variable parameter; applying a reactivity score to the target candidate customer And calculating the degree of similarity with the reaction customer.

In the step (d), the previous step selection target distribution function and the response customer distribution function may be normalized and compared.

The step (c) may transmit the marketing content to the predetermined target of the step.

According to another embodiment, a method of automating targeting comprises the steps of: (a) establishing a target customer; (b) selecting one of the target customers as a first stage target; (c) confirming a response customer for the marketing content transmitted to the first stage target; (d) analyzing characteristics of the reaction customer; (e) performing a step of estimating a response probability of a target candidate customer according to the analysis result, wherein the steps (c) to (e) are performed separately for each of the plurality of marketing contents, (f) And distributing the next target range for each of the plurality of contents so that the content having the highest probability of response or a predicted reaction probability of a predetermined reference value or more is transmitted.

In the step (f), after selecting a target candidate customer satisfying an advertiser's requirement for each of the plurality of contents among the target candidate customers, the next step target range may be set using the reaction probability.

Other aspects, features, and advantages will become apparent from the following drawings, claims, and detailed description of the invention.

According to the embodiment of the present invention, the following target is selected step by step on the basis of the response of some customers to the contents to be marketing (advertisement and promotion promotion) by the employer, It is possible to obtain excellent marketing effect by utilizing the behavior information of the customer even in a situation where the customer is insufficient.

In addition, by analyzing customer responsiveness to a specific content, it is possible to increase the marketing effect and to reduce the content transmission, which may be regarded as spam for other customers, by targeting the customers who are expected to respond to the content highly.

In addition, by responding to a specific content, the reaction probability is predicted according to the analysis result of the customer and the target range of the next step is set. If the reaction is expected to be below a predetermined level, There is an effect to be able to do.

In addition, the plurality of marketing contents are optimally distributed according to the response probability of the customers, thereby increasing the overall reaction rate, thereby increasing the overall advertising revenue.

1 is a block diagram of a configuration of a targeting automation apparatus according to an embodiment of the present invention;
FIG. 2 is a block diagram of a target-
FIG. 3 is a view illustrating a customer distribution function for analyzing characteristics of a reaction customer,
Figure 4 is an illustration of a customer table and a reactive score table,
5 is a diagram illustrating a reaction probability prediction curve for the next step target range setting,
6 is an exemplary diagram for explaining the reaction probability prediction;
7 is a flowchart of a method of automating a targeting process according to an embodiment of the present invention;
FIG. 8 is an exemplary view of a response probability table for each customer generated by the targeting automation apparatus according to another embodiment of the present invention; FIG.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It is to be understood, however, that the invention is not to be limited to the specific embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, the terms "comprises" or "having" and the like refer to the presence of stated features, integers, steps, operations, elements, components, or combinations thereof, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

Also, the term "part" or the like, as described in the specification, means a unit for processing at least one function or operation, and may be implemented by hardware, software, or a combination of hardware and software.

It is to be understood that the components of the embodiments described with reference to the drawings are not limited to the embodiments and may be embodied in other embodiments without departing from the spirit of the invention. It is to be understood that although the description is omitted, multiple embodiments may be implemented again in one integrated embodiment.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the like elements throughout. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.

FIG. 1 is a block diagram of a targeting automation apparatus according to an embodiment of the present invention. FIG. 2 is a block diagram of a target selecting unit according to a step. FIG. 3 is a diagram illustrating a customer distribution function FIG. 4 is a diagram illustrating a customer table and a response score table, FIG. 5 is a diagram illustrating a response probability prediction curve for setting a next step target range, and FIG. 6 is an exemplary diagram for explaining a reaction probability prediction .

The targeting automation apparatus according to an embodiment of the present invention selects a marketing target of the next stage based on a response of customers selected as a marketing target at a current stage of an arbitrary merchandise, So that it can exhibit excellent marketing effect even in a situation where there is no basic knowledge of the customer. In addition, it is possible to maximize the effect of the marketing cost by predicting the reaction probability and setting the next target range based on the prediction.

