JP5033166B2 - Advertisement creation method and advertisement creation system - Google Patents

Description

  The present invention relates to an advertisement creation method and an advertisement creation system for creating a product advertisement such as a flyer that is optimal for use in a retail business, particularly a store system that develops a plurality of sales outlets in different regions.

  As a known advertisement creation system, for example, Patent Document 1 discloses an advertisement creation support system in a head office connected to a store via a network. In this advertising creation support system, in order to produce highly effective advertising, the headquarters creates a flyer plan, and in response to this produced flyer plan, the dealer answers the expected sales volume of the product on the flyer. Depending on the expected sales volume, advertisements will be produced reflecting the posted size, position, color, border thickness, character size, etc. of each product.

  Further, for example, Patent Document 2 discloses a sales information processing system that is not an advertisement creation system, but displays so that the relationship between the product posted in the advertisement and the sales information of the product can be visually grasped. .

JP 2002-123666 A JP 2000-057209 A

  According to the advertisement creation support system disclosed in Patent Literature 1, an advertisement is produced by determining the product size and the like based on the sales expected amount predicted by the dealer. However, this system does not produce advertisements based on sales performance, but creates advertisements based on the estimated sales expected by each dealer, resulting in lack of objectivity and reliability. There is a risk of advertisement production. In addition, each dealer is forced to predict the expected sales amount each time, which is cumbersome and the speed of advertisement production cannot be obtained. Further, this Patent Document 1 does not disclose anything about how to determine the posting size, position, color, border thickness, character size, etc. of each product according to the sales forecast amount.

  On the other hand, Patent Document 2 only discloses a display related to a product posted in an advertisement and its sales information, and does not describe how to create an advertisement at all.

  Accordingly, it is an object of the present invention to provide an advertisement creation method and an advertisement creation system capable of creating an objective and reliable advertisement reflecting the sales results.

  Another object of the present invention is to provide an advertisement creation method and an advertisement creation system capable of creating an advertisement with a higher advertising effect.

  Still another object of the present invention is to provide an advertisement creation method and an advertisement creation system that can easily and quickly create an advertisement without human intervention.

  According to the present invention, PI values of all products for each time period are stored in the first database, and the PI values of the products posted in the advertisement distributed at the time corresponding to the distribution time of the advertisement to be created. Are read from this first database, the products listed in the distributed advertisement are ranked, the PI values of all the products at this corresponding time are read from the first database, and the read PI values are higher. The products that are listed in the advertisement that should be created by replacing the lower ranking product with the selected product are ranked, and the coefficient value based on the frame actual value is calculated in advance for each frame of the advertisement that should be created And store them in the second database, and rank the products in descending order of the coefficient values based on the actual frame values read from the second database in descending order of ranking. Allocation, ad creation method of ranking the items stored in the third database as an advertisement should create ads allocated to each frame is provided.

  Rank the products listed in the distributed ads by the PI value of the products listed in the ads distributed during the time corresponding to the distribution of the ads to be created (for example, the same week in the previous year) On the other hand, among the products in the corresponding period, the product with the higher PI value at that time is selected, and the products to be published in the advertisement to be created are ranked by replacing the products with the lower ranking with the selected products. Next, using the coefficient value (location correction value) based on the frame actual value calculated in advance for each frame of the advertisement to be created, the ranked products are ranked in descending order of ranking, and the coefficient value based on the frame actual value is high. Each frame of the advertisement to be created is determined by sequentially allocating to each. In this way, since the products are replaced by ranking with the PI values at the corresponding time in the past, it is possible to create an objective and reliable advertisement that reflects the sales performance. Further, since the product is divided into frames using the location correction value calculated in advance for each frame, it is possible to create an advertisement with a higher advertising effect. In addition, since the creation of a new advertisement is automatically performed from past data without human intervention, it is possible to create an advertisement easily and quickly.

  The first database preferably stores weekly PI values for all products.

  It is also preferable that the ranking is to rank products placed in advertisements distributed at a corresponding time by ABC analysis of PI values. In this case, it is more preferable that the lower ranking product is a C rank product.

