KR20220170322A - Method for Determining Triggering Method And Message Based Advertisement Platform Therefore - Google Patents

Abstract

The present invention provides a triggering type determining method and a message-based advertisement platform therefor. According to one aspect of the present invention, provided are the triggering type determining method and the message-based advertisement platform therefor which identify an attraction time point, which is a time point for sending an advertisement message, based on behavioral details of a customer, generate customer segment information, which is information for clustering the customer, and determine an advertisement message sending method for each customer segment by using the attraction time point and the customer segment information.

Description

Method for Determining Triggering Method And Message Based Advertisement Platform Therefore

The present invention relates to a method for determining a triggering method and a message-based advertisement platform therefor.

The contents described in this part merely provide background information on the present embodiment and do not constitute prior art.

Mobile communication terminals including smart phones are becoming popular, and message-based commerce advertisements are increasing. The message-based commerce advertisement is an advertisement method that sends an advertisement message to the user's mobile communication terminal to induce product purchase by using the user's personal information.

However, most of the advertising messages used in commerce advertisements are illegal spam messages due to leakage of personal information, or messages according to site subscription or use of consent to receive advertisement messages. Therefore, even if a user receives an advertisement message, he or she regards it as a spam message and does not check the content or deletes the message in many cases. Even if the user checks the content of the message, the advertisement message contains not advertisement contents personalized for each user, but the entire sent message, so that the purchase of individual users is not induced.

Korean Patent Publication No. 2020-0040498 (2020.04.20)

According to one aspect of the present disclosure, a method of sending an advertisement message to trigger a customer to purchase a product may be determined based on customer behavior details.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

According to one aspect of the present disclosure, in the triggering method for a message-based advertising platform to send an advertising message to customers in a customer group, behavior details of customers corresponding to advertising messages and / or behavior details of customers in the advertising target platform The process of collecting as customer behavioral information; generating attraction point information by determining an attraction time point, which is a time point for sending an advertisement message, based on the customer behavior information; generating customer segment information, which is information for clustering customers based on the customer behavior information; and determining a triggering method for each customer segment using the attraction point information and the customer segment information.

According to one aspect of the present disclosure, based on the customer's behavioral history, an attraction point, which is a point in time for sending an advertisement message, is determined, customer segment information, which is information for clustering customers, is generated, and the attraction point and customer segment information are used. Therefore, by determining a method of sending an advertisement message for each customer segment, the effect of sending an advertisement message is maximized.

The effects of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

1 is a conceptual diagram illustrating a message-based advertising platform according to an embodiment of the present disclosure.
2 is a block configuration diagram illustrating a triggering information management unit according to an embodiment of the present disclosure.
3 is a block diagram illustrating a service target management unit according to an embodiment of the present disclosure.
4 is a flowchart illustrating a method for determining a customer triggering method according to an embodiment of the present disclosure.

Hereinafter, some embodiments of the present disclosure will be described in detail through exemplary drawings. In adding a reference code to the components of each drawing, it should be noted that the same components have the same numbers as much as possible even if they are displayed on different drawings. In addition, in describing the present disclosure, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the present disclosure, the detailed description will be omitted.

In addition, in describing the components of the present disclosure, terms such as second and first may be used. These terms are only used to distinguish the component from other components, and the nature, order, or order of the corresponding component is not limited by the term. Throughout the specification, when a part 'includes' or 'includes' a certain component, it means that it may further include other components without excluding other components unless otherwise stated. . In addition, the '... Terms such as 'unit' and 'module' refer to a unit that processes at least one function or operation, and may be implemented as hardware, software, or a combination of hardware and software.

The detailed description set forth below in conjunction with the accompanying drawings is intended to describe exemplary embodiments of the present disclosure, and is not intended to represent the only embodiments in which the present disclosure may be practiced.

The message-based advertising platform of the present disclosure is a platform for providing customer-customized advertising messages to customers. Such an advertising message may be a message including a hashtag generated customized for the customer (hereinafter referred to as 'hashtag message') or a message including general promotional information such as coupon information (hereinafter referred to as 'promotional message'). not limited

Advertising messages in the present disclosure are short message service (Short Message Service, SMS), long message service (LMS) or multimedia message service (Multimedia Messaging Service, MMS), push alarm, in-app It may be provided using an In-App Messaging (IAM) method, but is not limited to this method.

