JP2005078497A - Server device and advertisement distribution method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To distribute an advertisement in accordance with a customer and his or her real environments regarding advertisement distribution which takes into account the customer's real environments or the like of a distributed party of the advertisement. <P>SOLUTION: In distributing the advertisement to a portable terminal of the customer via a network, with respect to 1 record of a distribution candidate customer file which records position information received from the customer's portable terminal, a shop information file is referred to (S310), a distributer shop candidate is extracted (S312-326), and a shop of advertisement to be distributed to the portable terminal corresponding to 1 record of the distribution candidate customer file is decided (S332 and S334) by referring to a customer information file and the shop information file every shop recorded in the distributer shop candidate file. Thus, in distributing the advertisement, personal information such as the real environments surrounding the customer who is going to receive the advertisement (a location, the weather, time, and so on) and the customer tastes is used, not in isolation but in combination, and the advertisement more useful to each customer can desirably be distributed. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a server device and an advertisement distribution method, and for example, relates to an advertisement distribution that takes into account the actual environment of a customer who is an advertisement distribution destination.

  The Internet is a means of being interactive and capable of exchanging information for individuals. Nonetheless, advertisements using homepages and e-mail on the Internet are mostly mass-targeted advertisements for unspecified people, and are annoying for customers who are not interested. For this reason, not only the advertising effectiveness and the contract rate are low, but conversely, a bad impression on the information provider may be planted.

  There are various reasons why consumers as individuals, not the general public, may be interested in advertising, or willing to purchase goods after seeing the advertisement. For example, an individual's hobbies and preferences are one factor that greatly influences decision making. In addition, external factors such as the actual environment (current location, weather, time, etc.) surrounding an individual when receiving advertisements are factors that indirectly influence decision making. These are used independently in the advertisement distribution support system as factors that influence decision making. However, for example, even if information on snacks is sent from an interested food manufacturer during work, the probability of being connected to purchase is low. Also, even if information about stores that are not of interest is sent just because you are in that location, the advertising effect will not increase. In other words, it can be said that the above-mentioned factors related to purchasing do not produce a great effect even if they are used independently.

  However, when an optimum advertisement is to be distributed in consideration of a plurality of factors such as individual hobbies / preferences and the actual environment surrounding the individual, a combination of a plurality of conditions must be handled and the processing becomes complicated. Furthermore, when other factors are added, it takes man-hours and time to cope with the change.

  There has been proposed a system that performs immediate direct marketing on a mobile phone monitor using the location information of the mobile phone and the e-mail function of the mobile phone (for example, page 4-7 of Patent Document 1). Figure 1-3). However, every time an e-mail is sent, there is a problem that a person must specify a condition from the screen to extract a monitor and create a mail text.

  Further, a product advertisement distribution method has been proposed in which a mobile phone customer can automatically obtain a desired product advertisement in conjunction with an area where the customer is moving (for example, Patent Document 2-3). (See page -5, Fig. 1-4). This method also needs to handle a combination of a plurality of conditions as described above, and there are problems that the process becomes complicated and that it takes man-hours and time to cope with the change.

JP 2002-366464 A Japanese Patent Laid-Open No. 2002-259821

  It is desirable to use personal information such as the actual environment (location, weather, time, etc.) surrounding the customer receiving the advertisement and the customer's hobbies / preferences in combination, not alone, and deliver more useful advertisements to the customer.

  In addition, it is desirable to be able to effectively advertise products that are desired to be preferentially advertised depending on the inventory status and campaign contents.

  The present invention solves the above-mentioned problems individually or collectively, and an object thereof is to distribute advertisements according to customers and their actual environments.

  Another object is to enable a store to effectively advertise a product that it wants to advertise preferentially.

  The present invention has the following configuration as one means for achieving the above object.

  The present invention records position information received from a customer's mobile terminal when distributing an advertisement to the customer's mobile terminal via a network in a distribution candidate customer file stored in a database, and For the record, extract at least the store information file stored in the database and extract the distribution source store candidate, record the extraction result in the distribution source store candidate file stored in the database, and store the distribution source store candidate file. The store of the advertisement to be distributed to the mobile terminal corresponding to one record of the distribution candidate customer file is determined with reference to the customer information file and the store information file stored in the database for each store recorded in And

  ADVANTAGE OF THE INVENTION According to this invention, the advertisement according to a customer and the real environment can be delivered.