  The category and the region correction value for each product category and each product sales area are obtained and stored in the fourth database in advance, and the above-mentioned product selection is performed by reading the obtained PI values of all products from the fourth database. It is also preferable that the correction is performed using the category and region correction values, and the corrected PI value is used to select a higher-order product. In this case, category and region correction values are stored in the fourth database for each month, and all the values obtained from the category and region correction values corresponding to the distribution month of the advertisement to be created, read from the fourth database. More preferably, the PI value of the product is corrected.

  The coefficient value based on the frame actual value of the advertisement to be created is preferably a location correction value calculated in advance from the past sales amount of each frame of the advertisement to be created and stored in the second database.

  This location correction value divides one advertisement into cells that are the smallest unit, and calculates the sales amount for each cell from the sales amount of the product corresponding to each frame in one advertisement and the number of cells in the frame. The cumulative value of the sales amount for each cell is calculated for a plurality of advertisements in a predetermined period in the past, and the composition ratio coefficient representing the composition ratio for each cell is calculated. The composition ratio coefficient is added to obtain the frame actual value corresponding to the sales amount of each frame, the frame average value that is the average value of the frame actual value of the advertisement to be created is obtained, and for each frame of the advertisement to be created, More preferably, (location correction value) = (frame actual value) / (frame average value) is calculated.

  It is also preferable to further correct the PI value of the product allocated to each frame of the advertisement to be created with the location correction value. In this case, it is more preferable to calculate the predicted sales amount of each frame and the entire advertisement by converting the PI value corrected by the location correction value of the advertisement to be created into the sales amount.

  According to the present invention, the first database that stores the PI values for all products for each time period, and the PI values of the products that are published in the advertisement distributed at the time corresponding to the distribution time of the advertisement to be created Is read from the first database to rank the products listed in the distributed advertisement, and the PI values of all the products at this corresponding time are read from the first database, and the read PI values are read out from the first database. Means to select the top product, ranking the products to be listed in the advertisement to be created by replacing the ranked product with the selected product, and the pre-calculated each frame of the advertisement to be created A second database that stores coefficient values based on the top frame actual value, and top frame actual values obtained by reading the ranked products from the second database in descending order of ranking And sequentially allocating means high coefficient value frame based, ad creation system comprising a third database ranking the items stored as advertisement should create ads allocated to each frame is provided.

  Rank the products listed in the distributed ads by the PI value of the products listed in the ads distributed during the time corresponding to the distribution of the ads to be created (for example, the same week in the previous year) On the other hand, among the products in the corresponding period, the product with the higher PI value at that time is selected, and the products to be published in the advertisement to be created are ranked by replacing the products with the lower ranking with the selected products. Next, using the coefficient value (location correction value) based on the frame actual value calculated in advance for each frame of the advertisement to be created, the ranked products are ranked in descending order of ranking, and the coefficient value based on the frame actual value is high. Each frame of the advertisement to be created is determined by sequentially allocating to each. In this way, since the products are replaced by ranking with the PI values at the corresponding time in the past, it is possible to create an objective and reliable advertisement that reflects the sales performance. Further, since the product is divided into frames using the location correction value calculated in advance for each frame, it is possible to create an advertisement with a higher advertising effect. In addition, since the creation of a new advertisement is automatically performed from past data without human intervention, it is possible to create an advertisement easily and quickly.

  The first database is preferably storage means for storing weekly PI values for all products.

  The ranking means is preferably a means for ranking products placed in advertisements distributed at corresponding times by ABC analysis of PI values. In this case, it is more preferable that the lower ranking product is a C rank product.

  A fourth database for storing categories and regional correction values for each product category and each product sales region is further provided, and the means for selecting the above-described product stores the PI values of all the obtained products from the fourth database. It is also preferable that the correction is performed by the read category and area correction value, and the corrected PI value is a means for selecting a higher-order product. In this case, the fourth database stores the category and region correction values for each month, and the means for selecting the product reads the category and region corresponding to the distribution month of the advertisement to be created, read from the fourth database. It is more preferable that the PI value of all the products obtained by the correction value is corrected.