Referring to FIG. 1, a targeting automation apparatus 100 according to an embodiment of the present invention includes a target customer setting unit 110, a target level selection unit 120, a customer response receiving unit 130, and a controller 150 . And may further include a content transmission unit 140, a user input unit 160, and / or a display unit 170 according to an embodiment.

The target customer setting unit 110 sets a customer to be a marketing target as a target customer for an arbitrary merchandise to be marketed.

Here, the target customer may be all or a part of members who subscribe to a predetermined service. The predetermined service means various online / offline services operated by a member such as an online shopping mall service, an Internet portal service, a web mail service, and the like.

In establishing the target customer, the target customer setting unit 110 may limit the total number of customers to which the marketing content (advertisement and sales promotion content) is to be transmitted, or may limit the number of target customers per each step.

For example, the total number of target customers can be limited to 10,000. Alternatively, 1000 persons may be selected for each step, and the content may be transmitted. If the response rate is less than the set value (for example, 10%) after the lapse of a predetermined time, the target customer selection at the next step may be stopped.

Also, the target customer setting unit 110 may define a target customer based on the member information. For example, a member can be selected as a target customer only if the membership score is above the reference value or if the membership level is above the reference level.

The content transmitting unit 140 may be configured to receive an advertisement including information on a product to be marketed for a target customer identified according to the stepwise target identification information transmitted from the target selecting unit 120, And transmits the content.

The marketing content is transmitted using the pre-registered contact information to the target customer. For example, the marketing content is transmitted to the target customer by using the contact information such as e-mail, SMS, multimedia message (MMS) A content transmission method can be used.

Or an advertisement displayed on an application executed by the target customer through the identification information capable of identifying the target customer may be a part of the marketing content. At this time, the content transmission unit 140 may transmit the marketing content (i.e., advertisement) to be displayed on the terminal based on the terminal identification information of the target customer.

The content transmitting unit 140 may be included as a component of the targeting automation apparatus 100 according to the present embodiment as shown in FIG. 1, but may be a content server connected to the targeting automation apparatus 100 It is possible to transmit the marketing contents to the target customers by stages while operating independently of the targeting automation apparatus 100. [

The customer response receiving unit 130 confirms the customer who has responded to the content transmitted to the selected target customer in the stage selected by the stepwise target selecting unit 120, which will be described later.

Responsive customers include, for example, those that select (eg click) a link, such as a URL contained in the content, or purchase a product that promotes and promotes sales through the content, Means a customer who has acted.

The customer response receiving unit 130 can identify the responding customer by collecting the identification information of the customer who performed the action when the matching action of the customer such as the click of the customer or the purchase of the product occurs. Here, the customer identification information may be collected by causing a return message including the customer identification information to be transmitted to the targeting automation apparatus 100 when the URL is clicked or the product is purchased.

Or the return message may be transmitted to the content server, and the content server may collect the customer identification information of the customer who has transmitted the return message itself or the return message, and may transmit the collected customer identification information to the targeting automation apparatus 100.

Here, the customer identification information may be at least one of a customer ID, a name, a resident registration number, a telephone number, and a device ID (e.g., a UDID (Unique Device ID)) of the customer holding terminal. Such customer identification information can be used only to distinguish the customer from other customers while preventing the disclosure of information about the customer's personal identity through encryption.

The step-by-step target selection unit 120 selects a customer who wants to transmit contents in a stepwise manner among the target customers set in the target customer setting unit 110.

In the case of the first stage target selection, it is possible to select a target customer as the first stage target as many as the predetermined number randomly according to the arbitrary selection method. Or a customer who meets a predetermined condition according to a condition selection method, the whole or a part thereof may be arbitrarily selected as a first stage target.

