Advertisement distribution method and ad distribution platform
The advertisement distribution platform improves targeting accuracy by analyzing user data to deliver ads to intended groups, enhancing click-through rates and reducing costs while minimizing fraud.
Patent Information
- Application Number
- JP2023223231
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-12-28
- Publication Date
- 2024-11-13
- Estimated Expiration
- 2043-12-28
AI Technical Summary
Existing advertisement distribution methods struggle to accurately target the intended user group, leading to wasted advertising costs and reduced profitability, particularly when displaying ads to non-intended groups such as foreigners residing in Japan who are not tourists.
An advertisement distribution platform that collects and analyzes user identifiers, geographic information, and date and time data to determine user movement and stay information, allowing targeted advertisement delivery to intended groups, such as traveling users, by excluding non-traveling individuals and compressing data to improve processing speed and reduce costs.
Enhances the accuracy of advertisement targeting, improves click-through rates, reduces costs per click, and minimizes ad fraud by delivering ads to the intended audience, contributing to increased revenue from inbound tourism.
Smart Images

Figure 0007583147000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a method for distributing advertisements to be displayed on a website to users by an advertisement distribution platform, and to the advertisement distribution platform. [Background technology]
[0002] In recent years, some advertisements displayed on websites are traded through RTB (real-time bidding) using an SSP (supply-side platform) and a DSP (demand-side platform) for the website's advertising space, and as a result, the advertising space is delivered by the advertiser who won the bid via the DSP (see, for example, Patent Document 1). [Prior art documents] [Patent documents]
[0003] [Patent Document 1] JP 2014-102663 A Summary of the Invention [Problem to be solved by the invention]
[0004] However, even if an ad space is won through RTB as described above and an ad is displayed, it is difficult for the advertiser to actually make a profit unless the ad is displayed to the target user group. For example, when trying to display advertisements for hotels and tourist spots to foreign tourists visiting Japan, if targeting is performed based on the language of the browser used by the user, the advertisements will often be displayed to foreigners residing in Japan who are staying in the country for a long time, resulting in a waste of advertising costs. The present invention has been made in consideration of the above problems, and an object of the present invention is to improve the accuracy of targeting of advertisement delivery to intended user groups. [Means for solving the problem]
[0005] (Modes of the Invention) The following aspects of the invention are illustrative of the configuration of the present invention, and are described in terms of items to facilitate understanding of the various configurations of the present invention. Each item does not limit the technical scope of the present invention. Therefore, while taking into consideration the best mode for carrying out the invention, even if some of the components of each item are replaced or deleted, or other components are added, they may be included in the technical scope of the present invention.
[0006] (1) A method for delivering advertisements to users for display on websites by an advertising distribution platform, comprising: acquiring a user identifier, geographical information, and date and time included in an advertisement display request from a website visited by the user, and a user identifier, geographical information, and date and time provided by an integrated service of the advertising distribution platform used by the user; storing these in chronological order for each user; analyzing the stored data; grasping movement information and stay information for each user based on the relationship between the user's stay location and the date and time; and targeting users to whom advertisements will be delivered based on the movement information and stay information.
[0007] The advertisement delivery method described in this section is executed by an advertisement delivery platform, and collects geographic information indicating the area where the user is staying, and performs targeting of advertisement delivery based on the collected information. Specifically, a user identifier, geographic information, and date and time contained in an advertisement display request transmitted from a website visited by the user via an SSP or the like are acquired. Furthermore, a user identifier, geographic information, and date and time are acquired from information provided by a linked service used by the user. The acquired data is then stored in chronological order for each user identified by the user identifier. The acquisition and storage of this data is performed continuously. The linked services here refer to websites, applications, information provision services, etc. that are in a collaborative relationship with the advertisement delivery platform being used and the company that operates it.