1 is a conceptual diagram illustrating a message-based advertising platform according to an embodiment of the present disclosure.

The message-based advertising platform 10 according to an embodiment of the present disclosure includes a datalake (100), a first priority product allocation unit (120), and a triggering information management unit (130). , a service target management unit (service target management unit, 140), an advertisement service provider (advertisement service provider, 160), and a log collector (log collector, 180) are included in whole or in part. The message-based advertising platform 10 shown in FIG. 1 is according to an embodiment of the present disclosure, and all components shown in FIG. 1 are not essential components, and some components may be added, changed, or deleted in other embodiments. can

Although FIG. 1 illustrates each component 100 to 180 of the message-based advertising platform 10 as a device, this is for convenience of explanation, and in another embodiment, each component 100 to 180 of the message-based advertising platform is It may be implemented as a software module or processor that performs the functions of.

Although FIG. 1 shows a data pipeline of one method for convenience of explanation, this is an example and the method of the data pipeline used in the message-based advertising platform 10 is limited to the method shown in FIG. I never do that.

The data lake 100 is a repository where data used to perform message-based advertisements is stored, and unstructured/semi-structured/structured data can be collected from various data sources. The data lake 100 may be located on all or part of an on-premise environment, an edge computing environment, and a cloud environment. The data lake 100 may include various interfaces for collecting, storing, accessing, searching, referencing, or cross-referencing data.

Data collected and stored by the data lake 100 include, for example, phone call history, message transmission/reception history, web usage history, app usage data, etc., and customer product purchase history , Customer activity data within the Internet shopping platform, such as shopping cart details, wish history, coupon usage history, rate plan, additional services, bill payment details, location of mobile communication terminal, loss history of mobile communication terminal, membership use information, etc. Use of communication services Data, media viewing content history, paid content content/rate plans, etc. Media service usage data, navigation usage history, navigation POI (Point of Interest) history, user driving habits, search history, navigation usage data, music purchase history, music streaming history However, it may include all or part of music service usage data, such as music source search history, etc., but is not limited to this.

The data lake 100 may further collect data generated or processed by the message-based advertising platform 10, customer behavior details in response to advertising messages, and/or customer behavior details within the advertisement target platform. For example, the data lake 100 includes various scores obtained by the first-priority product allocation unit 120 or data on the first-priority products allocated to each customer, information on determining the attraction point managed by the triggering information management unit 130, and customer segments. Information, a triggering method acquired by the service target management unit 140, a click probability prediction model, information on a hashtag message sending target, information on a promotion message sending target, report information on each advertisement message sending target, and an advertisement service providing unit 160 ) acquires personalized hashtags, hashtag messages, promotional messages, delivery history of each message, information on clicking links (e.g. URLs, etc.) of customers following message delivery, and access to specific web pages/apps by customers following message delivery Customer behavior logs or product information (such as sales, discount rate, sales quantity, etc.) All or part of the data related to information collected from the advertising target platform may be further collected, but is not limited thereto.

The first-priority product allocation unit 120 uses the data stored in the data lake 100 to select the first-priority product, which is the product that the customer is most likely to purchase, among the products in the product group, and advertises to the customer. Allocate as a product.

The first product allocating unit 120 converts a product interest score, which is a customer's interest in each product, or a product purchase score, which indicates the probability that the customer purchases at least one product, for each product in the customer's product group, which is the score of each product. earned as a score. The first priority product allocation unit 120 calculates a first score, which is a score for selecting the first priority product, using the product score. The first-priority product allocation unit 120 may further calculate a product attractiveness score representing a sales contribution according to the sending of an advertising message. The first priority product allocation unit 120 selects the customer's first priority product based on the first score, or calculates a final score in which the product attractiveness score is reflected as a weight in the first score, and based on the final score and the first priority score. You can select the 1st priority product with .

The first priority product allocating unit 120 may generate cut lines for all or some of the products in the product group. In this case, for the top N products having high final scores (N is a natural number), the first product allocation unit 120 selects products whose first scores are equal to or higher than the cutoff in order of highest final scores of each product. By comparing the first score of the product with the cut line until it is found, the first product may be selected and assigned to the customer.