In the step-by-step target selection of two or more stages, the customer response to the transmitted content for the current target is analyzed, and as a result, the customer who shows a relatively high degree of similarity in comparison with the relatively high- The target range of the next step can be set based on the reaction probability predicted for the next step.

The stage-specific target selection unit 120 may generate target-specific target identification information for identifying a selected customer after the target customer is selected in stages, and may transmit the generated target-identifying information to the content transmission unit 140.

Or the stepwise target selection unit 120 may transmit the stepwise target identification information to the control unit 150 to be transmitted to the content server (not shown) to which the marketing content is to be transmitted.

Referring to FIG. 2, the stepwise target selection unit 120 may include a response customer analysis unit 210, a response probability predicting unit 220, and a target range setting unit 230.

The response customer analysis unit 210 analyzes the characteristics of the responding customer who are interested in the content based on the customer response received at the customer response receiving unit 130.

This is to maximize the marketing effect by allowing the customers who have similar attributes as the responding customers to be the target of the priority when selecting the next target customers.

The response customer analysis unit 210 extracts the characteristics of the responding customers differentiated when compared with all of the target customers who have transmitted the content at the present stage, calculates the similarity scores with the responding customers to the target candidate customers using the extracted features do.

Here, the target candidate customer may be the rest of the target customer set by the target customer setting unit 110, excluding the customer who has transmitted the content to the current stage. Alternatively, the target candidate customer may be the entire target customer set in the target customer setting unit 110.

In analyzing the characteristics of the responding customer, customer characteristic variables including general variables such as customer profile information and customer behavior information, and extended parameters such as order history for a certain period of time are used.

The customer profile information may include variable parameters such as age, gender, and address. The customer action information may include order information such as order time, order price, order unique number, order category, and destination information such as destination time, destination address, destination unique number, and destination category as variable parameters.

Expansion variables can include parameters such as order total amount, order number grade, order average price grade, order item cluster, destination cluster, price sensitivity, time discount rate, transaction cost, imaginary budget, risk avoidance tendency for a certain period .

The response customer analysis unit 210 compares the distribution function of the target customer selected at the present stage with the distribution function of the response customer for at least one of the variable parameters included in the general variable and the expansion parameter, And the reactive score is calculated. Then, the reactivity score is applied to the variable parameter of the target customer, and the similarity is calculated with the reaction customer.

First, the calculation of the reactivity score may be performed by comparing the selected target customer distribution function value with the response customer function value. Here, the comparison operation may be, for example, a subtraction operation or a division operation.

Referring to Fig. 3, the distribution function f (x), g (x) of the target customer and response customer for an arbitrary parameter x is shown. In Fig. 3, a customer distribution function f (x), g (x) that has undergone a normalization process is shown for comparison.

For the value x1 in the variable parameter x, the relative value of the current target customer is f (x1) and the relative value of the responding customer is g (x1).

Here, the difference (g (x1) -f (x1)) or the ratio (g (x1) / f (x1)) between the relative value of the reaction customer and the relative value of the current target customer can be used for calculating the reactive score have.

The process of assigning the calculated reactivity score to the customer having the value of the variable parameter x1 may be repeated for every variable parameter so that the reactivity score can be given to the target candidate customers.

For example, if the variable parameter indicates the age, it is assumed that the number of current target customers is 1000 and the number of responding customers is 100. [

When the number of responding customers corresponding to '20s' is 200, and the number of responding customers corresponding to '20s' is 40, The relative value of the current target customer in the '20s' is 0.2 (= 200/1000) and the relative value of the responding customer is 0.4 (= 40/100).

In this case, when calculating the reactive score using the difference, the response score of the '20s' in the parameter parameter corresponding to the age becomes 0.2 (= 0.4-0.2). Since the reactive score has a positive value, it means that the '20s' shows high reactivity in the variable parameter indicating age.

Next, the score of similarity between each target candidate customer and the reaction customer is calculated by combining the calculated reactivity score for each parameter of each target candidate customer.