[0008] Next, the data acquired and stored as described above is analyzed for each user, and the movement information and stay information of each user are grasped from the relationship between the stay location grasped from the geographic information of each user and the date and time at that time. In other words, if the relationship between the stay location and the date and time for each user is known, the stay information such as how long the user stayed in a certain area and the movement information such as the user moving to another area at a certain point in time can be grasped. Then, the grasped movement information and stay information can infer the purpose of the user's stay in a certain area and the purpose of moving from one area to another area, and the like, and based on that information, the user to whom the advertisement is delivered is targeted. This improves the accuracy of targeting the advertisement delivery to the intended user group. As a result, the advertisement is delivered accurately to the intended user group, and it is expected that the click-through rate of the advertisement displayed on the website will be improved, the cost per click will be reduced, and ad fraud will be eliminated.
[0009] (2) In the above (1), an advertisement delivery method is provided, which targets a traveling user based on the movement information and the stay information, and delivers advertisements to the traveling user. The advertisement delivery method described in this section delivers advertisements to traveling users by targeting them based on the movement information and stay information of each user as understood by analyzing the stored data as described in (1) above. That is, if the period during which the user stays in each region, whether the user has moved from one region to another, and the frequency of the user's movement are known, it is possible to determine whether the user is traveling or not, and users who are not traveling are excluded from the targeting candidates. This allows advertisements for travelers to be displayed appropriately to traveling users. For example, if the travel destination is Japan, foreigners residing in Japan who have been staying in Japan for a long time are excluded, and advertisements for hotels, tourist spots, etc. are displayed to foreigners visiting Japan who are visiting Japan for sightseeing, which contributes to an increase in inbound revenue.
[0010] (3) In the above (2), an advertising delivery method includes extracting predetermined characteristics for each user from the stored data, and compressing the stored data by integrating multiple data and / or deleting unnecessary data based on the extracted predetermined characteristics. The advertisement delivery method described in this section extracts predetermined characteristics for each user, such as the area of stay, the duration of stay, and movement between areas, from data for each user acquired and stored from advertisement display requests and linked services. Then, based on the extracted predetermined characteristics, the data to be stored is compressed by integrating and compressing multiple pieces of data into a different valuable format, or by deleting unnecessary data that may be considered noise. This also contributes to improving processing speed and reducing costs, since a huge amount of data to be stored is compressed without losing valuable data for each user.
[0011] (4) The advertisement delivery method according to (3) above, further comprising extracting from the stored data at least data indicating a change in the area of stay of each user, and compressing the stored data. The advertisement delivery method described in this section is to extract data indicating changes in the area of stay of each user from data for each user acquired and stored from advertisement display requests and linked services. For example, a case will be described in which five pieces of data for a certain period of time for a certain user are stored in chronological order in the order of description. The data are "data on stay in area A", "data on stay in area A", "data on stay in area B", "data on stay in area B", and "data on stay in area A".
[0012] In this case, the third "data of staying in area B" indicating that the area of stay has changed from area A to area B, and the fifth "data of staying in area A" indicating that the area of stay has changed from area B to area A are extracted. In addition to this, the first "data of staying in area A" indicating the initial area of stay may be extracted. Then, the data to be saved is compressed by saving only the extracted data and deleting the data that was not extracted. This contributes to improving processing speed and reducing costs because the huge amount of data to be saved is compressed while leaving only the data necessary to grasp the movement information and stay information of each user.
[0013] (5) The above ( 3 ) In the advertising delivery method, as the specified characteristic, at least one of the following is extracted for each user for a specific period of time: the number of stays in a specific area, the duration of stay in each area when traveling between two or more specific areas, the cumulative duration of stay in any area, the number of trips between areas, the period elapsed since the previous trip between areas, and the period elapsed between data indicating trips between areas. The advertisement delivery method described in this section specifically specifies a predetermined feature to be extracted from the stored data for each user. That is, at least one of the following is extracted for a specific period for each user as a predetermined feature: the number of stays in a specific area, the duration of stay in each area when moving between two or more specific areas, the cumulative duration of stay in any area, the number of moves between areas, the period elapsed since the previous move between areas, and the period elapsed between data showing the move between areas. Then, based on each of the extracted features, data is integrated or deleted, and the stored data is compressed. As a result, the stored data is compressed while the stored opaque data is converted into more valuable data, thereby achieving both compression of the stored data and improvement of targeting accuracy.
