system

Abstract

We provide the system. [Solution] A means of collecting data from a database, removing confidential information, and providing it to an AI, A means of obtaining untapped company information via API, analyzing the characteristics of the companies, and providing this information to AI, A means by which AI anticipates potential challenges for each company and selects and generates proposal templates based on those challenges, A method for having AI generate a proposal email based on the generated proposal document and sending it together with the proposal document, A system that includes means to track the open and viewed status of sent emails, enabling follow-up.

Description

Technical Field

[0001] The technology of the present disclosure relates to a system.

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When a company develops new customers, there is a problem that it is difficult to understand specific proposal contents and create appropriate proposal documents. Also, it takes time to create proposal documents, and often the created proposals do not meet the needs of customers. As a result, there is a problem that the efficiency of the company's sales activities decreases and it becomes difficult to acquire new customers. Furthermore, manual mass sending and follow-up are also inefficient, hindering the speed and effectiveness of the entire sales activities.

Means for Solving the Problems

[0005] In the present invention, in order to improve the creation of proposal documents and the development of new customers, the following means are provided.

[0006] First, a system is established to collect data from a database, remove confidential information, and provide it to the AI. Next, a system is established to acquire untapped company information via an API, analyze the characteristics of each company, and provide that information to the AI. Then, a system is established for the AI ​​to anticipate potential challenges for each company and select and generate proposal templates based on those challenges. Furthermore, a system is established to have the AI ​​generate proposal emails based on the generated proposals and send them together with the proposals. Finally, a system is provided to track the open and viewed status of sent emails, enabling follow-up. This will improve the efficiency of proposal creation, enable appropriate responses to customer needs, and enhance the speed and effectiveness of sales activities.

[0007] A "database" is a system for collecting and managing data, efficiently storing structured information and enabling its rapid retrieval and use.

[0008] "Confidential information" refers to important information that should only be used within a limited scope, both inside and outside the company, and should be protected from leakage and unauthorized access.

[0009] "AI" is an abbreviation for "Artificial Intelligence," and refers to technology that enables machines to replicate some or all of human intelligence. Specifically, it includes functions such as data analysis and generation, and pattern recognition.

[0010] "API" is an abbreviation for "Application Programming Interface," and refers to an interface for exchanging data and functions between different software systems.

[0011] A "template" refers to a standard format or template for efficiently performing a series of tasks, with specific formatting and content pre-configured.

[0012] A "proposal" is an official document that describes the features, benefits, and proposed implementation of a product or service, and is used to explain and persuade customers.

[0013] "Email" is a system for sending and receiving messages through electronic means of communication, and is a method of transmitting text and attachments over the internet.

[0014] "Follow-up" refers to reviewing the results and reactions of an activity that has already been carried out, and taking additional actions or responses as needed.

[0015] "Tracking" refers to the process of tracking and recording the progress and results of specific activities or events, and is a means of accurately understanding data and information. [Brief explanation of the drawing]

[0016] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9]Shows an emotion map to which a plurality of emotions are mapped. [Figure 10] Shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Mode for Carrying Out the Invention

[0017] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0018] First, the language used in the following description will be explained.

[0019] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of a plurality of arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of a plurality of types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0020] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0021] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0022] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0029] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0036] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0037] Modes for carrying out the invention

[0038] This invention relates to a system that utilizes AI and an integrated data platform to automatically generate proposals for acquiring new customers and to send proposal emails and materials in bulk. The embodiments for carrying out the invention will be described in detail below.

[0039] System Configuration

[0040] This system consists of a database, server, AI, API, and user terminals.

[0041] database

[0042] Accumulate past proposals and related documents.

[0043] server

[0044] Data acquisition and preprocessing

[0045] Customer data acquisition and analysis

[0046] AI-powered proposal generation and email creation.

[0047] Mass email sending

[0048] Situation tracking

[0049] AI

[0050] Learning for Proposal Writing

[0051] Anticipated challenges for each company

[0052] Automated proposal and email generation

[0053] API

[0054] Data integration with integrated data platforms (e.g., Compass, Domo)

[0055] User terminal

[0056] Review of proposal and email content

[0057] Execute follow-up actions

[0058] Operating Procedure

[0059] Step 1: Data Collection and Preprocessing

[0060] server

[0061] Past proposals and related documents are collected from the database, confidential information is removed, and then the data is preprocessed for AI training. Specifically, filtering is performed to reduce unnecessary information.

[0062] Specific example

[0063] The server scans the "Proposals" folder from the database and centrally collects all files. Then, it uses the Python pandas library to identify and filter data containing sensitive information.

[0064] Step 2: Gathering and analyzing information on untapped customers

[0065] server

[0066] Use APIs to retrieve information on untapped companies (industry, number of employees, sales, etc.) from an integrated data platform.

[0067] We analyze the characteristics of companies and provide that data to AI.

[0068] Specific example

[0069] The server sends a request to the integrated data platform's API to retrieve information about company A. Then, using the scikit-learn library, the retrieved information is clustered and identified as belonging to the "IT industry".

[0070] Step 3: Proposal Generation

[0071] server

[0072] The AI ​​anticipates the challenges specific to each company and selects a proposal template based on those challenges.

[0073] By embedding company-specific information into templates, customized proposals are automatically generated.

[0074] Specific example

[0075] The AI ​​selects a template related to "IT security," fills in specific information such as "number of employees at company A" and "sales revenue," and generates a proposal document tailored to company A.