  The second database preferably stores a location correction value calculated in advance from the past sales amount of each frame of the advertisement to be created as a coefficient value based on the frame actual value of the advertisement to be created.

  Means for dividing one advertisement into cells that are the smallest unit, means for calculating the sales amount for each cell from the sales amount of the product corresponding to each frame in one advertisement and the number of cells in the frame, For each of a plurality of advertisements in a predetermined period of time, a means for calculating a cumulative ratio of sales amounts for each cell and calculating a composition ratio coefficient representing the composition ratio of the sales amount for each cell, A means for calculating the frame actual value corresponding to the sales amount of each frame by adding the composition ratio coefficient for each cell, and calculating the frame average value which is the average value of the frame actual value of the advertisement to be created, and the advertisement to be created Means for calculating (location correction value) = (frame actual value) / (frame average value), and means for storing the calculated location correction value in the second database. Also Masui.

  It is also preferable to provide means for further correcting the PI value of the product allocated to each frame of the advertisement to be created with the location correction value.

  It is also preferable to further include means for converting the PI value corrected with the location correction value of the advertisement to be created into a sales amount and calculating a predicted sales amount for each frame and the entire advertisement.

  According to the present invention, it is possible to create an advertisement with objectivity and reliability that reflects sales results. It is also possible to create an advertisement with a higher advertising effect. Furthermore, an advertisement can be easily and quickly created without human intervention.

It is a block diagram which illustrates roughly the structure of one Embodiment of the advertisement production system of this invention. FIG. 2 is a flowchart illustrating a schematic flow of advertisement creation processing by a computer in the embodiment of FIG. 1. It is a figure which shows the advertisement itself used as the foundation of the advertisement which should be produced. It is a figure which shows the list which ranked and represented each product published by the advertisement used as a foundation. FIG. 4 is a diagram in which an ABC rank is added to the basic advertisement shown in FIG. 3. It is a figure which shows a part of list which displayed all the goods in the order based on the correct | amended PI value. It is a figure which shows a part of category and area correction value which are memorize | stored in the 4th database. It is a flowchart explaining the schematic flow of a calculation process of a category and an area correction value. It is a figure which shows the list which displayed the goods of A and B rank, and the selected goods. It is a figure which shows the list which ranked the goods of FIG. 9 in the order of PI value. It is a figure which shows the location correction value of each flame | frame of the base advertisement memorize | stored in the 2nd database by frame correspondence. It is a flowchart explaining the schematic flow of a calculation process of a location correction value. It is a figure which shows the state which divided | segmented the advertisement into the cell which is the minimum division | segmentation area virtually. It is a figure explaining the procedure which calculates | requires the sales amount for every cell. FIG. 2 is a diagram illustrating an example of a composition ratio coefficient for each cell obtained in the advertisement of the embodiment of FIG. 1. It is a figure showing the relationship between the composition ratio coefficient for every cell, and each frame of an advertisement. It is a figure showing the frame performance value in each frame of the calculated advertisement to be created. It is a figure showing the location correction value of each calculated | required frame. It is a figure showing the location correction value of each calculated | required frame. It is a figure which shows PI value of each frame correct | amended with the location correction value after allocating goods. It is a figure which shows the estimated sales amount of each frame calculated | required by converting corrected PI value.

  FIG. 1 is a block diagram schematically illustrating a configuration of an embodiment of an advertisement creation system according to the present invention. The present embodiment relates to a system for creating a product advertisement such as a flyer advertisement in a large-scale retail store chain such as a supermarket having a plurality of stores in a plurality of regions.

In FIG. 1, 10 is an advertisement creation processing apparatus that actually performs advertisement creation processing, and 11 is connected to this advertisement creation processing apparatus 10, and the PI values and sales amounts of all products in all regions are classified by past week. The first database 12 that is stored is connected to the advertisement creation processing apparatus 10, and the second database 13 that stores the location correction value is connected to the advertisement creation processing apparatus 10 to create A third database 14 for storing information relating to the advertisement division and past advertisement frame division patterns and the products of each frame, 14 is connected to the advertisement creation processing device 10 and stores the category and area correction value. , 15 is a communication network such as the Internet, and 16 _{1 to} 16 _n (where n is a natural number) is an advertisement creation process via the communication network 15 The POS (point of sales) apparatus of each store comprised so that connection to the apparatus 10 is each shown.