[0014] (6) An advertisement delivery platform that delivers advertisements to users for display on websites, comprising: a data acquisition unit that acquires a user identifier, geographic information, and date and time included in an advertisement display request from a website visited by the user, and a user identifier, geographic information, and date and time provided from an integrated service of the advertisement delivery platform used by the user; a data storage unit that stores the data acquired by the data acquisition unit in chronological order for each user; a data analysis unit that analyzes the data stored by the data storage unit and grasps movement information and stay information of each user from the relationship between each user's stay location and date and time; and a targeting unit that targets users to whom advertisements are delivered based on the movement information and the stay information. an advertisement delivery unit that delivers advertisements to users targeted by the targeting unit; Including The targeting unit targets users who are traveling based on the movement information and the stay information, and the advertisement distribution unit distributes advertisements aimed at the users who are traveling. The data storage unit further includes a data extraction unit that extracts predetermined characteristics for each user from the data stored by the data storage unit, and integrates a plurality of data and / or deletes unnecessary data based on the extracted predetermined characteristics to compress the data to be stored. Ad serving platform. And, the ad distribution platform described in (6) is the above ( 3 ) advertising distribution method, 3 ) has an effect equivalent to that of the advertisement distribution method described above. Effect of the Invention
[0015] Since the present invention is configured in this manner, it is possible to improve the accuracy of targeting advertisement delivery to an intended user group. [Brief description of the drawings]
[0016] [Figure 1] 1 is a block diagram showing an example of a configuration of an advertisement distribution platform according to an embodiment of the present invention. [Diagram 2] 1 is a flow diagram showing an example of a procedure of an advertisement delivery method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0017] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Hereinafter, the same or corresponding parts will be indicated by the same reference numerals throughout the drawings. Furthermore, detailed description of the same or corresponding parts as those in the prior art will be omitted. Fig. 1 shows an example of the configuration of an advertisement delivery platform 10 according to an embodiment of the present invention. Here, the role of each component of the advertisement delivery platform 10 will be described, and specific operations will be described with reference to the flow diagram of Fig. 2 described later. Note that the configuration of the advertisement delivery platform 10 according to the embodiment of the present invention is not limited to the block diagram of Fig. 1, and may be such that some of the components shown in Fig. 1 are deleted, changed, or appropriately added depending on the procedure or situation of the advertisement delivery method, for example.
[0018] As shown in FIG. 1, the advertisement delivery platform 10 includes a data acquisition unit 12, a data storage unit 14, a data extraction unit 16, a data analysis unit 18, a targeting unit 20, and an advertisement delivery unit 22. The advertisement delivery platform 10 of the embodiment of FIG. 1 is used in an RTB (real-time bidding) environment, and corresponds to a so-called DSP (demand side platform). For this reason, the advertisement delivery platform 10 exchanges information with a plurality of websites 40 having advertisement frames to be bid on, and an SSP (supply side platform) 50. Each of the websites 40 has one or more advertisement frames to which advertisements are delivered, and is viewed by a plurality of users U via a terminal 30. Furthermore, the advertisement delivery platform 10 is also configured to receive information from a plurality of linked services 60, which will be described later.
[0019] The data acquisition unit 12 of the advertisement distribution platform 10 acquires necessary data from the advertisement display request received from the website 40 via the SSP 50 and information received from the linked service 60. For example, the data acquisition unit 12 acquires, from the advertisement display request of the website 40, the IP address of the terminal 30 used by the user U browsing the website 40, the transmission date and time (reception date and time) of the advertisement display request, and the like, and further acquires from these a user identifier capable of identifying each user U, geographical information indicating the area where the user U is thought to be staying, and the like.
[0020] Here, the linked services 60 are various services linked with the advertisement delivery platform 10 and the company that operates the advertisement delivery platform 10, and examples of such services include general websites, applications, and information provision services using a DMP (data management platform). From such linked services 60, the data acquisition unit 12 acquires, for example, the access log of the user U, the date and time, the IP address, the user agent, the profile (age, sex, occupation, nationality, etc.) entered by the user U, the language information set by the user U, and the date and time of the access. Furthermore, the data acquisition unit 12 acquires a user identifier that can identify each user U determined from the above, and geographical information indicating the area where the user U is thought to be staying. Examples of user identifiers acquired from the advertisement display request and the linked services 60 include cookies, advertisement IDs, and common IDs.