The information on the first-priority product of each customer generated/obtained/calculated by the first-priority product allocator 120 may be used by the service target management unit 140 to extract a target for sending a promotion message.

The triggering information management unit 130 uses the data stored in the data lake 100 to determine the time of attraction for sending an advertisement message, manages (creates and updates) information on the time of attraction, and selects customers based on the customer behavior log. Manages (creates and updates) customer segment information, which is clustered information. The triggering information management unit 130 will be described later in detail with reference to FIG. 2 .

The service target management unit 140 determines a triggering method for each customer segment using data stored in the data lake 100 and/or data managed by the triggering information management unit 130, and an advertisement to be sent to the customer based on the determined triggering method. It determines all or part of the type of message, the content of the advertising message, and the timing of sending the advertising message. Here, the triggering method means a method of sending an advertisement message for triggering a customer to purchase a product. The service target management unit 140 transmits triggering information, which is a triggering method determined for each customer segment, to the advertisement service provider 160 so that the advertisement service provider 160 generates and sends each advertisement message according to the triggering information. .

The service target management unit 140 may generate report information for each target customer (or each customer segment) of each advertisement message or perform statistical analysis and transmit the information to a predetermined manager. The service target management unit 140 will be described later in detail with reference to FIG. 3 .

The advertisement service provider 160 generates a hashtag message and/or a promotion message and sends it to the customer. The hashtag message may be generated based on a personalized hashtag including a product to be exposed to customers and an exposure order. The advertising service provider 160 generates an advertising message to be sent according to the customer segment to which each customer belongs based on the triggering information, but the contents of each advertising message (eg: promotional content, recommended products, personalized hashtags, etc.) Stars can be created.

The advertisement service providing unit 160 receives the customer's action history for the sent message and provides an advertisement service in a form corresponding to the customer's activity. For example, when a customer clicks a specific connection link (eg, URL or app launch link) in a message, the advertisement service provider 160 executes or exposes an app or web page according to the type of advertisement message sent to the customer. Advertising service can be provided so that the

The log collection unit 180 receives and transmits various data collected or generated by the message-based advertising platform 10 to the data lake 100 . The log collection unit 180 is an advertising target platform of the advertising message provided by the message-based advertising platform 10 (eg, a web service platform such as Internet shopping and web content subscription, an app service platform such as mobile shopping and app content subscription, etc. ) Customer behavior details generated in ) may be further received and transmitted to the data lake 100 .

The log collection unit 180 may pre-process the received data. For example, the log collection unit 180 may transform or process data received from various sources into a form that can be utilized by an analysis environment.

2 is a block configuration diagram illustrating a triggering information management unit according to an embodiment of the present disclosure.

The triggering information management unit 130 according to an embodiment of the present disclosure includes all or part of an attraction time discrimination unit 202 and a customer segment definition module 204. The triggering information management unit 130 shown in FIG. 2 is according to an embodiment of the present disclosure, and all components shown in FIG. 2 are not essential components, and some components may be added, changed, or deleted in other embodiments. there is.

Although FIG. 2 illustrates the triggering information management unit 130 as a device, this is for convenience of explanation. In another embodiment, the triggering information management unit may be implemented as a software module or processor that performs the functions of each component 132 to 134. can

The attracting time determination module 202 determines an attracting time point, which is a time point for sending an advertisement message, using the data stored in the data lake 100, and generates or updates information on the attracting time point. The attraction point information is the information on the purchase time according to the customer behavior type. Specifically, the attraction time determination module 202 determines the attraction time based on the customer's behavior history corresponding to the advertisement message and/or the customer's behavior history on the advertisement target platform (hereinafter collectively referred to as 'customer behavior history'). Customer behavior types are acquired by extracting customer behaviors that serve as standards. The customer's behavior to be extracted is, for example, when a customer churns out after clicking on an ad message's connection link, and there is a churn after clicking on an ad message connection link. Page) If there is a dropout after access, if there is a dropout after browsing the advertising target platform, if there is a dropout after putting the sale product of the advertising target platform in the shopping cart, if the purchase page of the advertising target platform (e.g., order form page, etc.) ) can extract all or part of the case of leaving after visiting, but it is not limited to this case.