  The advertisement creation processing device 10 includes a CPU (Central Processing Unit) 10a, a RAM (Random Access Memory) 10b, a ROM (Read Only Memory) 10c, a display 10d, an input / output device 10e such as a printer, a keyboard, and a mouse. , Each database has an interface 10f and a communication interface 10g.

  The “all products” may be all products handled by a large-scale retail chain having a plurality of stores, but in this embodiment, a large category of products handled by the large-scale retail chain, for example, confectionery , Vegetables, fruits, fish, meat, ham and sausages, tofu, natto, konjac, pickles, boiled and paste products, seasonings, ..., canned / bottled, frozen foods, daily necessities Assume that all products are in a large category such as a category. The PI value indicates the sales amount per 1000 people of the product, and is given by PI value = (sales amount) / (number of users) × 1000.

  FIG. 2 is a flowchart for explaining a schematic flow of advertisement creation processing by a computer in the embodiment of FIG. Hereinafter, the specific content of the advertisement creation processing in the present embodiment will be described with reference to FIG.

  In the present embodiment, the advertisement to be created is a case where the large category classification is a leaflet advertisement with one spread of confectionery. This advertisement is created every week.

  First, an advertisement in the same distribution area as the advertisement to be created and in the same week as the distribution week of this advertisement in the previous year is read from the third database 13 as a basic advertisement (step S1). In the third database 13, advertisements for each past week in each region are stored in advance including the frame division pattern and the products for each frame, and the distribution week of advertisements to be created via the input / output device 10e and When the distribution area is designated, the CPU 10a reads out the advertisement of that area and the same week of the previous year from the third database 13 and displays it on the display 10d.

  Next, the CPU 10a reads the PI value of the same week last year in the area from the first database 11 for the product of each frame of the advertisement that is the basis (step S2). The first database 11 stores in advance the PI values for all products for all products in all regions in advance, and after reading out the basic advertisement, the CPU 10 a The PI value is read from the first database 11 and displayed on the display 10d.

  FIG. 3 is a diagram showing the advertisement itself that is the basis of the advertisement to be created, and such an advertisement is displayed on the display 10d by the process of step S2. In FIG. 3, only the frame division is shown, but in actuality, a picture of the product, a selling price, and the like are displayed on each frame.

  Next, the CPU 10a performs ABC analysis of the PI values read for each product and ranks them (step S3). Furthermore, the CPU 10a displays a list arranged in descending order of the rank on the display 10d. In this embodiment, the ABC ranking is such that the upper 80% is the AB rank and the lower 20% is the C rank. The reason is that if the top 50% is AB rank, customer support products are ignored, and if the top 10% is AB rank, customer dissatisfaction is ignored, so customers, retail owners and retail employees As a result of the interview survey, it was found that it is most effective to set the lower 20% as the C rank to be replaced.

  FIG. 4 is a diagram showing a list that ranks and displays each product posted in the basic advertisement, and such a list is displayed on the display 10d by the process of step S3. This list includes ABC rank, product name, product code, retail price, PI value, product category, etc., and each product is displayed in the order of A, B, and C ranks. Within each rank, the higher PI values are displayed sequentially. It is desirable to display by color for each ABC rank.

  FIG. 5 is a diagram in which an ABC rank is added to the basic advertisement shown in FIG. 3, and it is desirable that the display as shown in FIG. 5 is also displayed on the display 10d after the process of step S3. In FIG. 5, only the frame division and the ABC rank are shown, but in reality, a photograph, a selling price, etc. of the product are displayed on each frame.

  Next, the CPU 10a reads the PI values for the same week in the previous year for all the products from the first database 11, and further stores the categories for each product category and each product sales region stored in the fourth database 14. And the area correction value is read out, and the PI value of the same week in the previous year of all products in all areas is corrected with the read category and area correction value (step S4). Further, the CPU 10a displays on the display 10d a list sequentially arranged from the corrected products with high PI values.