  Further, it is possible to effectively advertise a product that the store wants to advertise preferentially.

  The advertisement distribution system and method according to the embodiments described below are based on the effects of one-to-one marketing based on purchase internal factors such as customer hobbies and preferences, and external purchases such as location, weather, and time surrounding customers. The effect of dynamic marketing based on factors is correlated with each other to improve the rate at which customers are interested in advertising information. Further, it is intended to relieve the customer from annoying and troublesome advertising, and receive only advertising information that is more useful to the customer. In addition, for advertisers, to increase the consumer's response rate to advertisements and to carry out effective advertisements, at the same time, to avoid the risk of injuring consumers with bad advertisements It is.

  Hereinafter, an advertisement distribution support system and method according to an embodiment of the present invention will be described in detail with reference to the drawings.

[Constitution]
FIG. 1 is a block diagram illustrating a configuration example of an advertisement distribution support system.

  The advertisement distribution support system 1 includes an advertisement distribution support server 2, a database 3 held in an external storage device, a terminal 4 for operating the advertisement distribution server 2, and the like.

  The advertisement distribution support server 2 communicates with the store terminal 6 and the mobile terminal 7 via the network 5 such as the Internet or a mobile phone network, and also provides a server of a mobile terminal carrier not shown and a server for providing weather information. Communicate with various servers. The terminal 4 and the store terminal 6 are personal computers, and the mobile terminal 7 is a mobile phone, PHS (Personal Handy phone System), PDA (Personal Digital Assistants), or the like. Although FIG. 1 shows a configuration example in which the advertisement distribution support server 2 is a single unit, a plurality of advertisement distribution support servers 2 that share the database 3 are prepared and processed by dividing customers and stores described later. You can also. In addition, the non-volatile memory 2a such as a hard disk of the advertisement distribution support server 2 stores a program for the advertisement distribution support server 2 to execute processing to be described later.

  The database 3 holds a customer information file, a store information file, an advertisement distribution history file, a product category table, a weather table, a coefficient table, and the like that are created and managed by the advertisement distribution support server 2.

Product Category Table The product category table is a table for matching the preference and interest level of a customer's products with the store's product sales willingness. Information input from the terminal (for example, terminal 4) of the administrator of the advertisement distribution support server 2 and the advertisement distribution support system 1 is stored in the database 3 as table data of the product category table.

  FIG. 2 is a diagram showing an example of the product category table, and shows an example of fast food.

  The product categories are arranged in the horizontal direction, and conditions such as time zone, weather, and temperature are taken in the vertical direction, and have numerical values representing the ratio of vertical and horizontal mutual matching. For example, FIG. 2 shows that the rate of coincidence between the product category “hamburger” and the time zone condition “6:00 to 9:00” is “0.5”. Note that the advertisement distribution support server 2 stores, in the database 3, table data that is adjusted so that the total value of ratios for each category is “1” for each condition.

● Weather table The advertisement distribution support server 2 obtains weather information from an external weather information providing server via the network 5 at a predetermined time or at a preset time, and stores it in the database 3 according to the latest information obtained. Updated weather table. FIG. 3 is a diagram showing an example of a weather table, which has predicted values such as weather (clear, cloudy, rain, etc.), temperature, and humidity for each region and each time zone (in units of several hours).

Customer Information File The advertisement distribution support server 2 creates a customer information file based on registration information or questionnaire information input by the customer from the mobile terminal 7 and stores it in the database 3. Fig. 4 shows an example of a customer information file. A set of customer identification number, name, address, mobile phone number, e-mail address, gender, age, and other personal attribute information, and the name of the preference item and its preference ratio. Has a plurality of luxury goods sets.

  “Favorite product name” is selected from the product names registered in the product category of the product category table shown in FIG. 2, and is collected and aggregated by input from the customer terminal 7 or other sales performance collection system. Based on the above, the advertisement distribution support server 2 sets. The “preference ratio” is a numerical value indicating the ratio of preference for the preference product, and is set by the advertisement distribution support server 2 based on the data collected and aggregated by the customer questionnaire and other sales performance collection systems. For example, FIG. 4 shows that a customer whose customer identification number is “abc1234567” selects “coffee” as a preference item name and the preference ratio is “23” percent. The advertisement distribution support server 2 adjusts the total preference ratio to 100% for each customer when creating or changing the customer information file. For example, in the customer questionnaire data, if the luxury items are ordered from the first to the fourth, the advertisement distribution support server 2 sets the preference item ratio of these preference items to 40, 30, 20, 10%, etc., respectively. .