[0021] The data storage unit 14 stores the data acquired by the data acquisition unit 12 as described above and the data extracted by the data extraction unit 16 as described below. When the data acquired by the data acquisition unit 12 is stored, the data is stored in chronological order based on the date and time of each data for each user U based on the user identifier. Here, the user identifier acquired by the data acquisition unit 12 from the advertisement display request and the user identifier acquired from the linked service 60 may have different structures, but if it is determined that they are user identifiers indicating the same user U, they are stored together as data of the user U. It is possible to determine whether user identifiers with different structures belong to the same user U by using a common part between those user identifiers. An arbitrary database or the like may be used as the storage destination of the data by the data storage unit 14, and different storage destinations may be used for the data acquired by the data acquisition unit 12 and the data extracted by the data extraction unit 16.
[0022] The data extraction unit 16 extracts data satisfying specific conditions and preset characteristics for each user U from the data acquired by the data acquisition unit 12 and stored by the data storage unit 14 for the purpose of compressing the stored data, etc. Such specific conditions and preset characteristics will be described in detail later. The data extracted by the data extraction unit 16 and data derived from the characteristics extracted by the data extraction unit 16 as described below are stored by the data storage unit 14 as described above. The data analysis unit 18 analyzes the data stored by the data storage unit 14 and grasps the movement information and stay information of each user U from the relationship between the stay location of the user U determined from the geographic information and the date and time of the data indicating the stay location.
[0023] The targeting unit 20 targets users U to whom advertisements are to be delivered based on the movement information and stay information of each user U grasped by the data analysis unit 18. Since the user demographic to be targeted varies depending on the content of the advertisement, the targeting unit 20 targets advertisement delivery according to the characteristics of the user U grasped from the movement information and stay information of the user U and the content of the advertisement to be delivered. Note that targeting here also includes the meaning of excluding users U who are not targeted by a certain advertisement from candidates for delivery of that advertisement. The advertisement distribution unit 22 distributes an advertisement to a user U targeted by the targeting unit 20 for the advertisement, and distributes the advertisement to an advertisement space of the website 40 in response to an advertisement display request from the website 40 that the targeted user U is browsing.
[0024] Here, the components of the advertisement delivery platform 10 as described above are divided into functional units, and are not divided into units of software or hardware that actually construct the advertisement delivery platform 10. Furthermore, any software and hardware can be used for the software and hardware that construct the advertisement delivery platform 10. Furthermore, the terminal 30 used by the user U may be any terminal that can access the website 40 via a network, and may be, for example, a personal computer, a tablet computer, a smartphone, a mobile phone, or the like. Although not shown in FIG. 1, the advertisement delivery platform 10 is configured so that various settings and inputs are performed by an administrator terminal used by an administrator of the advertisement delivery platform 10, or the like.
[0025] Next, an advertisement delivery method according to an embodiment of the present invention, which uses the advertisement delivery platform 10 shown in FIG. 1 and is used to deliver advertisements to a foreign user U traveling in Japan, will be described along the flow of the flow diagram shown in FIG. 2, taking as an example a case where advertisements are delivered to the user U. For the configuration of the advertisement delivery platform 10, refer to FIG. 1 as appropriate. Note that the flow diagram shown in FIG. 2 shows an example of a procedure flow for explaining the advertisement delivery method according to the embodiment of the present invention. Therefore, the advertisement delivery method is not limited to these flow diagrams, and may be a flow in which some of the steps shown in FIG. 2 are deleted, changed, or appropriately added depending on the configuration or situation of the advertisement delivery platform 10, for example.
[0026] S10 (data acquisition): The data acquisition unit 12 acquires data including a user identifier, geographic information, and date and time from an advertisement display request transmitted from the website 40 via the SSP. Furthermore, the data acquisition unit 12 acquires data including a user identifier, geographic information, and date and time from the linked service 60. The data is acquired continuously, and data is acquired each time an advertisement display request is received and each time data is provided from the linked service 60.