  A specific correction example will be described. For example, when the PI value of the “actual product 2” in the same week in the previous year is 14,520 and the category and region correction value corresponding to the condition are 1.2, the corrected PI The value is 14,520 × 1.2 = 17,424.

  FIG. 6 is a diagram showing a part of a list in which all products are displayed in the order based on the corrected PI value, and such a list is displayed on the display 10d by the process of step S4. This list includes the product name, product code, retail price, product category, region, category used and region correction value, year / month / week, corrected PI value, etc., and each product is in the order of the corrected PI value. Is displayed.

  FIG. 7 is a diagram showing a part of the category and area correction values stored in the fourth database 14. The category and region correction values are obtained in advance by correcting the PI value by using a correction value obtained in advance for each product category, each product sales region, and each product sales month. It is for trying to legitimately evaluate.

  FIG. 8 is a flowchart for explaining the schematic flow of the category and region correction value calculation processing. Hereinafter, the calculation process of the category and the region correction value will be described with reference to FIG.

  First, the CPU 10a reads the regional PI value in a certain month of a certain product category from the first database 11 (step S81). For example, if the product category is “Chocolate”, the A and B districts exist by region from January to December, such as by region in Hokkaido, Tohoku, Tokyo, Chubu / Tokai, Kinki, China, Kyushu, etc. , C district, D district, E district, F district, G district, H district, I district, J district, and K district are obtained.

  Next, the CPU 10a obtains the average value of the obtained regional PI values for a certain month, and obtains the category and the regional correction value of the month from the ratio of each regional PI value to the obtained average value (step S82). For example, for January, the PI values for each region are summed, and this is divided by 12 to obtain the average value for January. By dividing the PI value for each region in January by this average value, Obtain category and region correction values.

  Next, the CPU 10a determines whether or not the category and area correction value have been obtained for all months (step S83), and obtains all the category and area correction values for January to December.

  Next, the CPU 10a determines whether or not the category and the region correction value have been obtained for all the product categories (step S84), and obtains the category and the region correction value for all the product categories.

  Thereafter, the CPU 10a stores the obtained category and area correction value in the fourth database 14 by product category and by month (step S85).

  After the process of step S4 shown in FIG. 2, the CPU 10a selects the same number of products as the number of products of the C rank among the products ranked in order from the product with the largest corrected PI value (step S5). That is, in the example shown in FIGS. 4 and 5, since four products have a C rank, four products are selected from the top in FIG.

  Next, the CPU 10a removes the C-ranked product from the list, replaces the selected product with that part, and further ranks in the order of the PI value (step S6). FIG. 9 is a diagram showing a list displaying products of ranks A and B and selected products. Such a list is displayed on the display 10d by the replacement process described above. In this list, AB rank, product name, product code, retail price, PI value, product category, etc. are described. Each product is displayed in the order of ranks A and B, and the replaced product has the original C rank. In position. After that, when these products are ranked in the order of PI values, a list as shown in FIG. 10 is displayed on the display 10d. In this list, AB rank, product name, product code, retail price, PI value, product category, etc. are described. Each product is displayed in the rank order of A and B, and the replaced four products have high PI values. Therefore, it is in the upper position.

  Thereafter, the CPU 10a reads the location correction value of each frame of the advertisement based on the second database 12 (step S7).

  FIG. 11 is a diagram showing the location correction value of each frame of the basic advertisement stored in the second database 12 corresponding to the frame. This location correction value is a correction value for legitimately evaluating the advertising power in each frame of the advertisement, and can be calculated when the frame pattern and distribution time of the advertisement are determined.

  FIG. 12 is a flowchart for explaining the schematic flow of this location correction value calculation process. Hereinafter, the location correction value calculation process will be described with reference to FIG.

  First, the CPU 10a reads one advertisement (advertising for a certain confectionery of a certain week) from the third database 13 in the distribution area of the advertisement in the previous year or the past year (the latest 52 weeks), and the advertisement. Are read from the first database 11 (step S121).