Referring to FIG. 4A, customer characteristic variables V1, V2, V3,... For a plurality of customers C1, C2, C3,. Where V1 is sex, V2 is residence address, and V3 is age. For example, in case of customer C1, the sex is male, the residence address is Guro-Dong, and the age is 30.

In the customer table, a reactivity score table as shown in FIG. 4B is generated by applying the reactivity score to the response customer calculated for each customer characteristic variable for each customer.

The degree of similarity score for each customer can be calculated by performing a predetermined calculation on the reactivity score for any customer. For example, it is possible to calculate the similarity score for each customer by applying various methods such as adding or multiplying by applying weights to customer characteristics. In the case of customer C1, w1 * a1 + w2 * b1 + w3 * c1 + ... may be the final similarity score. Here, w1, w2, and w3 are weights of the variable parameters V1, V2, and V3.

When the reaction customer analysis unit 210 calculates the similarity score with the current-stage reaction customer for the target candidate customer, the reaction-probability predicting unit 220 predicts the reaction probability using the score.

Here, the response probability is a probability that the target candidate customer having the similarity score is expected to respond when the content is transmitted. For example, it is possible to calculate the percentage of customers who are expected to respond when the content is transmitted to the target candidate customers in the order of the target candidate customers arranged in rank order having a high degree of similarity score, as the response probability.

Referring to FIG. 5, there is illustrated a response probability curve 300 in which the horizontal axis indicates the degree of similarity score and the vertical axis indicates the response probability. It is assumed that the response rate is 25.06% (1253 responses) when the target customer is 100,000 and the first stage target is 5000 customers.

In this case, the target candidate customers are 95000 persons excluding the first-stage target customers, and it is predicted that when the order is ranked according to the similarity score described above, the probability of response gradually decreases from about 0.52. For example, if the target range of the second stage is set to 10,000, the probability of the reaction can be estimated to be about 0.43.

This is because the higher the similarity rank is, the lower the degree of similarity with the current-level response customer, and the higher the probability that the content will not respond to the content.

The prediction of the response probability may be performed based on the similarity score calculated by the response customer analysis unit 210. The correlation between the similarity score and the response probability may be determined by a predetermined algorithm or may be determined experimentally or statistically .

One embodiment for predicting the response probability is illustrated in FIG. In the case where the number of target candidate customers is 1000, similarity score according to the similarity rank is shown.

6, the reaction probability can be determined by the following equation.

Response probability = target candidate customer's similarity score / highest similarity score

For example, a customer having a similarity rank of 1 can predict a response probability of 1 (= 100/100), and a customer having a similarity rank of 500 can predict a response probability of 0.80 (= 80/100).

This is merely an example, and the correlation between the similarity score and the response probability may be variously set.

The target range setting unit 230 sets the next step target range based on the reaction probability predicted by the response probability predicting unit 220. [

For example, the degree of similarity in which the response probability is equal to or greater than a preset reference value may be determined as the lower limit of the rank, and the target candidate customers whose rank is higher than the lower limit of the rank may be set as the next step target range.

For example, in the case of the response probability curve of FIG. 5, when the response probability is limited to 0.4 or more, 20000 is determined as the lower limit of the rank, and target candidate customers whose similarity rank is 1 to 20000 can be set as the next stage target range .

Or the degree of similarity at the point where the variation of the reaction probability exceeds a preset reference value, that is, the point at which the response probability is sharply lowered, as the lower limit of the rank, and the target candidate customers having the rank higher than the lower limit of rank, It can also be set.

For example, in the similarity degree and response probability table of FIG. 6, the response probability 0.80 of the similarity rank 500 and the response probability 0.10 of the similarity rank 501 are the response probability 0.70 while the similarity rank is changed by 1 (= 501-500) (= 0.80-0.10), and the variation of the reaction probability changes very rapidly. In this case, the similarity rank 500 is determined as the lower limit of the rank, and the target candidate range can be set to the next target range only for the target candidate customers with the similarity rank of 1 to 500.