[0027] S20 (data storage): The data acquired by the data acquisition unit 12 in S10 above is stored by the data storage unit 14. That is, data including a user identifier, geographic information, and date and time acquired from the advertisement display request, and data including a user identifier, geographic information, and date and time acquired from the linked service 60 are stored. At this time, the user identifier is used to identify each user U, and the date and time data is used to store data in chronological order for each user U. As in S10 above, data storage in this step is also performed continuously, and data is stored every time it is acquired in S10 above, accumulating data for each user U.
[0028] S30 (obtaining data of the relevant user): The data extraction unit 16 obtains data for any specific period, including the latest data, of the relevant user U (hereinafter referred to as the "relevant user U1" for ease of explanation) whose trends it is desired to understand from the data for each user U accumulated by being saved in the above S20. At this time, the data of the relevant user U1 saved in S70 and S90, which will be described later, may also be obtained. The acquisition of these data may be performed at any timing, in addition to the timing when new data is saved in the above S20.
[0029] S40 (judgment of change in stay area): The data extraction unit 16 judges whether the stay area (stay country in this embodiment) of the user U1 has changed based on the data of the user U1 acquired in S30. That is, it judges whether the stay country identified from the geographic information of the latest data of the user U1 has changed from the stay country identified from the geographic information of the previous data of the user U1. As a result, if it is judged that the stay country has changed (YES), the process proceeds to S50, and if it is judged that the stay country has not changed (NO), the process returns to S10 to continue acquiring and saving data.
[0030] S50 (determining the number of changes in region of stay): The data extraction unit 16 determines whether the number of changes in the region of stay (the country of stay in this embodiment) of the user U1 is equal to or greater than a preset certain number. That is, as will be described later in S60 and S70, data indicating that the country of stay of the user U1 has changed is extracted and stored, and the number of changes is determined based on that data. As a result, if it is determined that the number of changes in the country of stay is equal to or greater than the certain number (YES), the process proceeds to S80, and if it is determined that the number of changes in the country of stay is not equal to or greater than the certain number (NO), the process proceeds to S60.
[0031] S60 (data extraction): The data extraction unit 16 extracts the original data when it is determined in S40 that the country of stay of the user U1 has changed. That is, since the latest data of the user U1 acquired in S30 above is data indicating the change in the country of stay, this data is extracted. S70 (data compression and storage): The data extracted in S60 is stored separately from the data stored in S20 by the data storage unit 14. At this time, among the data of the user U1 stored in S20, data that is considered unnecessary, such as data older than the data stored in this step that does not indicate the transition of the country of stay, may be deleted.
[0032] S80 (feature extraction): The data extraction unit 16 extracts characteristic trends of the user U1 from the data of the change in the country of stay of the user U1, which exists in a certain number or more. In this embodiment, the "number of stays in a specific country", "stay duration in each country in a move between two or more specific countries", "cumulative stay duration in any country", "number of moves between countries", "elapsed period from the previous move between countries", and "elapsed period between data indicating moves between countries" are extracted. Here, each feature will be explained using an example in which the data "2023-12-01 @ Japan", "2023-12-02 @ China", "2023-12-03 @ Japan", "2023-12-04 @ China", and "2023-12-07 @ Japan" are stored as data indicating the change in the country of stay of the user U1 during a specific period.
[0033] "Number of stays in a specific country" refers to the number of stays in each country that the user U1 travels to and from, and in the above example, it can be confirmed that the user stayed in Japan three times and China twice. Therefore, while retaining the information that the user stayed in Japan three times and China twice, the data is compressed by setting the country of stay for those periods as Japan, since the number of stays in Japan is greater. The "stay period in each country during travel between two or more specific countries" refers to the stay period in each country while the user U1 is traveling between the countries. In the above example, it is one day in Japan from "2023-12-01@Japan" to "2023-12-02@China", one day in China from "2023-12-02@China" to "2023-12-03@Japan", one day in Japan from "2023-12-03@Japan" to "2023-12-04@China", and three days in China from "2023-12-04@China" to "2023-12-07@Japan". For this reason, while retaining that information, the data is compressed to include China, which is the longest stay period among them, for three days, as the country of stay during those periods.