  Next, the CPU 10a virtually divides the advertisement into cells that are the minimum division areas, and calculates the sales amount for each cell by dividing the sales amount of each frame by the number of cells constituting the frame. The cumulative value of the sales amount for each cell in the advertisement is obtained (step S122).

  Next, the CPU 10a determines whether or not the cumulative value of the sales amount for each cell has been obtained for all advertisements for the previous year or the past year (all advertisements for the confectionery) (step S123).

  When it is determined that all advertisements have not been calculated (in the case of NO), the CPU 10a reads the other advertisements of the previous year (the advertisements of confectionery of other weeks) from the third database 13 (step S124), step The processes of S121 to S123 are repeated. When it is determined that the calculation has been made for all advertisements (in the case of YES), the CPU 10a executes the process of the next step S125.

  Here, the processes of steps S121 to S124 described above will be described in detail below. FIG. 13 is a diagram showing a state in which an advertisement is virtually divided into cells that are the minimum divided regions, and FIG. 14 is a diagram for explaining a procedure for obtaining a sales amount for each cell when virtually divided in this way. It is. However, the frame allocation and the address assignment method in these drawings are different from the frame allocation of the present embodiment described above.

  As shown in FIG. 13, the advertisement is virtually divided into cells that are the minimum divided regions. This cell is defined as a minimum division area in which each frame of the advertisement can be divided by a natural number. Furthermore, as shown in FIG. 13, for example, the address is assigned to each cell so that the address of the cell 130 becomes “G-6”. Next, the sales amount of each cell in the frame is obtained by dividing the sales amount of each frame of the advertisement by the number of cells constituting the frame. For example, as shown in FIG. 14, since the number of cells constituting the frame 140 is 32, if the sales amount of the frame 140 is 2 million yen, the sales amount for each cell in the frame 140 is 2 million. Yen / 32 = 6250,000 yen. Similarly, the sales amount for each cell is obtained for all advertisements for the previous year or the past year, and the cumulative value is calculated.

  Thereafter, in step S125, the CPU 10a obtains a composition ratio coefficient representing the composition ratio of the sales amount for each cell from the cumulative value of the sales amount for each cell. FIG. 15 is a diagram illustrating an example of the component ratio coefficient for each cell obtained in this way in the advertisement of the present embodiment.

  Next, the CPU 10a calculates a frame actual value corresponding to the sales amount of each frame by adding the composition ratio coefficient for each cell for the frame pattern of the advertisement, and a frame that is an average value of the frame actual value of the advertisement. An average value is obtained (step S126). FIG. 16 is a diagram showing the relationship between the composition ratio coefficient for each cell and each frame of the advertisement, and FIG. 17 is a diagram showing the frame result value for each frame obtained as described above. The average value of the frame achievement values, that is, the frame average value is 100/20 = 5.0 because the total value of the frame achievement values in all the frames is 100.0 and the number of frames is 20.

  Next, the CPU 10a calculates a location correction value for each frame of the advertisement from the frame actual value and the frame average value thus determined (step S127). That is, (location correction value) = (frame actual value) / (frame average value) is calculated for each frame. For example, for the upper left frame in FIG. 17, since the frame actual value is 9.9, the location correction value is obtained as 9.9 / 5.0 = 1.98. 18 and 19 are diagrams showing the location correction value of each frame obtained in this way.

  Next, the CPU 10a stores the location correction value thus obtained in the second database 12 (step S128).

  Returning to the description of the advertisement creation process in FIG. 2, the CPU 10 a ranks the products in the order of PI values in step S <b> 6 and obtains a list as shown in FIG. 10. 2 are sequentially allocated to frames with high location correction values read from the database 12 (step S8).

  Next, the CPU 10a corrects the PI value of the product of each frame allocated in this way by multiplying the location correction value of that frame (step S9). FIG. 20 is a diagram showing the PI value of each frame corrected with the location correction value after the product is allocated. For example, if the PI value of the product in the lower right frame A2 is 25,290, the location correction value of this frame is 3.22, and therefore the corrected PI value is 25 as shown in FIG. 290 × 3.22 = 81,434. In the list of FIG. 9 and FIG. 10, the description of the first-ranked product that enters the frame A1 is omitted.