● Store information file The advertisement distribution support server 2 creates a store information file based on information entered by operating the store terminal 6 by a store manager or a person in charge (hereinafter sometimes referred to simply as a “store”). And save it in database 3. FIG. 5 is a diagram showing an example of a store information file. Store attribute information such as store identification number, store name, address, telephone number, store location, weather region, and product category and recommended ratios With a recommended product set.

“Store position” is position information obtained from GPS (Global Positioning System) and information of a communication carrier base station, and indicates the position in the following format.
Addd.mm.ss.xx
Bddd.mm.ss.xx
Where A is latitude and N (north latitude) or S (south latitude)
B represents longitude, E (east longitude) or W (west longitude)
ddd.mm.ss.xx is a numerical value that represents `` degree.minute.second.percent of a second ''

“Business hours” indicates the business hours of the store in the following format.
Wwhhmm.hhmm
Where Ww represents the day of the week, Su (Sun) Mo (Mon) Tu (Tue) We (Wed) Th (Thu)
One of Fr (Friday) Sa (Saturday)
hhmm.hhmm represents “opening time. closing time”.
When the store is open at 7:30 pm, it is “0900.1930”. Also,
The day of the week when the store is closed is expressed as “Ww0000.0000”. In addition, the year
If the store is closed during the New Year holidays, etc.
"Ww0000.0000" should be used

  The “weather area” indicates which area of the weather table is used as the weather of the store. “Product category” indicates a product selected by the store as a recommended product from among the product categories registered in the product category table. “Recommended ratio” indicates the degree to which the recommended product of the store should be advertised to the customer. The advertisement distribution support server 2 adjusts the recommended ratio so that the total recommended ratio for each store becomes 100.

  Of course, the store can operate the store terminal 6 connected to the website provided by the network 5 and the advertisement distribution support system 1 to change the information on the recommended product at any time, and the advertisement distribution that accepts the change information The support server 2 updates the store information file stored in the database 3 to reflect the change information.

● Advertisement distribution history file When the advertisement distribution support server 2 performs the advertisement distribution described later, the history is created as an advertisement distribution history file and stored in the database 3. FIG. 6 is a diagram showing an example of the advertisement distribution history file. For each customer identification number indicating an advertisement distribution destination, the distribution date and time of the advertisement, the store identification number of the distribution source store, the distribution destination position (of the mobile terminal 7 at the time of distribution) Position). The latest distribution information for the customer identification number is stored in the “most recent distribution record”, and the previous distribution for the customer identification number is stored in the “distribution history n” (n = 1, 2, 3,…). Stack information (gets older as n increases).

Coefficient table The advertisement distribution support server 2 accepts input from the terminal 4 of the administrator of the advertisement distribution support system 1 for some judgment values among the information stored in the database 3 and reflects them in the coefficient table. The processing content can be adjusted. FIG. 7 is a diagram showing an example of the coefficient table. Although details will be described later, time intervals and temperature information (items for determining whether the temperature is hot, warm, cool, or cold are used as items for the advertisement delivery standard) Value), a parameter value used to calculate a store score, which will be described later, and the like.

[processing]
The advertisement distribution support server 2 performs a position information inquiry process, a position information reception process, an advertisement distribution process, and the like, which will be described in order below.

Position Information Inquiry Process The advertisement distribution support server 2 activates and executes position information inquiry processing at the time interval indicated by the “activation interval” for the period indicated by the “operation time zone” in the coefficient table. For example, according to the coefficient table shown in FIG. 7, the position information inquiry process is activated and executed every 0.7 hours (42 minutes) from 7:00 to 23:00. FIG. 8 is a flowchart showing an example of the position information inquiry process.

  The advertisement delivery support server 2 reads the data of one customer registered in the customer information file from the customer information file and the advertisement delivery history file (S110), and compares the current time with the delivery date and time of the latest delivery record of the customer. Then, it is determined whether or not a time equal to or longer than the “reference delivery interval” (three hours in the example of FIG. 7) of the coefficient table has elapsed (S112). If it has elapsed, the customer is stored in a predetermined area of the RAM (not shown) as a customer for inquiring position information (S114). If it has not elapsed, the process of step S114 is skipped, and it is determined whether or not the determination of step S112 has been completed for all customers registered with the customer information file (S116). If not completed, the process returns to step S110 to perform the determination in step S112 for the next customer in which the customer information file is registered. In this way, when the determination in step S112 is completed for all customers (registered all customer information files), the process proceeds to step S118.