[0076] Step 4: Generating the proposal email

[0077] server

[0078] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[0079] Specific example

[0080] The AI ​​generates a proposal email containing the message, "We propose security enhancements for company A." This email includes a proposal document attached in PDF format.

[0081] Step 5: Send out the proposal and email to everyone.

[0082] server

[0083] The proposal and proposal email will be sent via the mail server based on the mass mailing list.

[0084] Record transmission logs and manage status.

[0085] Specific example

[0086] The server sends the proposal and proposal email to "Company A" using the SMTP protocol. Upon successful transmission, the status is logged.

[0087] Step 6: Follow-up

[0088] User

[0089] You can check email open rates and proposal view rates through the Domo dashboard.

[0090] Send follow-up emails or schedule meetings as needed.

[0091] Specific example

[0092] The user confirms in the Domo dashboard that the email sent to "Company A" has been opened, and the next step is to schedule a follow-up meeting.

[0093] conclusion

[0094] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. In addition, tracking of sending status and follow-up functions improve the overall effectiveness and speed of sales activities.

[0095] The following describes the processing flow.

[0096] Step 1: Data Collection and Preprocessing

[0097] server

[0098] Collect past proposals and related documents from the database. Specifically, use the Python pandas library to execute appropriate queries and retrieve the data.

[0099] Remove confidential information from the acquired data. Filter based on specific keywords or patterns to remove personal information and corporate confidential information.

[0100] The pre-processed data is provided to the AI. The data is converted to a standard format such as CSV and saved as training data for the AI.

[0101] Step 2: Gathering and analyzing information on untapped customers

[0102] server

[0103] Obtain information on untapped companies via the integrated data platform's API. For example, send an HTTP request and receive company information (in JSON format).

[0104] Based on the acquired data, we will analyze the characteristics of the companies. We will use the scikit-learn library to perform clustering analysis and feature extraction.

[0105] The analysis results are provided to the AI. The company characteristics data is converted into a format that the AI ​​can use as reference data for generating proposals (e.g., a data frame format).

[0106] Step 3: Select and generate a proposal template

[0107] server

[0108] AI anticipates potential challenges for each company. Based on company characteristic data, it identifies challenges by comparing them with similar past cases.

[0109] Based on the task, a proposal template is selected. Proposal templates are provided in advance, and the AI ​​selects the appropriate one.

[0110] The template is used to embed company-specific information and generate customized proposals. Specific data (e.g., company revenue or challenges) is embedded in the template's placeholders.

[0111] Step 4: Generating the proposal email

[0112] server

[0113] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[0114] Select the necessary email template and fill in your company's specific content. This will generate customized proposal emails tailored to individual companies.

[0115] Combine the proposal email and proposal document into a single package. Attach the proposal document in PDF format to the email and set the email body accordingly.

[0116] Step 5: Send out the proposal and email to everyone.

[0117] server

[0118] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[0119] Record transmission logs for each recipient and manage their status. Save success / failure results to a log file for later review.

[0120] Step 6: Tracking email open and view rates

[0121] server

[0122] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[0123] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[0124] Step 7: Follow-up Action

[0125] User

[0126] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[0127] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[0128] (Example 1)

[0129] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0130] The traditional process of creating and sending proposals for acquiring new customers is time-consuming and labor-intensive. Furthermore, manual data collection, analysis, and email sending are inefficient and prone to human error. Additionally, the lack of follow-up and tracking after sending results in low effectiveness of sales activities.

[0131] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0132] This invention includes a server that collects data from a database, removes confidential information, and provides it to the AI; a server that acquires untapped company information via an API, analyzes the characteristics of the companies, and provides it to the AI; a server that anticipates potential challenges for each company and selects and generates proposal templates based on those challenges; a server that generates proposal emails based on the generated proposals and sends them together with the proposals; and a server that tracks the open and viewed status of the sent emails to enable follow-up. This enables the automatic generation and mass sending of proposals and effective follow-up.

[0133] (List of definitions)

[0134] A "database" is a storage device used to accumulate past proposals and related documents.

[0135] "AI" refers to an artificial intelligence system designed for automated decision-making and proposal generation.

[0136] An "API" stands for Application Programming Interface, which is used to exchange data between external systems.

[0137] "Untapped company information" refers to data about potential new customer companies, including industry, number of employees, and sales revenue.

[0138] Clustering analysis is a process that uses the scikit-learn library to classify, analyze, and group company data.

[0139] A "template" is a pre-defined form used by AI to generate proposals, and it is a format for embedding company-specific information.

[0140] A "proposal" is a document that outlines a customized proposal tailored to a specific company.

[0141] A "proposal email" is a message used to send the contents of a generated proposal as an email.

[0142] An "SMTP server" is a server used to send emails using the Simple Mail Transfer Protocol.

[0143] The "Domo Dashboard" is a visualization tool for tracking email open rates and proposal view rates.

[0144] "Follow-up" refers to the process of taking additional actions or making contact regarding submitted proposals or emails.

[0145] "Filtering" is the process of removing information from data that should not be released externally.

[0146] Modes for carrying out the invention

[0147] This invention relates to a system for automatically generating and efficiently sending proposals and emails for acquiring new customers. In particular, it aims to automate proposal creation, analyze company characteristics, and improve the efficiency of email sending by utilizing AI and an integrated data platform. The specific methods for realizing this system are described below.