The attraction point determination module 202 calculates a customer behavior type based on the extracted customer behavior. For example, the attraction point determination module 202 may count the number of each customer's behavior and include the corresponding behavior in the customer's behavior type if there is a predetermined threshold or more. Alternatively, the attraction point determining module 202 may extract only customer behavior corresponding to a predetermined customer behavior type from the customer behavior details.

The attraction point determination module 202 is based on the extracted customer behavior or customer behavior type, and purchase, which is the time required for customers who took the same behavior (or customers of the same customer behavior type) to purchase the product from the customer behavior details. Calculate more time required. The purchase time may be calculated by, for example, calculating a median value or an average value of time required for customers who have taken the same action to purchase a product. Accordingly, the attraction point determination module 202 may obtain purchase time for each customer behavior type and generate attraction point information.

Since the customer behavior details are continuously accumulated, the attraction point determination module 202 may update the attraction point information to reflect the latest customer behavior details in the attraction point information. For example, the attraction point determination module 202 periodically or whenever the size of added customer behavior detail data exceeds a preset threshold, the customer behavior type and/or the purchase time required for each customer behavior type are updated. can

The customer segment definition module 204 uses data stored in the data lake 100 to create or update customer segment information, which is information for clustering customers. Specifically, the customer segment definition module 204 performs clustering by designating all or part of customer behavior history as a feature. Customer behavioral history may further include basic information about the customer (eg age, gender, etc.). The customer segment definition module 204 calculates all or some of the clusters calculated by clustering as customer segments, and designates features of the corresponding clusters as reference information of the corresponding customer segment to generate customer segment information. For example, the customer segment definition module 204 uses product-related page access information, advertisement target platform access information, and customer purchase information as features to determine the number of exits after product-related page inflow, the number of ad target platform visits, the number of product purchases, Product purchase performance (purchase amount) and the like can be designated as standard information for each customer segment. An example of customer segment information is shown in Table 1.

Here GMV (gross merchandise volume) means the total amount (monetary value) of goods sold through a specific market during a specific period.

The customer segment definition module 204 may generate customer segment information using a clustering model based on machine learning or deep learning. For example, customer segment definition module 204 may perform K-measn clustering. The customer segment definition module 204 loads a training dataset from the data lake 100 and obtains a centroid for each cluster by applying a K-means cluster algorithm. The customer segment definition module 204 assigns each customer to a cluster by calculating a distance from each centroid of target customers to be advertised. The customer segment definition module 204 aggregates behavior details of customers belonging to each cluster to determine the degree of response to advertisement messages for each customer segment (e.g., the number of products with clicks, the number of clusters that clicked products over a predetermined number). number, link click rate of advertisement messages, etc.) is analyzed. The customer segment definition module 204 may determine the number of customer segments (or number of clusters) based on the analysis result.

The customer segment definition module 204 may configure a learning dataset by extracting customer behavior details corresponding to a specific period from customer behavior details and setting customer behavior details after the corresponding period as GT (GroundTruth). The customer segment definition module 204 learns (trains and tests) the clustering model using the training dataset and determines the optimal K (where K is the number of clusters). The customer segment definition module 204 may determine the optimal K using silhouette analysis. The customer segment definition module 204 may perform feature selection to improve prediction accuracy. Meanwhile, the customer segment definition module 204 may generate customer segment information by cutting off some of the customer clusters based on the response level of the advertisement message. This is to omit sending of the advertising message to the customer cluster where the effect of sending the advertising message is greatly reduced.

Since the customer behavior details are continuously accumulated, the customer segment definition module 204 may update the customer segment information to reflect the latest customer behavior details in the customer segment information. For example, the customer segment definition module 204 may update customer segments and/or standard information of each customer segment periodically or whenever the size of added customer behavior history data exceeds a preset threshold.

The triggering information management unit 130 may transmit and store attraction time information and customer segment information to the data lake 100 . The triggering information management unit 130 may transmit the attraction time information and customer segment information to the service target management unit 140 to determine a triggering method. Alternatively, the service subject management unit 140 may determine a triggering method using the attraction point information and customer segment information stored in the data lake 100 .