  As described above, by determining the elapsed time in step S112, the advertisement is not continuously delivered to the same customer within the time set as the “reference delivery interval”, and the customer is prevented from feeling uncomfortable.

  The advertisement delivery support server 2 requests the mobile operator of the mobile terminal 7 to send a short message for inquiring the location information to the mobile terminal 7 for the customer stored as the customer inquiring the location information ( S118). This request is made for each number of people set in the “position request transmission unit” of the coefficient table. In step S114, the request in step S118 may be made each time the number of customers stored as customers inquiring position information reaches the set value of “position request transmission unit”.

  On the other hand, the customer can download the location information application program from the website provided by the advertisement distribution support system 1 to his / her mobile terminal 7. At that time, it is necessary to input the customer identification number, and the customer identification number is recorded in the downloaded location information application program. In the following description, it is assumed that the customer has already downloaded the location information application program.

  When the customer's mobile terminal 7 receives a short message inquiring location information from the carrier of the mobile terminal 7, the customer's mobile terminal 7 starts the location information application program (in this case, the startup type is “push”). . The location information application program is activated not only by the (automatic) activation associated with the reception of the short message, but also by the customer operating the mobile terminal 7 when the customer desires the advertisement distribution (in this case, activation) The type is “pull”). The location information application program communicates with the GPS, base station, and telecommunications carrier to obtain its current location information, even if the processing contents differ depending on the type of mobile terminal 7 and the telecommunications carrier of the mobile terminal 7. After acquisition, information such as the customer identification number, the acquired position information, its error, and the activation type is transmitted to the advertisement distribution support server 2 as a response to the inquiry about the position information by e-mail or HTTP protocol.

Position Information Reception Processing Next, position information reception processing performed by the advertisement distribution support server 2 will be described. Note that the position information receiving process continues to operate without stopping. FIG. 9 is a flowchart showing an example of position information reception processing.

  When the advertisement distribution support server 2 receives information such as a customer identification number, position information, its error, and activation type from the position information application program of the mobile terminal 7 (S120), the distribution candidate stored in the database 3 is received information A record is recorded at the end of the customer file (S122), and the process returns to step S120.

  FIG. 10 shows an example of a distribution candidate customer file, which has a customer identification number, request date and time, position information, error level, distribution type, and the like.

  “Position information” is position information received from the mobile terminal 7 and has the same format as the store position.

  The “error level” is an average value of errors in the position information received from the mobile terminal 7, and error information that differs for each carrier is unified for this system. For example, when an error in position information provided by a certain communication carrier is expressed in a distance range, the intermediate point is set to 25 m if the error range is 0 to 50 m. Also, if the error of location information provided by other carriers is given by error distance and accuracy, (error distance / accuracy) / 2 value, 65% accuracy within 30m error {(30+ 0) /0.65} / 2 = 23m.

  “Distribution type” indicates the activation type of the location information application program received from the mobile terminal 7, “push” when the location information application program is automatically activated upon receipt of a short message, and when activated by the customer's operation Becomes “pull”.

● Advertisement delivery process Next, the advertisement delivery process performed by the advertisement delivery support server 2 will be described. The advertisement distribution process continues to operate without stopping. FIG. 11 is a flowchart illustrating an example of the advertisement distribution process.

  The advertisement distribution support server 2 determines whether there is a record in the distribution candidate customer file (S210). If there is no record, the time set in the “waiting time” of the coefficient table (0.001 hour in the example of FIG. 7). (3.6 seconds)) is paused (S212), and then the process returns to step S210.

  If there is a record in the distribution candidate customer file, one record is read from the beginning of the distribution candidate customer file (S214), and based on the customer identification number of the read record, the corresponding customer's information is read from the customer information file and the advertisement distribution history file. Data is read (S216).