[0148] This system consists of a database, servers, AI, APIs, and user terminals.

[0149] database

[0150] A storage device is used to accumulate past proposals and related documents. This allows the AI ​​to learn from past proposal data and generate more accurate proposals.

[0151] server

[0152] The server is responsible for several processes, including the following:

[0153] 1. Data acquisition and preprocessing:

[0154] The server collects proposals and related documents from the database, filters the data using the Python pandas library, and removes confidential information.

[0155] Specific example: The server executes an SQL query against the database to retrieve proposal data. This data is then read using the pandas library, and confidential information is filtered out.

[0156] 2. Collection and analysis of untapped company information:

[0157] The server uses an API to retrieve information on untapped companies from an integrated data platform (e.g., Compass, Domo). This information includes industry, number of employees, and revenue.

[0158] The server uses the scikit-learn library to perform clustering analysis on company characteristics and provides the results to the AI.

[0159] Specific example: A server sends an API request to retrieve company characteristic data. Clustering analysis is performed using the scikit-learn library to identify company characteristics.

[0160] 3. Proposal generation:

[0161] The AI ​​anticipates the challenges specific to each company and selects a proposal template based on those challenges. It then embeds company-specific information into the template to generate a customized proposal.

[0162] Specific example: The AI ​​selects an appropriate template from a template database and uses the Jinja2 template engine to embed company information.

[0163] 4. Generating and sending proposal emails:

[0164] Based on the proposal document, the AI ​​generates a proposal email. The generated email includes a summary of the proposal and the reasons for the proposal.

[0165] The server uses the SMTP protocol to send proposal emails and proposal documents in bulk. It records transmission logs and manages the status.

[0166] Specific example: AI uses an email template to generate an email with a proposal PDF attached. The server sends the email via SMTP and records the sending log.

[0167] User terminal

[0168] The user terminal is a device used to review proposals and email content, and to perform follow-up actions. The Domo dashboard allows users to check email open rates and proposal viewing rates, and to send follow-up emails or schedule meetings as needed.

[0169] Example prompt statements

[0170] Data collection:

[0171] Write Python code to scan the 'Proposals' folder in the database and collect all files. Then, use pandas to identify and remove columns containing sensitive information.

[0172] Collection and analysis of corporate information:

[0173] "Write Python code that retrieves information from company A's API and performs clustering analysis using scikit-learn."

[0174] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. In addition, tracking of sending status and follow-up functions improve the overall effectiveness and speed of sales activities.

[0175] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0176] Program processing flow

[0177] Step 1: Data Collection and Preprocessing

[0178] Input: Database containing past proposals and related documents.

[0179] Output: Pre-processed data with sensitive information removed, ready to be provided to the AI.

[0180] Specific actions:

[0181] 1. The server connects to the database and uses SQL queries to collect proposal data.

[0182] 2. After obtaining the proposal data, the data is filtered using the Python pandas library to remove confidential information.

[0183] 3. Save the filtered data to provide it to the AI ​​model.

[0184] Specific data processing:

[0185] The server extracts proposal data from the database using SQL queries. This data is then read using pandas, and specific columns or values ​​are filtered to remove sensitive information. Finally, the pre-processed data is saved as a file.

[0186] Specific example:

[0187] The server downloads all files in the "Proposal" folder at once. The pandas drop function is used to remove columns containing confidential information.

[0188] Step 2: Gathering and analyzing information on untapped companies

[0189] Input: Company information obtained via API

[0190] Output: Analyzed company characteristics data

[0191] Specific actions:

[0192] 1. The server sends a request to the integrated data platform's API to retrieve data on untapped companies.

[0193] 2. Perform clustering analysis on the acquired data using the scikit-learn library.

[0194] 3. Save the analysis results to the database and prepare them for provision to the AI.

[0195] Specific data processing:

[0196] The server uses an API to retrieve company data (e.g., industry, number of employees, sales). This data is then analyzed using scikit-learn's clustering algorithm to group company characteristics. The analysis results are saved for use with the AI.

[0197] Specific example:

[0198] The server retrieves data from company A via an API request, performs clustering analysis using the scikit-learn KMeans algorithm, and identifies company A's industry.

[0199] Step 3: Proposal Generation

[0200] Input: Company characteristics data and past proposal templates

[0201] Output: Customized proposal

[0202] Specific actions:

[0203] 1. The server uses AI to anticipate the challenges specific to each company and selects a proposal template based on those challenges.

[0204] 2. Create customized proposals by embedding company-specific information into proposal templates.

[0205] Specific data processing:

[0206] The AI ​​selects an appropriate proposal template based on company characteristics data and uses the Jinja2 template engine to embed company information.

[0207] Specific example:

[0208] The AI ​​selects an "IT security" template from the template database and fills in specific information about company A (number of employees, sales revenue).

[0209] Step 4: Generating the proposal email

[0210] Input: Generated proposal

[0211] Output: Proposal email

[0212] Specific actions:

[0213] 1. The AI ​​analyzes the generated proposal and creates a proposal email.

[0214] 2. Attach the proposal PDF file to the email you created.

[0215] Specific data processing:

[0216] The AI ​​extracts necessary information from the generated proposal and inserts it into an email template. The proposal PDF is then attached to the email to complete the process.

