KR20240086442A - Advertising automation method using big data - Google Patents

Abstract

The present invention proposes an advertising automation method using big data, which processes structured/semi-structured data, app/web log data, member data, and CRM data collected from advertisers, advertising media, and advertising media visitors to determine and display optimal exposure advertisements specialized for the visitors and the advertisers, and compares the conversion probability of targets using big data to provide the exposure advertisements specialized for the visitors. The advertising automation method using big data comprises the steps of: collecting structured/semi-structured/unstructured data, app/web log data, member data and CRM data from the advertising media, the advertisers and the advertising media visitors; and creating a pre-segment for each advertiser target using the collected data, analyzing anti-fraud customers among the advertising media visitors, and analyzing the performance of the advertising media in real time.

Description

Advertising automation method using big data {ADVERTISING AUTOMATION METHOD USING BIG DATA}

The present invention relates to advertising automation, and processes structured/semi-structured, app/web log data, member data, and CRM data collected from advertisers, advertising media, and advertising media visitors to create optimal exposure advertisements specialized for visitors and advertisers. Advertising that can be determined, exposed, and used to predict the possibility of conversion (before the digital stage targeted by the advertiser) of customers (hereinafter referred to as targets) who may reach the advertiser's goal by using big data to provide specialized exposure advertisements to visitors. It concerns automation platform servers and advertising automation methods.

As media technology develops, the number of media through which consumers access information is gradually increasing. Users play media using a variety of devices, and advertising strategies are also changing in response to changes in the media environment.

For example, advertising media is moving away from passive forms such as TV, newspapers, and electronic billboards, and can provide advertisements based on selective viewing of media, and can also be expanded to mobile advertisements such as applications or web advertisements. It is becoming.

As an example of using various advertising media, recent websites are providing advertisements in connection with the content of the site. To this end, websites sometimes provide direct advertising or hire external advertising, but the majority of websites receive advertising from advertising suppliers (or advertising sales agencies) and create a separate advertising area on the web page to provide visitors with advertising. It is being exposed.

There are a variety of advertising providers at home and abroad, including Google, and recently, as global advertising providers enter the domestic market, more advertising providers are supplying ads to popular sites in the Korean market. In this environment, a website, which is an advertising media, can contract with multiple advertising providers rather than one advertising provider for a specific advertising area and periodically expose advertisements received from multiple providers.

In this way, in order to efficiently process advertisements from multiple advertising providers for a specific advertising area, the order of exposure between advertising networks can be assigned in advance, and mediation is performed in order to maximize revenue from exposure of advertisements. It is being proposed as a technology to call other advertising networks. Publishing companies are using multiple mediation platforms to increase their fill rate and increase total revenue.

The method of providing Internet advertising and its automated system, which are registered patents in Korea (No. 10-2203478, January 11, 2021), provide the same solution regardless of the advertiser's industry or what business concerns the advertiser has, so that data Since everything from collection to dashboard is unified, there is a disadvantage that neither advertisers nor consumers visiting advertising media are satisfied.

The technical problem that the present invention aims to solve is to process structured/semi-structured data, app/web log data, member data, and CRM data collected from advertisers, advertising media, and advertising media visitors to provide optimal exposure advertising tailored to visitors and advertisers. The goal is to provide an advertising automation method that determines and exposes and uses big data to predict the target's conversion potential and provide specialized exposure advertisements to visitors.

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

The advertising automation method using big data according to the present invention to achieve the above technical problem is performed on an advertising automation platform server including an information collection module, a preliminary analysis module, a real-time analysis module, and a communication module, and the information collection module is Collecting structured/semi-structured/unstructured data, app/web log data, member data, and CRM data from advertising media, advertisers, and visitors to the advertising media, and the preliminary analysis module using the data collected from the information collection module It includes the steps of creating a pre-segment for each target , analyzing anti-fraud customers among the advertising media visitors, and analyzing the performance of the advertising media.

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

As described above, the advertisement automation method using big data according to the present invention allows individual companies (advertisers) to provide customized advertisements to selected advertisement visitors in real time, using the advertisement visitor's access information. By exploring and providing new exposure advertisements that may be of interest to the visitor, advertisers can provide optimal advertisements to visitors, and visitors can create new advertisements that reflect their interests analyzed through their access information. There is an advantage to being able to access advertisements.