  Thereafter, the CPU 10a obtains the predicted sales amount of each frame by multiplying the correction PI value of each frame thus obtained by the number of users and dividing by 1000 (step S10). Further, the predicted sales amount of the entire advertisement may be obtained by adding the predicted sales amounts of all the frames. FIG. 21 is a diagram showing the predicted sales amount of each frame obtained by converting the corrected PI value. For example, in FIG. 20, since the product of the lower left frame A3 has a corrected PI value of 55,060, it is multiplied by 27,934, which is the number of users, and divided by 1000, as shown in FIG. The predicted sales amount is 1,538,042 (yen). That is, 55,060 × 27,934 / 1000 = 1,538,042 (yen).

  Next, the CPU 10a stores the advertisement layout, the predicted sales amount, and the like thus created in the third database 13 (step S11).

  As described above, according to this embodiment, the PI value of the product posted in the basic advertisement distributed in the same week in the previous year with respect to the distribution time of the advertisement to be created is used as the basic advertisement. Products listed in the advertisement that ranks the listed products, selects the product with the highest PI value among all the products in the same week last year, and replaces the selected product with the product with C rank Ranking. Further, the advertisements to be created are determined by sequentially allocating the ranked products to the frames having the highest location correction value in the descending order of ranking, thereby creating an advertisement. In this way, since the products are replaced by ranking according to the PI value in the same week of the previous year, it is possible to create an objective and reliable advertisement reflecting the sales performance. Further, since the product is divided into frames using the location correction value calculated in advance for each frame, it is possible to create an advertisement with a higher advertising effect. In addition, since the creation of a new advertisement is automatically performed from past data without human intervention, it is possible to create an advertisement easily and quickly.

  All the embodiments described above are illustrative of the present invention and are not intended to be limiting, and the present invention can be implemented in other various modifications and changes. Therefore, the scope of the present invention is defined only by the claims and their equivalents.

10 Advertisement creation processing device 10a CPU
10b RAM
10c ROM
10d display 10e input / output device 10f interface 10g communication interface 11 first database 12 second database 13 third database 14 fourth database 15 communication network 16 _{1 to} 16 _n POS device of each store

Claims (20)