[0217] Specific example:

[0218] The AI ​​generates an email stating, "We propose security enhancements for company A," and attaches a PDF of the proposal.

[0219] Step 5: Send out the proposal and email to everyone.

[0220] Input: Proposal email and proposal PDF

[0221] Output: Recording of transmission logs and status information

[0222] Specific actions:

[0223] 1. The server uses the SMTP protocol to send proposal emails and proposal PDFs simultaneously.

[0224] 2. Log the status and manage it when the transmission is complete.

[0225] Specific data processing:

[0226] Send emails via the SMTP server. The sending status and details of the email are recorded in the sending log.

[0227] Specific example:

[0228] The server uses SMTP to send a proposal document and proposal email to company A, and records this in the transmission log.

[0229] Step 6: Follow-up

[0230] Input: Email open status and proposal viewing status

[0231] Output: Follow-up actions (e.g., sending an email, scheduling a meeting)

[0232] Specific actions:

[0233] 1. Users can check email open rates and proposal view rates through the Domo dashboard.

[0234] 2. Send follow-up emails or schedule meetings as needed.

[0235] Specific data processing:

[0236] Check the open and view rates on the Domo dashboard and take the next action (send a follow-up email, schedule a meeting).

[0237] Specific example:

[0238] The user confirms on the Domo dashboard that "the email sent to company A has been opened" and sets up a follow-up meeting.

[0239] (Application Example 1)

[0240] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0241] Traditional customer proposal creation systems required manual proposal creation and customer information analysis, which was time-consuming and labor-intensive. Furthermore, the process of generating personalized proposals was cumbersome, hindering the efficiency of new customer acquisition. Additionally, there was a lack of management functions to track the effectiveness of generated proposals and provide appropriate follow-up.

[0242] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0243] This invention includes a server that provides data to an AI after collecting information from a database and removing confidential information; a server that provides data to an AI after obtaining untapped company information via an API, analyzing the characteristics of the companies, and providing this information to the AI; a server that allows the AI ​​to anticipate potential challenges for each company and select and generate proposal templates based on those challenges; a server that allows the AI ​​to generate proposal emails based on the generated proposals and sends them together with the proposals; a server that tracks the open and viewed status of sent emails to enable follow-up; a server that collects customer information from a sales site, analyzes it based on past purchase history and visit data, and generates customized product proposals as part of the proposals; and a server that generates personalized proposal emails based on the customized product proposals, sends them all at once via a mail server, and provides in-app notifications. This makes it possible to automatically and efficiently generate personalized proposals and proposal emails for new customers and send them all at once. In addition, it is possible to track the open and viewed status of emails and perform effective follow-up.

[0244] A "database" is a data storage system that centrally stores information such as past documents and proposals, and allows users to retrieve data as needed.

[0245] "Confidential information" refers to non-public information about a company or its customers that should not be leaked to the outside.

[0246] "AI" is an abbreviation for artificial intelligence, which refers to technologies that automate or optimize specific tasks through data analysis and machine learning.

[0247] "API" stands for Application Programming Interface, which is an interface for exchanging data and functions between different software programs.

[0248] "Untapped company information" refers to information about companies with which no business transactions have yet been concluded, including industry, number of employees, and sales figures.

[0249] A "template" is a basic model used when creating documents such as proposals or emails, and it includes a specific format and content.

[0250] A "proposal" is a document that formalizes a proposal made to a company or customer, and it includes details of the proposal, the reasons for it, and specific benefits.

[0251] A "proposal email" is an email sent to a customer to which a proposal and related information are presented. It is an email that outlines the proposal and its benefits.

[0252] A "sales site" refers to a website or online platform used for selling products.

[0253] "Customer information" refers to all information about a customer, such as purchase history and website visit history.

[0254] "Customized product recommendations" refer to suggestions tailored to the customer's needs, based on their purchase history and visit data, to propose the most suitable products for that customer.

[0255] A "personalized suggestion email" is a suggestion email that contains content that is individually customized for a specific customer.

[0256] "Follow-up" refers to ongoing support and tracking activities conducted after the initial proposal, primarily to check email open rates and the effectiveness of the proposed content.

[0257] This invention relates to a system for automatically generating personalized product proposals and sending them out via email and accompanying materials to customers on a sales website in order to acquire new customers. The embodiments for carrying out the invention will be described in detail below.

[0258] System Configuration

[0259] This system consists of a database, server, AI, API, and user terminals.

[0260] database

[0261] The system stores information such as past proposals, purchase history, and visit data.

[0262] server

[0263] Data acquisition and preprocessing

[0264] Customer data acquisition and analysis

[0265] AI-powered proposal generation and email creation.

[0266] Mass email sending and in-app notifications

[0267] Situation tracking

[0268] AI

[0269] Learning for Proposal Writing

[0270] Anticipating challenges for each customer

[0271] Automated proposal and email generation

[0272] API

[0273] Data integration with the integrated data platform

[0274] User terminal

[0275] Review of proposal and email content

[0276] Execute follow-up actions

[0277] Program processing

[0278] The server first collects past purchase history and visit data from the database and performs preprocessing. Specifically, it uses the Python pandas library to format the data and filters out sensitive information. Next, it retrieves untapped customer information via the integrated data platform's API. The collected customer information is classified using clustering analysis with the scikit-learn library.