This is because the marginal utility is likely to be minus (-) in terms of the effect of marketing cost at a point where the variation of the response probability variation with respect to the degree of similarity rank change is high, and the point can be set to the lower rank limit .

The target range setting unit 230 may be omitted according to the embodiment. In this case, the reaction probability information of the target candidate customer predicted by the response probability predicting unit 220 is transmitted to the advertiser server or the advertisement agency server, Alternatively, the advertisement agency server may set the target range according to a predetermined criterion based on the response probability information of the target candidate customer, and then perform the target marketing.

In addition, the stepwise target selection unit 120 may further include a customer filtering unit 240 for filtering a customer satisfying a predetermined condition among the target candidate customers included in the next step target range.

The customer filtering can be performed for a customer listed in advance or for a customer satisfying a predetermined filtering condition. For example, a list of customers who have previously transmitted content over a specified number of times to a previous product is listed in advance, and customers on the list can be effectively filtered if the current content can be considered as spam.

Or if the customer who has already transmitted the content at the present stage has a high similarity score, if the content is continuously received, the content may be regarded as spam, which may be the object of filtering.

Therefore, the filtering condition may be, for example, whether or not the current content is sent, whether or not the content is sent a predetermined number of times or more, and the filtering target customers satisfying the filtering conditions may be listed in advance.

Referring again to FIG. 1, the targeting automation apparatus 100 may include a user input unit 160 and a display unit 170.

The user input unit 160 may receive user input information such as the number of target customers from the user, the selection condition of the target customer, the content and type of the marketing content, the reference value for determining the lower limit, and the filtering condition.

Here, the user input unit 160 may be a keyboard, a mouse, or a touch sensitive screen interlocked with the display unit 170.

The display unit 170 may display the response probability curve drawn by the response probability predicting unit 220 on the screen so that the user can confirm the response probability curve. The user can check the response probability curve displayed on the screen through the display unit 170, determine a reference value for determining the lower limit for the target range setting, and the like, and input the determination.

Also, the display unit 170 can display text, a graph or the like regarding the number of customer reactions, the response rate, the importance of the parameters involved in calculating the similarity degree, and the frequency of use.

The control unit 150 controls each component of the targeting automation apparatus 100 to perform the above-described functions.

In addition, the controller 150 may determine whether or not a predetermined end condition is satisfied, and may terminate the targeting automation if the end condition is satisfied.

The termination condition may be limited to the total number of shipped items and / or the ratio of the reaction rate.

First, the detailed termination condition can be configured step by step when the total number of destinations is limited. The step configuration items may include the number of objects to be shipped per step, the response wait time for next shipment for each step, and the like.

As a method of constructing the steps, the number of shipment items per step may be divided into equal steps (for example, 1: 1: 1: 1: 1 at the time of 5th shipment) 1: 3: 6 at the time of the third stage dispatch). In addition, the response wait time for the next shipment by step can be set manually (for example, 20 minutes).

Next, in the case of limiting the reaction rate, detailed termination conditions can also be configured step by step. Step configuration items may include the number of items to be shipped per step as well as the number of items to be shipped, and the waiting time for the next shipment for each shipment.

As a step composition method, it is possible to equally divide the number of objects to be shipped per step (for example, 1000 persons per step). In addition, if the reaction wait time for the next shipment by stage is manually set, or if there is more than a predetermined rate of reaction according to the result of measuring the response rate after waiting time, the next step is constituted. It will be possible to terminate without constructing a step.

In addition, the reaction rate to be compared may be used by estimating the reaction rate after a certain time (for example, one day) as a result of the reaction so far.

Hereinafter, a targeting automation method performed in the above-described targeting automation apparatus 100 will be described with reference to FIG.

7 is a flowchart of a method of automating a targeting according to an embodiment of the present invention. Each step of FIG. 7 may be performed by each of the components of the targeting automation apparatus 100 shown in FIG.