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

Figure 1 is an example of an advertising automation platform server used by the present invention.
Figure 2 explains an example of data collected by the information collection module.
3 illustrates an example of a pre-segment.
Figure 4 illustrates an example of analyzing media performance in real time through Viewability in the advertising media performance analysis unit.
Figure 5 illustrates an example of media performance analysis using viewability, exposure, and click count indicators in the advertising media performance analysis unit.
Figure 6 explains an example of determining the possibility of conversion based on big data.
Figure 7 is an example of the configuration of an advertising automation system used by the present invention.
Figure 8 is an example of an advertising automation method using big data according to the present invention.

In order to fully understand the present invention, its operational advantages, and the objectives achieved by practicing the present invention, reference should be made to the accompanying drawings illustrating exemplary embodiments of the present invention and the contents described in the accompanying drawings.

Hereinafter, the present invention will be described in detail by explaining preferred embodiments of the present invention with reference to the accompanying drawings. The same reference numerals in each drawing indicate the same member.

Figure 1 is an example of an advertising automation platform server used by the present invention.

Referring to FIG. 1, the advertising automation platform server 100 according to the present invention includes an information collection module 110, a preliminary analysis module 120, a real-time analysis module 130, and a communication module 140.

Figure 2 explains an example of data collected by the information collection module.

Referring to FIG. 2, the information collection module 110 collects structured/semi-structured/unstructured data (structured data/semi structured data/unstructured data), app/web log data from advertising media, advertisers who are advertising clients, and advertising media visitors. , collects member data and CRM data. Here, CRM (Customer Relationship Management) maximizes customer-centered resources by analyzing and integrating internal and external data related to customers, such as gender, age, residence, and purchase history, and conducts marketing activities tailored to customer characteristics based on this. It refers to the process of planning, supporting, and evaluating.

The pre-analysis module 120 generates pre-segments for each advertiser in advance, analyzes anti-fraud customers among advertising media visitors, and analyzes the performance of advertising media. For this purpose, the pre-segment creation unit 121 ), anti-fraud customer analysis unit 122, and advertising media performance analysis unit 123.

The pre-segment generator 121 generates a pre-segment for each target. The dictionary segment includes information that groups visitors or members by their age, gender, region, and purchasing ability. These will become the basis for creating target segments later.

3 illustrates an example of a pre-segment.

Referring to Figure 3, it can be seen that dictionary segment 1 is grouped by the visitor's age, dictionary segment 2 is grouped by the member's gender, dictionary segment 3 is grouped by region, and dictionary segment 4 is grouped by purchasing ability.

The anti-fraud customer analysis unit 122 will be used to distinguish abnormal visitors from normal visitors by using landing page visit data compared to clicks of advertising media visitors and conversion/cancellation data after visiting the landing page.

The advertising media performance analysis department (123) will analyze the exposure area (View-ability) ratio & number of impressions/reach (Unique Imps/Clicks) of the advertisement in the advertising media to verify the efficiency of the advertisement in the specific advertising media. will be.

Figure 4 illustrates an example of analyzing media performance in real time through Viewability in the advertising media performance analysis unit.

Referring to Figure 4, although the advertisement is exposed to advertising media, a distinction is made between cases where only about 30% of the advertisement is exposed as shown on the left and cases where 100% of the advertisement is exposed as shown on the right. Of course, it is clear that advertising media with 100% exposure area is effective for the advertisement.

Figure 5 illustrates an example of media performance analysis using viewability, exposure, and click count indicators in the advertising media performance analysis unit.

Referring to Figure 5, examples where the exposure area ranges from 7% to 61% can be distinguished by the length of the bar, where the number of impressions (Unique Impression or Imps) refers to the number of times the advertisement is shown to the visitor, and the number of clicks (Unique Impression or Imps) refers to the number of times the advertisement is shown to the visitor. Clicks) is the number of clicks on the ad.

The real-time analysis module 130 uses the media visitor information analysis unit 131 to perform the function of determining exposure advertisements using advertising media visitor information or recycling effective target search and collection data using visitor access information. ), exposure advertisement decision unit 132, visitor access information analysis unit 133, conversion prediction unit 134, effective target search unit 135, and post-collection data reuse unit 136.

The media visitor information analysis unit 131 uses visitor information at the same time a specific advertising visitor visits and creates an effective segment based on the target segment in real time.

The exposure advertisement decision unit 132 determines the optimal advertisement matching the target segment and the visited advertisement media in the media visitor information analysis unit 131. The decided advertisement will be exposed to visitors through advertising media.

The visitor access information analysis unit 133 analyzes visitor access information including the visitor's visit time to the advertising media, number of visits, number of visits to the landing page, and number of purchases, and determines the visitor's interest and purchase tendency.