[0034] "Cumulative stay period in any country" is the cumulative stay period in each country where the user U1 stayed during a specific period. In the above example, the total stay period in Japan is 2 days, from "2023-12-01@Japan" to "2023-12-02@China" and "2023-12-03@Japan" to "2023-12-04@China", and the total stay period in China is 4 days, from "2023-12-02@China" to "2023-12-03@Japan" and "2023-12-04@China" to "2023-12-07@Japan". For this reason, while retaining that information, the data is compressed for China, which has the longest cumulative stay period, as the country of stay during those periods. "Number of travels between countries" refers to the trend of the number of travels between countries in a certain period. For example, in the above example, if the period is divided into four days each, "2023-12-01~04" and "2023-12-04~07", the number of travels can be confirmed as three in the former period and one in the latter period. Therefore, while retaining this information, the former period is considered as noise, and the data is compressed as "2023-12-01~03@Japan", "2023-12-04@China", and "2023-12-07@Japan".
[0035] "The period of time elapsed since the previous movement between countries" refers to the period of time elapsed between movements, and is used, for example, when compressing data, to determine whether data indicating a movement between countries is a valid movement. For example, in the above example, when determining whether the movement from "2023-12-02@China" to "2023-12-03@Japan" is valid, if the period of time elapsed since the movement immediately before this movement is a predetermined number of days or more, it is determined to be valid. The "period of time elapsed between data showing movement between countries" is used to see the frequency of movement, and is used, for example, to exclude users U whose movement information is observed frequently in order to eliminate noise and bots such as crawlers. For example, in the above example, the period of time elapsed between movements is at least one day and at most three days, and by comparing this with the overall trend, users U who are noise are excluded. In addition to the above-mentioned characteristics, any characteristics may be extracted, such as travel from a specific country to another specific country, a cumulative stay in a specific country exceeding a predetermined period, a number of travels between specific countries exceeding a predetermined number during a specific period, and no information on travel to countries other than the specific country.
[0036] S90 (data compression and storage): The data compressed based on the features extracted in S80 is stored separately from the data stored in S20 by the data storage unit 14. At this time, the data may be stored permanently in anticipation of various future uses. S100 (Data Analysis): The data analysis unit 18 analyzes the data stored in S20, S70, and S90 to grasp the movement information and stay information of the user U1. The movement information is information indicating a movement from one country to another and information related thereto, and the stay information is information indicating a stay in a country and information related thereto. In this embodiment, since some data has already been analyzed by the data extraction unit 16 through S60 and S80, this is used.
[0037] S110 (information amount determination): The targeting unit 20 determines whether or not sufficient information has been collected to target the advertisement to be delivered to the user U1 based on the movement information and stay information of the user U1 grasped in S100. In this embodiment, the advertisement to be delivered is an advertisement aimed at foreign tourists visiting Japan. As a result, if it is determined that sufficient information has been collected (YES), the process proceeds to S120, and if it is determined that sufficient information has not been collected (NO), the process returns to S10 to continue acquiring and storing data.
[0038] S120 (Targeting): The targeting unit 20 determines whether or not to target the advertisement to be delivered to the user U1, in other words, whether or not the user U1 is a foreign tourist visiting Japan. If it is determined that the user U1 is a foreign tourist visiting Japan and therefore should be targeted (YES), the process proceeds to S130, and if it is determined that the user U1 is not a foreign tourist visiting Japan and therefore should not be targeted (NO), the process proceeds to S140.
[0039] S130 (Advertisement delivery): The advertisement delivery unit 22 delivers the advertisement to be delivered, which in this embodiment is an advertisement for foreign tourists visiting Japan, to the user U1 targeted in S120. The advertisement is delivered to the advertisement space of the website 40 visited by the user U1 who sent the advertisement display request. S140 (Exclude relevant user): The targeting unit 20 determines that the relevant user U1 is not a foreign tourist visiting Japan, but rather a foreign resident of Japan who is staying in Japan for a long period of time, and therefore excludes the relevant user U1 from targeting candidates (delivery candidates) for the advertisement to be delivered.