  Next, the advertisement distribution support server 2 determines whether or not the distribution type value of the read record is “push” (S218), and if so (hereinafter referred to as “push-type customer”), the read record The error level is compared with the set value of “allowable error” in the coefficient table (S220), and if the error level <the allowable error, the process for the push-type customer is performed (S222). If error level ≧ allowable error, the process proceeds to step S226. In other words, for mobile terminals 7 with large positional information errors, the distance between the mobile terminal 7 and the store cannot be accurately grasped, and the distance between the two may be far, so there is a possibility that the advertisement is less effective without distributing the advertisement. Avoid the possibility of delivery.

  If the value of the delivery type is “pull” (hereinafter referred to as “pull type customer”), processing for the pull type customer is performed (S224), and then the process proceeds to step S226. For pull-type customers, processing for the pull-type customers is performed without examining the error level.

  In step S226, the top record in the distribution candidate customer file, that is, the processed record is deleted, and then the process returns to step S210.

Processing for Push Type Customer FIG. 12 is a flowchart for explaining the flow of processing for a push type customer.

  The advertisement distribution support server 2 reads one record of the store information file (S310), first determines from the business hours data whether or not the store is in operation (S312), and if it is not in operation, the processing step Proceed to S328.

If the store is in operation, the distance between the store and the mobile terminal 7 is calculated from the “store location” in the store information file and the “position information” in the distribution candidate customer file (S314). For example, when calculating using the Huveni distance calculation formula, find the average value of latitude at both locations as P, the difference in latitude as dP, the difference in longitude as dR, and then calculate the meridian radius of curvature M and 卯The shoreline radius of curvature N is calculated by the following formula.
M = 6334834 / √ (1-0.006674 ・ sin ² P) ³
N = 6377397 / √ (1-0.006674 ・ sin ² P)

Then, the distance D between them is calculated by the following equation.
D = √ {(M ・ dP) ² + (N ・ cos P ・ dR) ² }

  For example, if the store position is 35 degrees 32 minutes 57.53 seconds north, 139 degrees 31 minutes 22.82 seconds east, the position of the mobile device is 35 degrees 32 minutes 56.00 seconds north, 139 degrees 31 minutes 17.00 seconds east, the distance D between them is Calculated as 154 meters by the formula. The advertisement distribution support server 2 compares the calculated distance D with the “store distance” in the coefficient table (S316). If the distance D <store distance, the process proceeds to step S318. If the distance D ≧ store distance, the process is performed. Proceed to step S328.

  Next, the advertisement distribution support server 2 refers to the advertisement distribution history file to determine whether or not the store is registered in the latest distribution results of the target customer (S318). It is checked whether or not a time equal to or greater than the “same store continuous delivery interval” in the coefficient file has elapsed since the delivery date (S320). Also, if the store is not registered in the most recent distribution record, the distribution history is referred to check whether the store is registered in the distribution history n (S322). It is checked whether or not a time equal to or greater than “same store shortest delivery interval” in the coefficient table has elapsed (S324).

  For example, as shown in FIG. 6, for a customer with a customer identification number “abc1234567”, if the store with the store identification number “sh11223” is registered in the latest delivery record, even if the store and the mobile phone 7 (target customer) Even if the distance is short, it is determined that the distribution condition is not satisfied until the same store continuous distribution interval (“125 hours” in FIG. 7) elapses from the distribution date and time. In addition, when the store with the store identification number “sh11223” is registered in the distribution history n, the distribution condition is satisfied until the same customer shortest distribution interval (“27 hours” in FIG. 7) elapses from the distribution date and time. Judge that there is no. Of course, if it is not registered in the distribution history n, it is determined that the condition is satisfied. This determination avoids discomforting the customer by repeatedly delivering advertisements from the same store in a short period of time.

  Then, the advertisement distribution support server 2 proceeds to step S326 if the distribution condition is satisfied, and proceeds to step S328 if the distribution condition is not satisfied.

  Next, the advertisement distribution support server 2 sets the store identification number and the distance D of the store closer to the distance D as a set for the number set in the “number of stores narrowed down” in the coefficient table (“10” in FIG. 7). Then, store in the distribution source store candidate file stored in the database 3 (S326), determine whether all records of the store information file have been referenced (S328), and if there is an unreferenced record, proceed to step S310 If there is no unreferenced record, the process proceeds to step S330.