The conversion prediction unit 134 predicts the conversion possibility for each target segment using visitor information and visitor access information or big data. That is, by analyzing at least one of visitor information, visitor access information, and big data, a potential customer target segment with similar characteristics to existing visitors is created.

Figure 6 explains an embodiment of determining the conversion possibility for each target segment based on big data.

Referring to Figure 6, it can be seen that the target is segmented through big data analysis and targeted intensively to the optimal target group.

For example, the target that is most meaningful to advertisers is classified into grades 1 to 5 depending on the conversion potential. Tier 1 has the highest conversion potential but smallest target size, while Tier 5 has the lowest conversion potential but largest target size.

The effective target search unit 135 searches for a valid advertisement that matches the target segment of the visitor created according to the conversion probability of the visitor in the advertising media and conversion prediction unit 134.

The post-collection data reuse unit 136 sets a new target segment for the visitor and displays advertisements that are highly relevant to the content of the website according to the visitor's tendency identified from the visitor information and visitor access information when the visitor accessed the advertising media. Select and expose.

For example, contextual targeting is used to display advertisements that are highly related to website content based on the visitor's preferences. Contextual targeting means that if visitors are interested in border news and visit economic news, securities company advertisements are exposed first, and if visitors are interested in sports news and visit sports news, advertisements for sporting goods are exposed.

The communication module 140 is centered around a DSP (Demand Side Platform) installed outside the advertising automation platform server 100, and FDS (Fraud Detection System) and DMP (Data Management Platform) are used for data management. Platform), ADX (Ad Exchange), and SSP (Supply Side Platform) are connected in terms of communication.

Figure 7 is an example of the configuration of an advertising automation system used by the present invention.

Referring to Figure 7, the advertising automation system 700 according to the present invention includes an advertising automation platform server 100, FDS 710, DSP 720, DMP 730, ADX 740, SSP 750, and It can be seen that it includes advertising media 760.

The advertising automation platform server 100 uses the built-in communication module 140 to detect FDS (Fraud Detection System, 710) and DMP (Data) centered on DSP (Demand Side Platform, 720). Management Platform, data management platform, 730), ADX (Ad Exchange, ad exchange platform, 740), SSP (Supply Side Platform, 750) and advertising media (760) are connected to the network.

Figure 8 is an example of an advertising automation method using big data according to the present invention.

Referring to FIG. 8, the advertising automation method 800 using big data according to the present invention includes an information collection step 810, a preliminary analysis step 820, and a real-time analysis step 830.

In the information collection step 810, the information collection module 110 collects structured/semi-structured/unstructured data, app/web log data, member data, and CRM data from advertising media, advertisers, and visitors to the advertising media.

In the pre-analysis step 820, the pre-analysis module 120 uses the data collected by the information collection module 110 to create a pre-segment for each advertiser, and identifies anti-fraud customers among the advertising media visitors. Analyze and analyze the performance of the advertising media.

In the real-time analysis step 830, the real-time analysis module 130 uses advertising media visitor information to determine exposure advertisements, or uses visitor access information to predict conversions and recycle search and collection data for effective targets. do.

Below, the preliminary analysis step 820 and the real-time analysis step 830 will be described in detail.

The pre-analysis step (820) is a step of generating pre-segments for each target according to big data analysis (821, pre-segment creation step), visit data of the landing page compared to clicks of advertising media visitors, and conversion/cancellation data after visiting the landing page. The step of distinguishing normal visitors from abnormal visitors (822, anti-fraud customer analysis step) and the exposure area (View-ability) ratio & number of impressions/reach (Unique Imps/Clicks) of the advertisement in advertising media. It includes a step of analyzing and verifying the efficiency of the advertisement in a specific advertising media (823, advertising media performance analysis step).

Here, the pre-segment is a grouping of visitors based on at least one of their age, gender, region, and purchasing ability.

In the real-time analysis step (830), a target segment is created in real time by analyzing the visitor's tendency using the visitor information entered by the advertising visitor at the same time as the visitor (831, media visitor information analysis step), the target segment and the visited advertisement. A step of determining the optimal advertisement to be exposed that matches the media (832, exposure advertisement determination step), visitor access information including the visitor's visit time to the advertising media, number of visits, number of visits to the landing page, and number of purchases. A step of analyzing and identifying the visitor's interest and purchase tendency (833, visitor access information analysis step), a step of predicting the conversion possibility for each target of a specific visitor using visitor information and visitor access information or big data (834) , conversion prediction step), a step of searching for a valid advertisement matching the target segment of the visitor created according to the conversion probability of the visitor predicted by the advertising media and the conversion prediction unit 134 (835, effective target search step), and According to the visitor information entered by the visitor when accessing the advertising media and the visitor's tendency identified from the visitor access information, setting a new target segment for the visitor, selecting and exposing advertisements highly related to the content of the website (step 836 , post-collection data reuse step).