[0040] Here, the advertisement delivery platform 10 according to the embodiment of the present invention and the advertisement delivery method according to the embodiment of the present invention are not limited to the above-mentioned embodiment, and various modifications can be made by those skilled in the art within the technical concept of the present invention. For example, the data acquired from the advertisement display request or the linked service 60 is not limited to a user identifier, geographic information, and date and time, and various other information may be acquired and used. In the explanation using FIG. 2, the geographic information is handled on a country basis in order to target foreign tourists visiting Japan, but the granularity of the geographic information is not limited to countries, and may be at the city, town, or village level. Furthermore, the content of the advertisement and its targeting candidates are not limited to foreign tourists visiting Japan, and may be various user groups that can be determined from geographic information of various granularities stored in chronological order. In addition, the advertisement space to which the advertisement is delivered is not limited to the advertisement space won in the RTB environment, and may be an advertisement space in which the advertiser has acquired the right to post the advertisement in any manner.
[0041] According to the embodiment of the present invention having the above configuration, the following effects can be obtained. That is, the advertisement delivery method according to the embodiment of the present invention is executed by an advertisement delivery platform 10 as shown in FIG. 1, for example, and collects geographic information indicating the area where the user U is staying, and performs targeting of advertisement delivery based on the collected information. Specifically, the user identifier, geographic information, and date and time contained in the advertisement display request transmitted from the website 40 visited by the user U via the SSP 50 are acquired. Furthermore, the user identifier, geographic information, and date and time are acquired from information provided by the linked service 60 used by the user U (see S10 in FIG. 2). Then, the acquired data is stored in chronological order for each user U identified by the user identifier (see S20 in FIG. 2). The acquisition and storage of these data are performed continuously.
[0042] Next, the data acquired and stored as described above is analyzed for each user U, and the movement information and stay information of each user U are grasped from the relationship between the stay location grasped from the geographic information of each user U and the date and time at that time (see S100 in FIG. 2). That is, if the relationship between the stay location and the date and time for each user U is known, it is possible to grasp the stay information such as how long the user U stayed in a certain area, and the movement information such as the user U moving to another area at a certain point. Then, from the grasped movement information and stay information, it is possible to infer the purpose of the user U staying in a certain area, the purpose of moving from one area to another area, etc., and based on such information, the user U to whom the advertisement is delivered is targeted (see S120 in FIG. 2). This makes it possible to improve the accuracy of targeting the advertisement delivery to the intended user layer. Therefore, it is possible to accurately deliver the advertisement to the intended user layer (see S130 in FIG. 2), and it is also possible to expect an improvement in the click-through rate of the advertisement displayed on the website 40, a reduction in the cost per click, and the elimination of ad fraud.
[0043] In addition, the advertisement delivery method according to the embodiment of the present invention may target users U who are traveling based on the movement information and stay information of each user U grasped by analyzing the stored data, and deliver advertisements to the users U who are traveling. That is, if the period during which the user U stays in each region, whether the user U has moved from one region to another, the frequency of the user U's movement, etc. are known, it is possible to know whether the user U is traveling or not, and it is possible to exclude users U who are not traveling from targeting candidates. This makes it possible to accurately display advertisements for travelers to users U who are traveling. For example, if the travel destination is Japan, it is possible to display advertisements for hotels, tourist spots, etc. to foreigners who are visiting Japan for sightseeing, excluding foreigners residing in Japan who are staying in Japan for a long time (see S140 in FIG. 2), and to contribute to an increase in inbound revenue.
[0044] Furthermore, the advertisement delivery method according to the embodiment of the present invention extracts data indicating a change in the area where each user U stays from the data for each user U acquired and stored from the advertisement display request or the linked service 60 (see S40 and S60 in FIG. 2). Then, the data to be stored is compressed by storing only the extracted data and deleting the data that was not extracted (see S70 in FIG. 2). This makes it possible to compress a huge amount of data to be stored while retaining the data necessary to grasp the movement information and stay information of each user U, which contributes to improving processing speed and reducing costs.