When all the records of the store information file are referenced, the advertisement distribution support server 2 determines whether or not there is a distribution source store candidate (S330), and if there is, stores the store score S for each stored distribution source store candidate. (S332). If there is no distribution source store candidate, the process for the push-type customer is terminated.
S = Σ (Wm ・ Pm)

  Here, Wm is a weight, and its value is registered in the coefficient table. For example, FIG. 7 shows distance weights, preference weights, time weather weights, and distribution performance weights. Pm is a point calculated by the advertisement distribution support server 2, and a calculation example is shown below.

(1) The distance point Pd is calculated by the following formula. In the following equation, L is the set value of “store distance” in the coefficient table (“500 meters” in the example of FIG. 7), and D is the distance between the store and the mobile terminal 7. That is, the distance point Pd indicates the degree of proximity between the store and the customer.
Pd = {(L-D) / L} ・ 100

(2) The preference point Pf is calculated by the following formula. In the following formula, the value in () on the right side is calculated for each product category, and Cn is the preference ratio if the product category corresponding to the preference item name in the customer information file of the customer to be processed is registered, not registered If it is, it is “0”. Kn is the recommended ratio if the product category corresponding to the product category of the store information file is registered, and “0” if not registered. In other words, the product category registered in the name of the preference item in the customer information file and the product category registered in the recommended product in the store information file, that is, the one in which the customer's preference product matches the recommended product in the store is calculated. The total is defined as a preference point Pf.
Pf = Σ (Cn ・ Kn)

For example, when comparing a preference product with customer identification number “abc1234567” (see FIG. 4) and a recommended product with store identification number “sh11223” (see FIG. 5), “general dessert” matches, and the others do not match The preference point Pf is as follows.
pf = Σ (Cn ・ Kn) = 18 ・ 20 = 360

(3) Calculate the hourly weather point Pt using the following formula. In the following equation, the values in parentheses on the right side are calculated for each product category, and Kn is the recommended ratio of the product category registered in the store information file, like the preference point pf described above.
Pt = Σ (Bn ・ Kn)
Where Bn = G ・ g + H ・ h + J ・ j

  In the above equation, Bn is calculated based on the current time and the current weather / temperature indicated by the weather table, the coefficient table, and the product category table at the location corresponding to the location information of the mobile terminal 7. Specifically, G is the set value of the “time coefficient” in the coefficient table (“5” in the example of FIG. 7), H is the set value of the “weather coefficient” (“1” in the example of FIG. 7), J Is the set value of the “temperature coefficient” (“2” in the example of FIG. 7). In addition, g is a set value of “time zone” corresponding to the current time in the product category of the product category table, h is a set value of “weather” corresponding to the current weather, and j is equivalent to the current temperature. This is the setting value of “temperature”.

For example, for the recommended product “dessert cold” (see FIG. 5) with the store identification symbol “sh11267”, if the current time is 16:00, then the product category table time zone “13:00 to 17:00” shown in FIG. Get = 0.6.
G ・ g = 5 ・ 0.6 = 3

The weather “cloud” (see FIG. 3) from 15:00 to 18:00 of “XX city” is acquired by referring to the weather table based on the weather area “XX city” (see FIG. 5) of the store. However, since the product category “dessert cold” in the product category table does not have a value corresponding to the weather “cloudy”, h = 0 is set.
H ・ h = 1 ・ 0 = 0

Similarly, referring to the weather table, the temperature “18” (see FIG. 3) of the weather region “XX city” from 15:00 to 18:00 is acquired. 18 degrees corresponds to “warm” when referring to the coefficient table, and the value corresponding to the temperature “warm” of the product category “dessert cold” in the product category table from the recommended product “dessert cold” of the store (see FIG. 5). To j = 2.
J ・ j = 2 ・ 0.2 = 0.4

Therefore, Bn of the product category “dessert cold” in the city of XX at the current time is calculated as follows, and the preference point Bn · Kn of “dessert cold” of the store is as follows from the recommended ratio “80”: Is calculated.
Bn = G ・ g + H ・ h + J ・ j = 3 + 0 + 0.4 = 3.4
Bn ・ Kn = 3.4 ・ 80 = 272

(4) The distribution result point Pr is calculated by the following formula. In the following equation, R is the number of times the advertisement is distributed within the set value of “delivery result count time” (“24 hours” in the example of FIG. 7) of the coefficient table from the current time. The number of distributions for each store within the distribution result count time can be checked with reference to the advertisement distribution history file. Moreover, although the distribution result point Pr becomes a negative value, this is intended to correct the disparity in the number of advertisement distributions between stores with good conditions such as those located in downtown areas and stores that are not.
Pr = -R