First, the target customer setting unit 110 sets all or a part of the member customers as a target customer for the advertisement and the promotion target merchandise (step S400).

The stepwise target selection unit 120 selects a first stage target among the target customers set in the target customer setting unit 110 (step S410). As described above, the arbitrary selection method or the condition selection method may be applied to the first step target selection.

When the first stage target is selected, the content transmitter 140 transmits the content to the selected target (step S420). Alternatively, the targeting automation apparatus 100 may be connected to the content server instead of the content transmission unit 140 so that the content transmission may be performed by transmitting the identification information about the target of the first stage to the content server.

The customer response receiving unit 130 identifies a response customer who is interested in the content among the first stage targets (step S430). The response customer may be a customer who has performed an action in accordance with the marketing objective, for example, a customer who clicks a URL included in the content or purchases a product advertised in the content.

The confirmation of the reaction customer can be performed by sending a return message including the customer identification information to the targeting automation apparatus 100 or the content server when a marketing purpose matching action occurs. When a return message is sent to the content server, the content server may transmit the customer identification information that has sent the return message to the targeting automation apparatus 100.

The control unit 150 confirms whether or not a predetermined end condition is satisfied (step S440). Here, the termination condition may include a limitation on the total number of objects to be shipped and / or a reaction rate limit.

If the termination condition is satisfied, the flow advances to step S480 to terminate the targeting automation.

If the termination condition is not satisfied, the reaction customer analysis unit 210 analyzes the characteristics of the reaction customer interested in the content based on the customer reaction received at the customer reaction receiving unit 130 (step S450).

The analysis of the responding customer can be done by extracting the characteristics of the responding customer differentiated from the total of the target customer who transmitted the content at the present stage and calculating the similarity score with the responding customer for all the customers from this characteristic. The feature extraction and similarity score calculation of the reaction customer has been described in detail with reference to FIG. 3 and FIG. 4 in advance.

Upon completion of the analysis of the current-stage reaction customer, the response-probability predicting unit 220 predicts the response probability of the target candidate customer based on the similarity (step S460).

Here, the response probability is a probability that responds when a content is transmitted to a customer having the similarity degree score. For example, for the target candidate customers arranged in order of similarity score ranks, The percentage of customers who are expected to respond if they are sent can be calculated as a response probability.

The target range setting unit 230 sets the next step target range based on the predicted reaction probability (step S470). The target candidate customers having the reaction probability itself or a variation thereof equal to or greater than a predetermined reference value may be included in the next step target range. Here, the reference value may be manually set by the user, or may be a change amount of the response probability versus the similarity rank change amount in which the marginal utility based on the cost-effectiveness marketing effect becomes negative.

The customer filtering may be performed such that excluding the customer who has already transmitted the content or the customer who has transmitted the content other than the predetermined number of times in step S470.

When the next step target range is set, the process returns to step S420 to transmit the marketing content to customers belonging to the next step target range.

It will be appreciated that the above-described targeting automation method may be performed in an automated procedure in a time-sequential order by a software program or the like embedded in the digital processing apparatus (targeting automation apparatus 100). The codes and code segments that make up the program can be easily deduced by a computer programmer in the field. In addition, the program is stored in a computer-readable information storage medium, and the program is read and executed by a computer to implement the method. The information storage medium includes a magnetic recording medium, an optical recording medium, and a carrier wave medium.

In the above description, the case where there is one marketing content for promoting advertisement and sales has been mainly described, and a case where there are a plurality of marketing contents will be described with reference to the following drawings.

8 is a diagram illustrating an example of a response probability table for each customer generated by the targeting automation apparatus according to another embodiment of the present invention.

When there are a plurality of marketing contents for promoting advertisement and sales to a target customer set by the target customer setting unit 110, it is very inefficient to transmit all marketing contents to all target customers in terms of cost .