[0045] Moreover, the advertisement delivery method according to the embodiment of the present invention extracts, for each user U, predetermined characteristics related to, for example, the area of stay, the duration of stay, and movement between areas from the data for each user U acquired and stored from the advertisement display request or the linked service 60 (see S80 in FIG. 2). Then, based on the extracted predetermined characteristics, the data to be stored is compressed by integrating and compressing a plurality of data into another valuable format, or by deleting unnecessary data that may be considered as noise (see S90 in FIG. 2). This also makes it possible to compress a huge amount of data to be stored without losing valuable data for each user U, which contributes to improving processing speed and reducing costs.
[0046] In addition, in the advertisement delivery method according to the embodiment of the present invention, the predetermined feature extracted from the stored data for each user U may be as follows. That is, at least one of the following is extracted for a specific period for each user U as the predetermined feature: the number of stays in a specific area, the duration of stay in each area when moving between two or more specific areas, the cumulative duration of stay in any area, the number of moves between areas, the elapsed period from the previous move between areas, and the elapsed period between data indicating the move between areas. Then, based on each of the extracted features, data integration or deletion is performed to compress the data to be stored. As a result, the stored data can be compressed while converting the stored opaque data into more valuable data, making it possible to achieve both compression of the stored data and improvement of targeting accuracy.
[0047] On the other hand, when used in the advertising delivery method according to the embodiment of the present invention as described above, the advertising delivery platform according to the embodiment of the present invention can achieve the same effect as that of the advertising delivery method according to the embodiment of the present invention. [Explanation of symbols]
[0048] 10: Advertising distribution platform, 12: Data acquisition unit, 14: Data storage unit, 18: Data analysis unit, 20: Targeting unit, 40: Website, 60: Linked service, U (U1): User
Claims
1. A method for delivering advertisements to be displayed on a website to a user by an advertisement delivery platform, comprising: Acquire a user identifier, geographic information, and date and time included in an advertisement display request from a website visited by the user, and a user identifier, geographic information, and date and time provided by a linked service of the advertisement distribution platform used by the user, and store them in chronological order for each user; The stored data is analyzed to grasp the movement information and stay information of each user from the relationship between the stay location of each user and the date and time of each user, Targeting users to whom advertisements are to be delivered based on the movement information and the stay information; Targeting the traveling user based on the movement information and the stay information, and delivering advertisements to the traveling user; An advertising delivery method characterized by extracting predetermined characteristics for each user from the stored data, and compressing the stored data by integrating multiple data and / or deleting unnecessary data based on the extracted predetermined characteristics.
2. 2. The advertisement delivery method according to claim 1, further comprising extracting from said stored data at least data indicating a change in the area where each user is staying, and compressing the stored data.
3. The advertising delivery method according to claim 1, characterized in that as the specified characteristic, at least one of the following is extracted for each user for a specific period of time: the number of stays in a specific area, the duration of stay in each area when traveling between two or more specific areas, the cumulative duration of stay in any area, the number of trips between areas, the period elapsed since the previous trip between areas, and the period elapsed between data indicating trips between areas.
4. An advertisement distribution platform that distributes advertisements to users to be displayed on a website, a data acquisition unit that acquires a user identifier, geographic information, and date and time included in an advertisement display request from a website visited by a user, and a user identifier, geographic information, and date and time provided by an associated service of the advertisement distribution platform used by the user; a data storage unit that stores the data acquired by the data acquisition unit in chronological order for each user; a data analysis unit that analyzes the data stored by the data storage unit and grasps movement information and stay information of each user from a relationship between the stay location of each user and a date and time of each user; a targeting unit that targets users to whom advertisements are to be delivered based on the movement information and the stay information; an advertisement delivery unit that delivers advertisements to users targeted by the targeting unit, The targeting unit targets the traveling user based on the movement information and the stay information, and the advertisement distribution unit distributes advertisements aimed at the traveling user, The advertising distribution platform further includes a data extraction unit that extracts predetermined characteristics for each user from the data stored by the data storage unit, and compresses the data to be stored by integrating multiple data and / or deleting unnecessary data based on the extracted predetermined characteristics.
Citation Information
Patent Citations
On-line advertisement distribution system
JP2014102663A
Data analysis system and data analysis method
JP2019023851A
Method for providing advertisement for TV program
JP2020167645A
Point retargeting affiliate advertisement distribution method
JP2020197964A
System for user attribute estimation based on IP address
WO2019207771A1