Store PI values for all products by time in the first database,
The PI value of the product published in the advertisement distributed at the time corresponding to the distribution time of the advertisement to be created is read from the first database, and the products listed in the distributed advertisement are ranked. ,
Reading out the PI value at the corresponding time of all the products from the first database, and selecting the product with the higher PI value read out,
Rank the products that are listed in the advertisement to be created by replacing the selected products with the lower ranking products,
For each frame of the advertisement to be created, a coefficient value based on the frame actual value is calculated in advance and stored in the second database,
Sequentially assign the ranked products to frames with high coefficient values based on the frame actual values read from the second database in order of ranking.
An advertisement creation method, wherein an advertisement in which the ranked products are allocated to each frame is stored as an advertisement to be created in a third database.   The advertisement creation method according to claim 1, wherein the first database stores weekly PI values for all products.   The advertisement creating method according to claim 1 or 2, wherein the ranking is a method of ranking products placed in the advertisement at the corresponding time by ABC analysis of PI values.   The advertisement creation method according to claim 3, wherein the lower ranking product is a C rank product.   A category and a region correction value for each product category and each product sales region are obtained and stored in a fourth database in advance, and the product selection is performed by using the fourth database to obtain the PI values of all the obtained products. The advertisement creation method according to any one of claims 1 to 4, wherein the advertisement is corrected with the read category and region correction value, and the corrected PI value selects a higher-order product.   In the fourth database, the category and region correction values are stored for each month. The category and region correction values read from the fourth database and corresponding to the distribution month of the advertisement to be created are obtained as described above. The advertisement creation method according to claim 5, wherein PI values of all products are corrected.   The coefficient value based on the frame actual value of the advertisement to be created is a location correction value calculated in advance from the past sales amount of each frame of the advertisement to be created and stored in the second database. The advertisement creation method according to claim 1, wherein the advertisement creation method is characterized. The location correction value is
Divide one ad into cells that are the smallest unit,
Calculate the sales amount for each cell from the sales amount of the product corresponding to each frame in the one advertisement and the number of cells in the frame,
Calculate the cumulative value of the sales amount for each cell for a plurality of advertisements in a predetermined period in the past to obtain the composition ratio coefficient that represents the composition ratio of the sales amount for each cell,
For the frame pattern of the advertisement to be created, add the composition ratio coefficient for each cell to obtain the frame actual value corresponding to the sales amount of each frame,
The frame average value which is the average value of the frame actual value of the advertisement to be created is obtained, and (location correction value) = (frame actual value) / (frame average value) is calculated for each frame of the advertisement to be created. The advertisement creation method according to claim 7, wherein the advertisement creation method is obtained.   9. The advertisement creation method according to claim 7 or 8, wherein the PI value of the product allocated to each frame of the advertisement to be created is further corrected with the location correction value.   The advertisement creation method according to claim 9, wherein the PI value corrected by the location correction value of the advertisement to be created is converted into a sales amount to calculate a predicted sales amount for each frame and the entire advertisement. A first database for storing PI values by period for all products;
The PI value of the product posted in the advertisement distributed at the time corresponding to the distribution time of the advertisement to be created is read from the first database, and the products listed in the distributed advertisement are ranked. Means,
Means for reading PI values at the corresponding time of all products from the first database, and selecting a product having a higher PI value.
Means for ranking the products to be posted in the advertisement to be created by replacing the ranking-subordinate product with the selected product;
A second database for storing a coefficient value based on a frame actual value for each frame of the advertisement to be created that is calculated in advance;
Means for sequentially allocating the ranked products to frames having a high coefficient value based on the frame actual value read from the second database in the descending order of the ranking;
An advertisement creation system comprising: a third database that stores advertisements in which the ranked products are allocated to each frame as advertisements to be created.   12. The advertisement creating system according to claim 11, wherein the first database is storage means for storing weekly PI values for all products.   13. The advertisement creation system according to claim 11 or 12, wherein the ranking means is a means for ranking products placed in the advertisement at the corresponding time by ABC analysis of PI values.   14. The advertisement creation system according to claim 13, wherein the lower ranking product is a C rank product.   A fourth database for storing categories and region correction values for each product category and each product sales region, wherein the product selection means stores the PI values of all the obtained products in the fourth database; The advertisement creation system according to any one of claims 11 to 14, wherein the advertisement creation system is a unit that corrects with the category and region correction value read from the data, and selects the high-order product with the corrected PI value.   The fourth database stores the category and region correction values for each month, and the selecting means reads the category and region correction corresponding to the distribution month of the advertisement to be created, read from the fourth database. 16. The advertisement creation system according to claim 15, wherein the advertisement creation system is configured to correct PI values of all the obtained products by values.   12. The location correction value calculated in advance from the past sales amount of each frame of the advertisement is stored as the coefficient value based on the frame actual value of the advertisement in the second database. 17. The advertisement creation system according to any one of items 16 to 16. A means of dividing one advertisement into cells, which are the smallest units,
Means for calculating the sales amount for each cell from the sales amount of the product corresponding to each frame in the one advertisement and the number of cells in the frame;
Means for calculating the cumulative value of the sales amount for each cell for a plurality of advertisements in a predetermined period in the past and obtaining a composition ratio coefficient representing the composition ratio of the sales amount for each cell;
For the frame pattern of the advertisement to be created, a means for calculating the frame actual value corresponding to the sales amount of each frame by adding the composition ratio coefficient for each cell,
Means for obtaining a frame average value that is an average value of the frame actual values of the selected advertisements and calculating (location correction value) = (frame actual value) / (frame average value) for each frame of the advertisement to be created. When,
18. The advertisement creation system according to claim 17, further comprising means for storing the calculated location correction value in the second database.   The advertisement creation system according to claim 17 or 18, further comprising means for further correcting the PI value of the product allocated to each frame of the advertisement with the location correction value.   The advertisement creation according to claim 19, further comprising means for converting the PI value corrected with the location correction value of the advertisement into a sales amount to calculate a predicted sales amount for each frame and the entire advertisement. system.