Here, in the step 834 of predicting the possibility of conversion, the target is segmented through big data analysis, meaningful target groups are selected, and groups with a high conversion probability are predicted and targeted intensively.

Referring to FIG. 8, the advertising automation method 800 is centered on a DSP (Demand Side Platform) installed outside the advertising automation platform server 100, and includes an FDS (Fraud Detection System), It may further include a step (not shown) of connecting DMP (Data Management Platform), ADX (Ad Exchange), SSP (Supply Side Platform), and advertising media to a network.

In the above, the technical idea of the present invention has been described along with the accompanying drawings, but this is an exemplary description of a preferred embodiment of the present invention and does not limit the present invention. In addition, it is clear that anyone skilled in the art can make various modifications and imitations without departing from the scope of the technical idea of the present invention.

100: Automation platform server
110: Information collection module
120: Pre-analysis module
121: Pre-segment creation unit 122: Anti-fraud customer analysis unit
123: Advertising media performance analysis department
130: Real-time analysis module
131: Media visitor information analysis unit 132: Exposure advertisement decision unit
133: Visitor access information analysis unit 134: Conversion prediction unit
135: Effective target search unit
136: Post-collection data reuse department
140: Communication module

Claims (11)

It is performed on an advertising automation platform server that includes an information collection module, preliminary analysis module, real-time analysis module, and communication module.
The information collection module collecting at least one of structured/semi-structured/unstructured data, app/web log data, member data, and CRM data from advertising media, advertisers, and visitors to the advertising media; and
The pre-analysis module uses the data collected from the information collection module to generate at least one of the following: creation of a pre-segment for each advertiser, analysis of anti-fraud customers among visitors to the advertising media, and analysis of the performance of the advertising media. Steps to perform in real time
Advertising automation method using big data including.
In paragraph 1:
The real-time analysis module uses the advertising media visitor information to determine exposure advertisements, or uses the visitor's access information to predict the possibility of conversion and performs a function of recycling search and collection data of valid targets.
Advertising automation method using big data including more.
In clause 2, the step of collecting the data is,
Generating the dictionary segment for each target according to big data analysis;
Distinguishing between normal visitors and abnormal visitors using landing page visit data compared to clicks of the advertising media visitors & conversion/cancellation data after visiting the landing page; and
A step of verifying the efficiency of the advertisement in a specific advertising media by analyzing the media performance according to the proportion of meaningful exposure (Viewability) and exposure and clicks among the total exposure of the advertising material in the advertising media.
Advertising automation method using big data, including:
In claim 3, the dictionary segment is:
An advertising automation method using big data that groups visitors' age, gender, region, and purchasing ability according to preset criteria.
In claim 3, the step performed in real time is,
Analyzing the visitor using visitor information as soon as the advertising visitor visits and creating a target segment in real time; and
A step of determining the optimal advertisement to be exposed that matches the target segment and the visited advertising media.
Advertising automation method using big data including.
In claim 5, the step performed in real time is,
A step of analyzing visitor access information, including the visitor's visit time to advertising media, number of visits, number of visits to the landing page, and number of purchases, to identify the visitor's interest and purchase tendency.
Advertising automation method using big data including more.
In claim 6, the step performed in real time is,
A step of predicting the conversion possibility for each target segment of a specific visitor using the visitor information and the visitor access information or using big data.
Advertising automation method using big data including more.
In clause 7, the step of predicting the conversion possibility is,
Segment targets through big data analysis, select meaningful target groups, predict groups with high conversion potential, and target intensively.
Advertising automation method using big data.
In claim 8, the step performed in real time is:
A step of searching for valid advertisements that match the advertising media and the target segment of the visitor created according to the visitor's industry conversion possibility predicted by the conversion prediction unit.
Advertising automation method using big data including more.
In claim 9, the step performed in real time is,
According to the visitor information and visitor access information that accessed the advertising media, the steps of setting a new target segment for the visitor and selecting and displaying advertisements that are highly related to the content of the website are carried out.
Advertising automation method using big data including more.
In paragraph 1:
The communication module is centered around a DSP (Demand Side Platform) installed outside the advertising automation platform server, FDS (Fraud Detection System), and DMP (Data Management Platform). , steps to connect ADX (Ad Exchange), SSP (Supply Side Platform), and advertising media into a network.
Advertising automation method using big data including more.