[0279] The AI ​​selects a suitable product suggestion template for each categorized customer and generates customized product suggestions based on their purchase history and interests. During this process, a generation AI model is used to generate prompts and create specific suggestion content. The generated suggestion emails are sent to customers simultaneously via a mail server, along with in-app notifications.

[0280] Hardware and software

[0281] Hardware:

[0282] server

[0283] User terminal

[0284] software:

[0285] Python

[0286] pandas

[0287] scikit-learn

[0288] smtplib

[0289] requests <00​​​​​​​​​​​​​​​​​​​​​​​​​​​​ The server collects past purchase histories and visit data from the database. Specifically, it reads and formats these data using the pandas library in Python. Since the collected data may contain confidential information, filtering is performed to remove the confidential information.

[0298] Input: Purchase history and visit data in the database

[0299] Output: A dataset that has been formatted and from which confidential information has been removed

[0300] Step 2:

[0301] The server obtains unexploited customer information via the API of the integrated data platform. Specifically, it sends an API request to obtain information such as the customer's industry, number of employees, and sales volume. Then, it reads the obtained data using the pandas library and converts it into an appropriate format. [[ID=I9]]

[0302] Input: API request of the integrated data platform

[0303] Output: Dataset of characteristic information of unexploited customers [[ID=I7]]

[0304] Step 3:

[0305] The server performs clustering analysis on the obtained customer information using the scikit-learn library. Through the clustering analysis, the customers are classified into clusters based on their characteristics.

[0306] Input: Dataset of characteristic information of unexploited customers

[0307] Output: Customer data with cluster labels assigned

[0308] Step 4:

[0309] The AI ​​selects the optimal proposal template for each customer based on cluster labels. After selecting a template, it automatically generates a customized proposal by embedding customer characteristic information.

[0310] Input: Customer data with cluster labels assigned.

[0311] Output: Customized proposal

[0312] Step 5:

[0313] The AI ​​creates a personalized proposal email based on the generated proposal. The proposal email includes a summary of the proposal and the reasons for the proposal.

[0314] Input: Customized proposal

[0315] Output: Personalized suggestion email

[0316] Step 6:

[0317] The server sends proposals and proposal emails via the mail server based on a mass mailing list. It also simultaneously sends in-app notifications. It records sending logs and manages the status.

[0318] Input: Personalized suggestion email

[0319] Output: Sending log, in-app notification status

[0320] Step 7:

[0321] Users can check email open rates and proposal viewing rates through the dashboard. They can also send follow-up emails or schedule meetings as needed.

[0322] Input: Sending log, in-app notification status

[0323] Output: Follow-up actions, such as follow-up emails or meeting scheduling.

[0324] Through the above processing steps, this system can automatically and efficiently generate personalized proposals and emails for new customers and send them in bulk. Furthermore, it can track email open and view rates, enabling effective follow-up.

[0325] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0326] Modes for carrying out the invention

[0327] This invention relates to a system that utilizes AI, an integrated data platform, and an emotion engine to automatically generate proposals and emails for acquiring new customers, and further customizes them based on the user's emotions. The embodiments for carrying out the invention will be described in detail below.

[0328] System Configuration

[0329] This system consists of a database, server, AI, API, emotion engine, and user terminal.

[0330] database

[0331] It stores past proposals, related documents, and user sentiment data.

[0332] server

[0333] Data acquisition and preprocessing

[0334] Customer data acquisition and analysis

[0335] AI-powered proposal generation and email creation.

[0336] Emotional engine integration

[0337] Mass email sending

[0338] Situation tracking

[0339] AI

[0340] Learning for Proposal Writing

[0341] Anticipated challenges for each company

[0342] Automated proposal and email generation

[0343] API

[0344] Data integration with integrated data platforms (e.g., Compass, Domo)

[0345] Emotional Engine

[0346] We analyze user emotions and provide emotion data in real time. Based on this emotion data, we customize proposals and email content.

[0347] User terminal

[0348] Review of proposal and email content

[0349] Execute follow-up actions

[0350] Operating Procedure

[0351] Step 1: Data Collection and Preprocessing

[0352] server

[0353] Past proposals and related documents are collected from the database, confidential information is removed, and then the data is preprocessed for AI training. For example, filtering is performed to reduce unnecessary information.

[0354] The emotion engine is used to analyze user sentiment data regarding past suggestion emails and store it in a database.

[0355] Specific example

[0356] The server scans the "Proposals" folder from the database and retrieves the data using the Python pandas library. It filters out sensitive information and provides the preprocessed data to the AI. Additionally, an emotion engine analyzes user responses in past emails and stores the emotion data in the database.

[0357] Step 2: Gathering and analyzing information on untapped customers

[0358] server

[0359] We will use an API to retrieve information on untapped companies from an integrated data platform. Specifically, we will send an HTTP request and receive company information (in JSON format).

[0360] Based on the acquired data, we analyze the characteristics of the company. For example, we perform clustering analysis using the scikit-learn library.

[0361] The analysis results are provided to the AI. The company characteristics data is used by the AI ​​as reference data for generating proposals.

[0362] Specific example

[0363] The server sends a request to the integrated data platform's API to retrieve information about company A. Then, using scikit-learn, it identifies that company A belongs to the "IT industry".

[0364] Step 3: Select and generate a proposal template

[0365] server

[0366] The AI ​​anticipates potential challenges for each company and selects a proposal template based on those challenges. The optimal template is chosen based on company characteristic data.