Therefore, it is necessary to distribute and transmit a plurality of marketing contents. At this time, the marketing automation device 100 is used to select and transmit the marketing contents expected to have a high reaction probability among a plurality of marketing contents for each target customer, And increase overall advertising revenues by increasing overall customer response.

The targeting automation apparatus and the stepwise target selecting unit according to the present embodiment have the same configuration as the targeting automation apparatus 100 and the stepwise target selecting unit 120 shown in FIG. 1 and FIG. 2, .

The response probability predicting unit 220 predicts the response probabilities of the target candidate customers for each marketing content, and generates a response probability table for each customer. Such a customer-specific response probability table is illustrated in FIG.

8, it is predicted that customers A, B, and C will have a response probability of 0.5, 0.7, and 0.2 for the advertisement 1, and the response probabilities of the customers A, B, and C are 0.7, , And it is predicted that customers A, B, and C will have response probabilities of 0.2, 0.1, and 0.5 for Advertisement 3, respectively.

In other words, the response probability of customer A is 0.5, 0.7, and 0.2 for advertisement 1, 2, and 3, the response probability of customer B is 0.7, 0.4, and 0.1, and the response probability of customer C is 0.2, 0.3, .

In this case, the target range setting unit 230 according to the present embodiment selects optimal marketing contents in a form that maximizes profits while satisfying the various requirements of the advertiser by utilizing the reaction probability predicted for each customer, So that the target range can be set for each content.

For example, when an advertiser sets requirements such as age, sex, and residence area for a marketing target, it first determines whether the customer meets the requirement, and then increases the maximum profit by utilizing the response probability The advertisement (i.e., marketing content) can be distributed.

If the response probability is utilized, the target range can be set so that the advertisement having the highest response rate is transmitted to the customer satisfying the predetermined requirement. If the response probability is predicted as illustrated in FIG. 8, The response probability of advertisement 1 is highest for customer B, and the response probability of advertisement 3 is the highest for customer C, so that the customer A is included in the target range of the advertisement 2 content and the customer B is the advertisement 1 < / RTI > content, and the customer C may be set to be included in the target range of the advertisement C. [

Alternatively, the target range may be set such that an advertisement showing a reaction probability of a predetermined reference value or more is transmitted to a customer satisfying a predetermined requirement. 8, when the criterion value is 0.5, the customer A is included in the target range of the advertisement 1 content in addition to the customer B so that not only the advertisement 2 but also the advertisement 1 can be transmitted to the customer A. .

According to this, a plurality of marketing contents can be distributed to customers so as to exhibit an optimal advertisement effect, and a higher response rate can be obtained as compared with a random advertisement.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention as defined in the following claims And changes may be made without departing from the spirit and scope of the invention.

100: Targeting automation apparatus 110: Target customer setting unit
120: step-by-step target selection unit 130:
140: content transmission unit 150:
160: user input unit 170: display unit
210; Reaction customer analysis unit 220: Reaction probability predicting unit
230: target range setting unit 240: customer filtering unit

Claims (17)