3 is a block diagram illustrating a service target management unit according to an embodiment of the present disclosure.

The service subject management unit 140 according to an embodiment of the present disclosure includes a triggering method determination unit 302, a triggering information transmission module 304, and a report module 306. include all or part of The service target management unit 140 shown in FIG. 3 is according to an embodiment of the present disclosure, and all components shown in FIG. 3 are not essential components, and some components may be added, changed, or deleted in other embodiments. there is. For example, in other embodiments the reporting module may be omitted. In another embodiment, the service target management unit further includes a hashtag sending target determination unit (not shown) that determines a hashtag sending target by estimating the probability of clicking a customer's connection link when sending a hashtag message. can do.

Although FIG. 3 illustrates the service target management unit 140 as a device, this is for convenience of description. In another embodiment, the service target management unit may be implemented as a software module or processor that performs the functions of each component 142 to 146. can

The triggering method determination module 302 determines a triggering method for each customer segment using the attraction point information and customer segment information. The triggering method determining module 302 determines all or part of the type of advertisement message to be sent for each customer segment, the combination of advertisement messages to be sent, the content type of the advertisement message, the timing of sending the advertisement message, and the sending period of the advertisement message as a triggering method. However, it is not limited to this method, and any advertising message sending method for triggering a customer to purchase a product may be a triggering method in the present disclosure.

The triggering method determination module 302 may determine an attraction time point, ie, an advertisement message sending time point, as a triggering method for each customer segment using the attraction time information and customer segment information. The triggering method determining module 302 may map a preset triggering method and a customer segment as a triggering method of each customer segment. Alternatively, the triggering method determination module 302 may determine a triggering method for allowing customers of a corresponding customer segment to purchase a product, based on customer behavior details of customers belonging to each customer segment. The triggering method determining module 302 may determine a triggering method for each customer segment by statistically analyzing customer behavioral details or using a learning model pre-learned from customer behavioral details.

The triggering method determining module 302 determines the hash tag message sending target and the promotion message sending target (customer segment) among the advertising message types using the attraction point information and customer segment information, and as a triggering method, the advertising message type for each customer segment. can determine The triggering method determining module 302 is configured to send a first hashtag message, which is a hashtag message about a specific product (eg, a first-ranked product) assigned to a customer among hashtag message sending targets, and a list of a plurality of recommended products. An advertisement message type for each customer segment may be determined by further determining a target to which the second hashtag message, which is the hashtag message, is to be sent.

Alternatively, the triggering method determining module 302 may determine a combination of advertisement messages to be sent for each customer segment and/or a sending time point of the corresponding advertisement message. For example, for customers in the new visitor segment or the returning visitor segment, the triggering method determining module 302 determines a triggering method to send a hashtag message for a popular product of an advertisement target platform and a promotion message for coupon payment at a corresponding incentive time point. can decide

For the breakaway segment, the triggering method determining module 302 may determine a triggering method such that a hash tag message for a product of a product-related page that has been introduced in the past or a promotion message related to a corresponding product is sent at a corresponding attraction time point.

With respect to the new buyer segment, the triggering method determination module 302 may determine a triggering method to send a promotion message (eg, coupon payment promotion message, etc.) for inducing repurchase as soon as a purchase event occurs.

For the high performer segment, the triggering method determining module 302 sends a hashtag message (or a first hashtag message) for a new product and/or a hashtag message (or a second hashtag message) for a list of recommended products. A triggering method can be determined to be dispatched at the time of attraction.

For the low performer segment, the triggering method determining module 302 is related to coupon payment with a lower limit of the purchase amount or a promotion message for recommending a high-priced product and/or a hash tag message related to a high-priced product at an incentive time corresponding to You can decide the triggering method to dispatch.

For the dormant customer segment, the triggering method determining module 302 may determine a triggering method such that a promotion message and/or a hashtag message for a product under an event is sent at a corresponding attraction time point.