  In this way, the advertisement distribution support server 2 calculates the store score S for each store of distribution source store candidates, and then determines the store having the largest store score S as the distribution source store, and an example is shown in FIG. The contents of delivery are determined with reference to the mail draft text file shown (S334). Specifically, based on the distribution source store and the current time in the mail draft file, the content of the e-mail that distributes the content of the corresponding store identification number and the mail content of the distribution time zone to the mobile terminal 7 of the target customer To. The store can change the delivery time zone and the number and contents of the mail contents by operating the store terminal 6 at any time.

  The advertisement distribution support server 2 transmits the email of the distribution content determined in step S334 to the mobile terminal 7 of the target customer, updates the latest distribution result of the customer's advertisement distribution history file (S336), and The process ends. Note that the data stored as the latest distribution record immediately before is stacked in the distribution history n. At this time, the distribution history n for which the time equal to or greater than the set value of “distribution history storage time” in the coefficient table has been deleted.

Processing for Pull Type Customer FIG. 14 is a flowchart for explaining the flow of processing for the pull type customer. In FIG. 14, the same processing contents as those in FIG. 12 are denoted by the same reference numerals, and detailed description thereof is omitted. In the following, description will be made centering on portions where the processing differs between FIG. 12 and FIG.

  Since the pull-type customer is a customer who voluntarily requests advertisement distribution, the advertisement may be immediately distributed regardless of the distribution history. For this reason, the advertisement distribution support server 2 does not perform the determination of the distance D (S316) and the determination of the passage of time after distribution (S318-S324) shown in FIG. If there is no distribution source store candidate in step S330, the store identification number is left blank and the process proceeds to step S366. The mail content in which the store identification number in the mail draft file is blank, in the example of FIG. Is sent to notify that the store where the advertisement should be distributed cannot be found.

  Thus, according to the first embodiment, when receiving an advertisement, the actual environment (position, weather, time, etc.) surrounding the customer, personal information such as the customer's hobbies / preferences, and the location of the store of the advertisement source The product category, the recommendation level, and the like are parameterized, the score is calculated based on the parameters when the advertisement is distributed, and the advertisement is distributed according to the obtained score (representing the matching degree between the customer and the store). Accordingly, it is possible to provide an advertisement that the customer would like more. In addition, by distributing advertisements in consideration of advertisement distribution history, an advertisement distribution support system that can avoid discomfort such as dripping advertisements and unwanted advertisements as much as possible to customers who have received advertisement distribution can do.

[Modification 1]
In the above embodiment, a fast food restaurant has been described as an example of a store. However, the following modifications may be applied depending on the type of business to be applied.

  (1) When applied to a store in the travel industry, the scale of the reference distribution interval, the same store shortest distribution interval, and the same store continuous distribution interval in the coefficient table is changed from time to number of days.

  (2) When applied to a department store or a supermarket that handles a large number of business categories, it is preferable to register in a store information file for each sales floor.

  (3) In the case of chain stores of limited items (coffee, ice cream, hamburgers, etc.), it is preferable to register individual product names such as fish burgers and iced coffee instead of product categories.

  (4) When applying to a store in the cosmetics industry, it is preferable to adopt “skin condition” or the like as the condition of the product category table. For example, if the following conditions are set: bulkiness = low temperature / humidity, cassava = medium temperature / low humidity, moist = medium temperature / humidity, stickiness = high temperature / humidity, etc., the correspondence with weather conditions can be taken finely.

[Modification 2]
In the above description, the weather table, the distribution candidate customer file, and the advertisement distribution history file are constantly updated to the latest state, but can be stored for a certain period. Then, referring to those files, let the advertisement distribution support server 2 or other servers execute statistical processing at times other than the operating hours, and display the advertisement distribution results by time zone and store. Can be displayed on a graph.

  Also, according to the instructions from the administrator's terminal 4, change the setting value of the coefficient table and the value of the product category table, and distribute based on the temporary setting time different from the actual time of the advertisement distribution support server 2 If the value of “request date” and “location information” in the candidate customer file is used, the time until the ad delivery source store determination (S334) of the ad delivery process is simulated (no actual ad delivery is performed). It is possible to create a temporary advertisement distribution history file by simulating what the advertisement distribution status for each band and store will be. Based on such data, the advertisement distribution support server 2 performs statistical processing, creates and displays a table and graph of advertisement distribution results by time zone and store, and evaluates data for evaluating the setting values of the appropriate coefficient table Can be prepared.