[0367] The system automatically generates customized proposals by embedding company-specific information into proposal templates.

[0368] Using an emotion engine, template selection and proposal content are adjusted based on the user's past emotional data.

[0369] Specific example

[0370] The AI ​​selects a template related to "IT security" and fills in specific information about company A, such as "number of employees" and "sales revenue." At the same time, the emotion engine refers to past emotion data and adjusts the content to elicit a more positive response.

[0371] Step 4: Generating the proposal email

[0372] server

[0373] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[0374] Select the necessary email template, embed company-specific content, and customize the email content by reflecting sentiment data analyzed by the sentiment engine.

[0375] Combine the proposal email and proposal document into a single package.

[0376] Specific example

[0377] The AI ​​generates a proposal email containing the message, "We propose strengthening security for company A," and attaches the proposal as a PDF document. The emotion engine adjusts the wording and tone of the email to suit the user's preferences.

[0378] Step 5: Send out the proposal and email to everyone.

[0379] server

[0380] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[0381] Record transmission logs for each recipient and manage their status.

[0382] Specific example

[0383] The server uses the SMTP protocol to send the proposal and proposal email to "Company A," and logs the status upon successful transmission.

[0384] Step 6: Tracking email open and view rates

[0385] server

[0386] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[0387] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[0388] Specific example

[0389] The server embeds tracking pixels in emails to detect open rates, which are then displayed in real time on the Domo dashboard.

[0390] Step 7: Follow-up Action

[0391] User

[0392] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[0393] Based on the data analyzed by the emotion engine, the next follow-up action will be suggested.

[0394] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[0395] Specific example

[0396] The user confirms in the Domo dashboard that the email sent to "Company A" was opened, and based on the sentiment engine's suggestions, sets up a follow-up meeting as the next step.

[0397] conclusion

[0398] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. Tracking and follow-up features improve the overall effectiveness and speed of sales activities. In addition, the integration of an emotion engine enables customization based on user emotions, further enhancing the accuracy and effectiveness of proposals.

[0399] The following describes the processing flow.

[0400] Step 1: Data Collection and Preprocessing

[0401] server

[0402] Collect past proposals and related documents from the database. Specifically, use the Python pandas library to execute appropriate queries and retrieve the data.

[0403] Remove confidential information from the acquired data. For example, use regular expressions based on specific keywords or patterns to filter out confidential information and remove personal and corporate confidential information.

[0404] The pre-processed data is provided to the AI. The data is converted to a standard format such as CSV and saved as a training dataset for the AI.

[0405] In addition, an emotion engine is used to analyze users' emotional data regarding past suggestion emails and store it in a database. The emotion engine uses natural language processing technology to analyze the content of emails and assigns emotional labels such as positive, negative, and neutral.

[0406] Step 2: Gathering and analyzing information on untapped customers

[0407] server

[0408] We will use an API to retrieve information on untapped companies from an integrated data platform. Specifically, we will send an HTTP request and receive company information (in JSON format).

[0409] Based on the acquired data, we analyze the characteristics of companies. For example, we use the scikit-learn library to perform clustering analysis and feature extraction to reveal characteristics such as the company's industry, size, and sales revenue.

[0410] The analysis results are provided to the AI. The company characteristics data is converted into a format that the AI ​​can use as reference data for generating proposals (e.g., a data frame format).

[0411] Step 3: Select and generate a proposal template

[0412] server

[0413] AI anticipates potential challenges for each company. Based on company characteristic data, it compares it with similar past cases and industry trends to identify the most relevant issues.

[0414] Based on the selected issue, a proposal template is chosen. Proposal templates are prepared in advance, and the AI ​​selects the appropriate one.

[0415] The template is used to embed company-specific information and generate customized proposals. Specific data (e.g., company revenue or challenges) is embedded in the template's placeholders.

[0416] Using an emotion engine, template selection and proposal content are adjusted based on the user's past emotional data. For example, expressions and designs that received many positive responses are prioritized.

[0417] Step 4: Generating the proposal email

[0418] server

[0419] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[0420] Select the necessary email template and fill in your company's specific content. The sentiment engine analyzes emotional data to customize the email content for greater effectiveness. For example, it adopts phrasing and tones that elicit positive responses from users.

[0421] The proposal email and proposal document will be packaged together, with the proposal document attached to the email in PDF format.

[0422] Step 5: Send out the proposal and email to everyone.

[0423] server

[0424] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[0425] Record transmission logs for each recipient and manage their status. Save success / failure results to a log file for later review.

[0426] Step 6: Tracking email open and view rates

[0427] server

[0428] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[0429] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[0430] Step 7: Follow-up Action

[0431] User

[0432] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[0433] Based on the data analyzed by the emotion engine, the next follow-up action is suggested. For example, if the user showed a positive response to the previous email, a more assertive approach is suggested.

[0434] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[0435] (Example 2)

[0436] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0437] This invention relates to a system that significantly reduces the time and effort required for acquiring new customers in a company's sales activities and makes proposals to customers more effective. Conventional systems require a great deal of time and effort to create proposals and generate proposal emails to customers, and they are not sufficiently customized based on customer sentiment. Therefore, there has been a need for a system that supports new customer acquisition in an efficient and effective way.