The triggering method determining module 302 predicts the probability of clicking the access link of the advertisement message of the customers of each customer segment based on the customer behavior details (hereinafter, "click probability"), and determines the advertisement to be sent based on the predicted click probability. A message type can be determined as a triggering method. For example, it may be determined to send a hashtag message only to a specific number of customer segments having a high probability of clicking the access link of the hashtag message or when the probability of clicking the access link of the hashtag message is greater than a predetermined probability.

The triggering method determination module 302 may predict a click probability using a click probability prediction model pre-learned from information about past hash tag message transmission details of customers of each customer segment, based on customer behavior details. Here, the information on the hashtag message transmission history includes information on the number of times the conventional hashtag message has been sent to the customer, information on the number of clicks on the customer's connection link to the sent hashtag message, and/or clicks on the customer's hashtag message probabilities may be included. The click probability of the customer's hashtag message can be used to train a click probability model as label data of training data. On the other hand, the click probability for the hashtag message may be a click probability per period, a maximum click probability within a specific period, a click probability during a preset period, etc. according to the calculation method of the triggering method determination module 302, but is not limited thereto. .

The click probability prediction model may be a model that performs ensemble learning. The click probability prediction model can perform ensemble learning based on a decision tree. regression analysis model), a Bayesian network, or a probability model.

The triggering information transmission module 304 transmits the triggering information to the advertisement service provider 160 so that the advertisement service provider 160 generates and sends each advertisement message according to the triggering information. The triggering information transmission module 304 may transmit the triggering information of the corresponding customer segment to the service provider 160 at the time when the advertisement message sending time is determined as a triggering method. The triggering information transmission module 304 may transmit triggering information of a corresponding customer segment to the service providing unit 160 for each corresponding period, when a sending period of an advertisement message is determined as a triggering method.

The reporting module 306 generates reporting information about the triggering information. Reporting information on triggering information includes the content of generated or updated triggering information and/or details of customer behavior collected after sending an advertising message according to the triggering information (e.g., average score of click probability, expected conversion rate, product purchase history information, etc.), and the like. The reporting module 306 may transmit reporting information to a preset electronic address or a preset terminal. The reporting module 306 may perform statistical analysis on triggering information for each customer segment and customer behavior history after sending an advertising message, and transmit the analysis result to a preset electronic address or a preset terminal.

After the reporting module 306 transmits the triggering information, the message-based advertising platform 10 transmits target information, data stored in the data lake 100, and log data acquired by the log collection unit 180 from the advertising target platform. Based on all or part of the information analysis to provide a customer-customized advertisement message may be performed.

4 is a flowchart illustrating a method for determining a customer triggering method according to an embodiment of the present disclosure.

The triggering information management unit 130 generates attraction point information and customer segment information based on the customer's behavior details corresponding to the advertisement message and/or the customer's behavior details on the advertisement target platform (S400). The triggering information management unit 130 may determine an attraction time point, which is a time point for sending an advertisement message based on customer behavior information, and generate attraction time information. The triggering information management unit 130 may generate customer segment information by clustering all or part of customer behavior details as features. Each of the generated attraction point information and customer segment information may be updated as customer behavior information is accumulated.

The service target management unit 140 determines a triggering method for each customer segment using the attraction point information and customer segment information generated in step S300 and generates triggering information (S402). Such a triggering method may include, for example, all or part of the type of advertisement message to be sent, the combination of advertisement messages to be sent, the content of the advertisement message, the timing of sending the advertisement message, and the sending period of the advertisement message.

The message-based advertising platform transmits information on a target to send a hash tag message and/or information on a target to send a promotion message to the service target management unit 140 (S404).

The message-based advertising platform generates reporting information about triggering information and transmits it to a specific electronic address or terminal (S406).

In FIG. 4, it is described that each process is sequentially executed, but this is merely an example of the technical idea of an embodiment of the present disclosure. In other words, those skilled in the art to which an embodiment of the present disclosure pertains may change and execute the sequence described in FIG. 4 or perform one or more of each process without departing from the essential characteristics of the embodiment of the present disclosure. Since it can be applied by various modifications and variations by executing in parallel, it is not limited to the time-sequential order of FIG. 3 .