  The above simulation is effective when there are relatively many additions / deletions of stores, additions / deletions of merchandise categories, additions / deletions of customers, changes in seasons, score calculation formulas, and point calculation formulas. Thereby, it is also possible to connect to the control and leveling of the advertisement distribution amount between stores.

[Modification 3]
In the above, the advertisement distribution support server 2 has been described as having one set of time zone, weather, and temperature conditions in the product category table. However, as shown in FIG. 15, these conditions are set for each recommended product in the store information file. It may be held. Moreover, about the time conditions, it is possible to easily reflect the feature of each store by making it possible to specify a time zone for each store. Even in this case, the advertisement distribution support server 2 adjusts the total value for each condition (time zone, weather, temperature) to be “1” for one category, and the disparity between the stores. Try to reduce.

A block diagram showing a configuration example of an advertisement delivery support system, The figure which shows an example of a goods category table, The figure which shows an example of a weather table, Figure showing an example of customer information file, The figure which shows an example of a store information file, A diagram showing an example of an ad delivery history file, The figure which shows an example of a coefficient table, A flowchart showing an example of the position information inquiry process, A flowchart showing an example of position information reception processing; The figure which shows an example of a delivery candidate customer file, A flowchart showing an example of an advertisement distribution process; A flowchart explaining the flow of processing for a push-type customer; Figure showing an example of an email draft file A flowchart for explaining a flow of processing for a pull-type customer; It is a figure which shows the other example of a shop information file.

Claims (9)

A server device that distributes advertisements to customer mobile terminals via a network,
Receiving means for recording location information received from the customer's mobile terminal in a delivery candidate customer file stored in the database;
For one record of the delivery candidate customer file, the delivery source store candidate is extracted with reference to at least the store information file stored in the database, and the extraction result is stored in the database. Extraction means to record in,
For each store recorded in the distribution source store candidate file, refer to at least the customer information file and the store information file stored in the database, and distribute to the mobile terminal corresponding to one record of the distribution candidate customer file And a determination unit that determines a store of the advertisement to be performed. Further, referring to the advertisement distribution history file stored in the database and the customer information file, request means for requesting the location information of the customer's mobile terminal,
2. The server apparatus according to claim 1, further comprising history update means for updating the advertisement delivery history file based on the advertisement delivery result.   The extraction means refers to the advertisement distribution history file in addition to the store information file when the record of the distribution candidate customer file corresponds to a portable terminal that returned position information in response to a request from the request means. 3. The server device according to claim 2, wherein the distribution source store candidate is extracted.   2. The extraction unit according to claim 1, wherein the extraction unit narrows down the distribution source store candidates based on the location information of the mobile terminal and store sales information or location information recorded in the store information file. 3. The server device described in any one of 3.   2. The server device according to claim 1, wherein the determination unit further determines a store of an advertisement to be distributed with reference to a current time, a product category table stored in the database, and a weather table. .   The determination means includes a distance between the mobile terminal and the distribution source store candidate, customer preference information recorded in the customer information file, and product category information of the distribution source store candidate recorded in the store information file. For each distribution source store candidate, the correspondence and the relationship between the current time and the information recorded in the product category table or the weather table and the product category information of the distribution source store candidate recorded in the store information file 6. The server apparatus according to claim 5, wherein the distribution source store candidate having the maximum score is determined as an advertisement store to be distributed. An ad distribution method for distributing an advertisement to a customer's mobile terminal via a network,
Record the location information received from the customer's mobile device in the distribution candidate customer file stored in the database,
For one record of the delivery candidate customer file, the delivery source store candidate is extracted with reference to at least the store information file stored in the database, and the extraction result is stored in the database. Recorded in
For each store recorded in the distribution source store candidate file, the customer information file stored in the database and the store information file should be referred to be distributed to the mobile terminal corresponding to one record of the distribution candidate customer file. An advertisement distribution method characterized by determining an advertisement store.   8. A program for controlling an information processing apparatus to realize the advertisement distribution process according to claim 7.   9. A recording medium on which the program according to claim 8 is recorded.

Images