[0438] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0439] This invention includes a server that provides data to an AI after collecting information from a database and removing confidential information; a server that provides data to an AI after obtaining untapped company information via an API and analyzing the characteristics of the companies and providing this information to the AI; a server that allows the AI ​​to anticipate potential challenges for each company and select and generate proposal templates based on those challenges; a server that allows the AI ​​to generate proposal emails based on the generated proposals and sends them together with the proposals; a server that tracks the open and viewed status of sent emails and enables follow-up; a server that uses an emotion analysis engine to accumulate past emotional data of users and customize the content of proposals and proposal emails; a server that sends proposal emails and proposals all at once; a server that uses a Domo dashboard to display the sending status in real time; and a server that suggests follow-up actions. This enables the efficient and effective generation and sending of proposals and proposal emails, and makes it easier to realize optimal proposals based on the user's emotions.

[0440] "Document collection" refers to the act of obtaining past proposals and related documents from a database.

[0441] "Removal of confidential information" is the process of filtering and removing information from collected materials that should not be released to the public.

[0442] "Means of providing data to AI" refers to the process of inputting pre-processed data into a machine learning system.

[0443] "Acquiring company information" means collecting data on untapped companies via APIs.

[0444] "Company characteristic analysis" refers to evaluating the characteristics of each company based on acquired company information using methods such as clustering analysis.

[0445] "Identifying challenges" refers to AI predicting potential problems and areas for improvement specific to each company.

[0446] "Selecting a proposal template" is a method of choosing the most suitable proposal format based on the anticipated challenges.

[0447] "Proposal generation" refers to the process of creating a customized proposal by embedding company-specific information into a template.

[0448] "Proposal email generation" refers to the process where AI creates an email based on the generated proposal document, outlining the proposal's summary and reasons.

[0449] "Mass email sending" refers to the process of simultaneously sending a generated proposal document and proposal email to multiple recipients based on a list.

[0450] "Open tracking" refers to tracking and recording whether or not an email that has been sent has been opened.

[0451] "Follow-up action" refers to the act of conducting follow-up surveys or making additional contact with users after sending a proposal email, depending on their situation.

[0452] A "sentiment analysis engine" is a system that analyzes a user's past emotional data and optimizes the suggested content based on that data.

[0453] The "Domo Dashboard" is a tool for visualizing data and displaying real-time tracking information.

[0454] This invention relates to a system that utilizes AI, an integrated data platform, and an emotion analysis engine to automatically generate proposals and emails for acquiring new customers, and further customizes them based on the user's emotions. The detailed configuration and operation of the system are described below.

[0455] System Configuration

[0456] This system consists of a database, server, AI, API, sentiment analysis engine, and user terminal.

[0457] database

[0458] It stores past proposals, related documents, and user sentiment data.

[0459] server

[0460] It performs data collection and preprocessing, customer data acquisition and analysis, AI-powered proposal generation and email creation, sentiment analysis engine integration, mass email sending, and status tracking.

[0461] AI

[0462] This system provides learning opportunities for proposal creation, anticipates challenges specific to each company, and automatically generates proposals and emails.

[0463] API

[0464] Establish data integration with integrated data platforms (e.g., Compass, Domo).

[0465] Emotion analysis engine

[0466] We analyze user emotions and provide emotion data in real time. Based on this emotion data, we customize proposals and email content.

[0467] User terminal

[0468] Review the proposal and email content, and take follow-up actions.

[0469] Operating Procedure

[0470] The following explains how the system works using specific examples for each step.

[0471] Step 1: Data Acquisition and Preprocessing

[0472] server

[0473] The server collects past proposals and related documents from the database and removes confidential information. This process uses the Python pandas library to load, filter, and clean the data. It also uses a sentiment analysis engine to analyze user sentiment data regarding past proposal emails and stores it in the database.

[0474] Specific example

[0475] The server scans the "Proposal" folder from the database, retrieves data using the pandas library, filters it, and provides the preprocessed data to the AI. The sentiment analysis engine analyzes user responses contained in past emails and stores them in the database as sentiment data.

[0476] Step 2: Gathering and analyzing information on untapped customers

[0477] server

[0478] The server uses an API to retrieve information on untapped companies from the integrated data platform. Specifically, it sends an HTTP request and receives company information in JSON format. Then, it uses the scikit-learn library to perform clustering analysis, extract company characteristics, and provide them to the AI.

[0479] Specific example

[0480] The server sends an HTTP request to the integrated data platform's API to retrieve information about company A, and then uses scikit-learn to identify that company A belongs to the "IT industry".

[0481] Step 3: Select and generate a proposal template.

[0482] server

[0483] The AI ​​anticipates potential challenges for each company and selects proposal templates based on those challenges. Based on company characteristic data, it selects the optimal template, embeds company-specific information into the template, and automatically generates a customized proposal. In addition, it uses an emotion analysis engine to adjust template selection and proposal content based on the user's past emotion data.

[0484] Specific example

[0485] The AI ​​selects a template related to "IT security" and fills in information such as company A's "number of employees" and "sales revenue." Simultaneously, an emotion analysis engine refers to past emotion data and adjusts the content to elicit a positive response from the user.

[0486] Step 4: Generate the proposal email

[0487] server

[0488] Based on the generated proposal document, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its proposal, and its content is customized using a sentiment analysis engine. The proposal email and proposal document are then combined into a single package.

[0489] Specific example

[0490] The AI ​​generates a proposal email containing the message, "We propose strengthening security for company A," and attaches the proposal as a PDF document. An emotion analysis engine adjusts the wording and tone of the email to suit the user's preferences.