Various implementations of devices, units, processes, steps, etc. described herein may include digital electronic circuits, integrated circuits, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), computer hardware, firmware, software, and/or combinations thereof. These various implementations may include being implemented as one or more computer programs executable on a programmable system. A programmable system includes at least one programmable processor (which may be a special purpose processor) coupled to receive data and instructions from and transmit data and instructions to a storage system, at least one input device, and at least one output device. or may be a general-purpose processor). Computer programs (also known as programs, software, software applications or code) contain instructions for a programmable processor and are stored on a “computer readable medium”.

A computer-readable recording medium includes all kinds of recording devices that store data that can be read by a computer system. These computer-readable 　recording media include non-volatile or non-transitory storage devices such as ROM, CD-ROM, magnetic tape, floppy disk, memory card, hard disk, magneto-optical disk, and storage device. It may further include a medium or a transitory medium such as a data transmission medium. In addition, the computer-readable recording medium may be distributed in computer systems connected through a network, and computer-readable codes may be stored and executed in a distributed manner.

Various implementations of the systems and techniques described herein may be implemented by a programmable computer. Here, the computer includes a programmable processor, a data storage system (including volatile memory, non-volatile memory, or other types of storage systems, or combinations thereof) and at least one communication interface. For example, a programmable computer may be one of a server, network device, set top box, embedded device, computer expansion module, personal computer, laptop, personal data assistant (PDA), cloud computing system, or mobile device.

The above description is merely an example of the technical idea of the present embodiment, and various modifications and variations can be made to those skilled in the art without departing from the essential characteristics of the present embodiment. Therefore, the present embodiments are not intended to limit the technical idea of the present embodiment, but to explain, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The scope of protection of this embodiment should be construed according to the claims below, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of rights of this embodiment.

10: Message-based advertising platform

Claims (9)

A method for triggering a message-based advertising platform to send an advertising message to customers in a customer group,
A process of collecting customer behavior details corresponding to advertising messages and/or customer behavior details on advertisement target platforms as customer behavior information;
generating attraction point information by determining an attraction time point, which is a time point for sending an advertisement message, based on the customer behavior information;
generating customer segment information, which is information for clustering customers based on the customer behavior information; and
Determining a triggering method for each customer segment using the attraction point information and the customer segment information
Characterized in that it comprises, a triggering method determining method. According to claim 1,
The process of generating the attraction point information,
A method for determining a triggering method, characterized in that by extracting a customer's behavior that is a criterion at the time of attraction from the customer behavior information, and calculating a customer behavior type as the customer behavior information. According to claim 2,
The process of generating the attraction point information,
Based on the customer behavior type, a purchase time, which is a time required for customers of the same customer behavior type to purchase a product, is further calculated as the customer behavior information. According to claim 1,
The process of generating the customer segment information,
The method for determining the triggering method, characterized in that the customer segment information is generated by performing clustering using all or part of the customer behavior history as a feature. According to claim 4,
The process of generating the customer segment information,
A method for determining a triggering method, characterized in that all or part of clusters calculated by clustering are calculated as the customer segments, and the customer segment information is generated by designating a feature of a cluster corresponding to each customer segment as reference information. According to claim 5,
The process of generating the customer segment information,
As the customer behavior history, product-related page access information, advertisement target platform access information, and all or part of customer purchase information are set as the feature, and the number of exits after inflow of product-related pages, the number of visits to advertisement target platforms, the number of product purchases, and A method for determining a triggering method, characterized in that all or part of product purchase performance is designated as the reference information. According to claim 1,
The process of determining the triggering method,
All or part of the type of advertisement message to be sent for each customer segment, the combination of advertisement messages to be sent, the content of the advertisement message, the timing of sending the advertisement message, and the sending period of the advertisement message are described by using the information on the attraction point and the customer segment information. Characterized in that determining as a triggering method, a triggering method determining method. According to claim 1,
The process of determining the triggering method,
Based on the customer behavior details, predicting the probability of clicking the access link of the advertising message of each customer segment, and determining the type of advertising message to be sent for each customer segment based on the predicted click probability as a triggering method. A method for determining a triggering method. A computer program stored in one or more computer-readable non-transitory recording media to execute each process included in the method for determining a triggering method according to any one of claims 1 to 8.

Images