A targeting automation device that automatically selects a target for transmitting content for a marketing target product for advertisement and promotion,
A target customer setting unit for setting a target customer;
A step-by-step target selecting unit for selecting, as a step-by-step target, a customer who wishes to transmit the content step by step among the target customers; And
And a customer response receiver for confirming a response customer for the content transmitted to the stepwise target,
Wherein the stepwise target selection unit predicts a response probability of a target candidate customer according to a result of analyzing the characteristics of the response customer to set a next step target range.
The method according to claim 1,
Wherein the target candidate customer is the remaining target of the target customer except for the stepwise target up to the present stage.
The method according to claim 1,
Wherein the stepwise target selection unit comprises:
A reaction customer analysis unit for analyzing the characteristics of the reaction customer to calculate the degree of similarity between the target candidate customer and the reaction customer;
A response probability predicting unit for predicting a response probability of the target candidate customer based on the similarity;
And sets a similarity rank when the reaction probability itself or the amount of change is equal to or greater than a predetermined reference value as a lower limit of the rank of the next step target range.
The method of claim 3,
The response customer analyzer compares the distribution function of the target selected in the previous step with a distribution parameter of a given variable parameter to calculate a reactivity score, applies a reactive score to the target candidate customer, And calculates a degree of similarity with the customer.
5. The method of claim 4,
Wherein the response customer analysis unit normalizes and compares the previous step selection target distribution function and the response customer distribution function.
The method according to claim 1,
And a content transmitting unit for transmitting the content to the stepwise target selected by the stepwise target selecting unit.
The method according to claim 1,
If the content is plural,
Wherein the stepwise target selection unit comprises:
A reaction customer analysis unit for analyzing the characteristics of the responding customer for each of the plurality of contents to calculate the degree of similarity between the target candidate customer and the responding customer;
A response probability predicting unit for predicting a response probability of the target candidate customer based on the similarity for each of the plurality of contents;
And a target range setting unit that sets a next target range for each of the plurality of contents so that a content having a highest response probability or a reaction probability of a predetermined reference value or more is transmitted to the target candidate customer.
8. The method of claim 7,
Wherein the target range setting unit first selects a target candidate customer satisfying an advertiser's requirement for each of the plurality of contents among the target candidate customers and then distributes and sets the next target range using the reaction probability. .
A targeting automation method for automatically selecting a target for transmitting marketing and marketing promotion contents to a marketing target in a targeting automation device,
(a) establishing a target customer;
(b) selecting one of the target customers as a first stage target;
(c) confirming a response customer for the marketing content transmitted to the selected target;
(d) analyzing characteristics of the reaction customer;
(e) predicting a response probability of the target candidate customer according to the analysis result;
(f) setting a next step target range based on the response probability to select a next step target, and returning to step (c).
10. The method of claim 9,
Wherein the target candidate customer is a remainder of the target customer, excluding the target for each step up to the present stage.
10. The method of claim 9,
Wherein the step (d) includes analyzing characteristics of the reaction customer to calculate a degree of similarity between the target candidate customer and the reaction customer,
Wherein the step (e) predicts a response probability of the target candidate customer based on the degree of similarity,
Wherein the step (f) sets the similarity rank when the reaction probability itself or the amount of change is equal to or greater than a predetermined reference value to the lower limit of the rank of the next step target range.
12. The method of claim 11,
The step (d)
Calculating a reactivity score by comparing a distribution function of a target selected in a previous step with a distribution function of a reaction customer for an arbitrary variable parameter;
And calculating a degree of similarity between the target customer and the reaction customer by applying a reactive score to the target candidate customer and combining the reactivity score.
13. The method of claim 12,
Wherein the step (d) normalizes and compares the previous step selection target distribution function and the response customer distribution function.
10. The method of claim 9,
Wherein the step (c) transmits the marketing content to the predetermined target of the step.
A targeting automation method for automatically selecting a target for transmitting a plurality of marketing contents for advertising and sales promotion to a marketing target in a targeting automation apparatus,
(a) establishing a target customer;
(b) selecting one of the target customers as a first stage target;
(c) confirming a response customer for the marketing content transmitted to the first stage target;
(d) analyzing characteristics of the reaction customer;
(e) predicting a response probability of a target candidate customer according to the analysis result,
Wherein the steps (c) to (e) are performed separately for each of the plurality of marketing contents,
(f) distributing a next step target range for each of the plurality of contents so that a content predicted to have a highest response probability or a response probability higher than a predetermined reference value is transmitted to the target candidate customer.
16. The method of claim 15,
Wherein, in the step (f), the target candidate customer satisfying an advertiser's requirement is first selected for the plurality of contents among the target candidate customers, and then the next step target range is set to be distributed using the reaction probability. Way.
A recording medium on which a program that can be read by a digital processing apparatus is recorded for performing the targeting automation method according to any one of claims 9 to 16.

Images