[0491] Step 5: Send proposals and emails to all recipients.

[0492] server

[0493] The server uses the SMTP protocol to send proposals and proposal emails in bulk. Specifically, it uses the smtplib library to send emails based on a mass mailing list. It also records a transmission log and manages the transmission status.

[0494] Specific example

[0495] The server uses the SMTP protocol to send the proposal and proposal email to "Company A," and logs the status upon successful transmission.

[0496] Step 6: Track email open and view rates

[0497] server

[0498] The server uses embedded tracking pixels to track email open rates and proposal viewing rates. The collected tracking data is reflected in the Domo dashboard in real time.

[0499] Specific example

[0500] The server embeds tracking pixels in emails to detect open rates, which are then displayed in real time on the Domo dashboard.

[0501] Step 7: Follow-up Action

[0502] User

[0503] Users can check email open rates and proposal view rates through the Domo dashboard. Based on the data analyzed by the sentiment analysis engine, the system suggests the next follow-up actions. Users can then take additional follow-up actions as needed, such as sending follow-up emails or scheduling meetings.

[0504] Specific example

[0505] The user confirms on the Domo dashboard that "Company A" has opened the proposal, and then sets up a follow-up meeting based on the sentiment analysis engine's recommendations.

[0506] Examples of prompt statements

[0507] Select a proposal template and generate a proposal tailored to Company A. Adjust the content based on past sentiment data and output the optimal proposal in PDF format. Also, send the proposal and proposal email based on the mass email mailing list and track the open rates.

[0508] As described above, this system utilizes AI and emotion analysis engines to enable the automated generation and customization of efficient and effective proposals and emails for new customers, and further optimizes sales activities through tracking of sending status and suggesting follow-up actions.

[0509] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0510] Step 1: Data Collection and Preprocessing

[0511] server

[0512] The server collects past proposals and related documents from the database. It executes SQL queries against the database to read the relevant data. Then, it removes confidential information from the collected data. Specifically, it uses the Python pandas library to filter and clean the data. It also uses a sentiment analysis engine to analyze past proposal email data, extract user sentiment data, and store it in the database.

[0513] input

[0514] Proposals, related documents, and past proposal email data collected from the database.

[0515] output

[0516] Confidential information has been removed, and the proposal data has been pre-processed and analyzed for sentiment.

[0517] Specific actions

[0518] The server retrieves data from the database, filters the necessary data using the pandas library, removes sensitive information, and performs preprocessing.

[0519] Step 2: Gathering and analyzing information on untapped customers

[0520] server

[0521] The server uses an API to retrieve information on untapped companies from an integrated data platform. It sends an HTTP request and receives company information in JSON format. Then, it performs clustering analysis using the scikit-learn library to extract company characteristics. The characteristic data is analyzed and provided to the AI.

[0522] input

[0523] Information on untapped companies obtained via API (in JSON format).

[0524] output

[0525] Company characteristic data extracted through clustering analysis.

[0526] Specific actions

[0527] The server sends an API request to retrieve information on untapped companies. This data is then clustered and analyzed using scikit-learn to extract company characteristics.

[0528] Step 3: Select and generate a proposal template

[0529] server

[0530] The AI ​​anticipates potential challenges for each company and selects a proposal template based on those challenges. It uses company characteristic data to choose the optimal proposal template and embeds company-specific information into it. Furthermore, it uses an emotion analysis engine to adjust the proposal content based on the user's past emotional data.

[0531] input

[0532] Company characteristic data and historical sentiment data.

[0533] output

[0534] A customized proposal tailored to each company.

[0535] Specific actions

[0536] AI analyzes the data and selects the optimal proposal template. An emotion analysis engine then adjusts the content.

[0537] Step 4: Generating the proposal email

[0538] server

[0539] Based on the generated proposal, the AI ​​creates a proposal email. Using the Jinja2 template engine, it generates an email containing an overview of the proposal and the reasons for the proposal. Using an emotion analysis engine, the email content is customized based on the user's emotion data.

[0540] input

[0541] Generated proposals, past sentiment data.

[0542] output

[0543] Customized proposal email.

[0544] Specific actions

[0545] The AI ​​generates emails based on the proposal, and the sentiment analysis engine adjusts the content.

[0546] Step 5: Send out the proposal and email to everyone.

[0547] server

[0548] The server uses the SMTP protocol to send proposals and proposal emails in bulk. It uses the smtplib library to send emails based on a designated mailing list. It records sending logs and manages the sending status.

[0549] input

[0550] Customized proposal emails, proposal documents, and mass mailing lists.

[0551] output

[0552] Log and transmission status after transmission is complete.

[0553] Specific actions

[0554] The server sends an email using the SMTP protocol and logs the status.

[0555] Step 6: Tracking email open and view rates

[0556] server

[0557] The server uses embedded tracking pixels to track email open rates and proposal view rates. The collected data is reflected in the Domo dashboard in real time.

[0558] input

[0559] A proposal email with embedded tracking pixels.

[0560] output

[0561] Tracking data on opening and browsing activity.

[0562] Specific actions

[0563] The server uses tracking pixels to detect open and viewed status and displays it on the Domo dashboard.

[0564] Step 7: Follow-up Action

[0565] User

[0566] Through the Domo dashboard, users can check email open rates and proposal view rates. Based on the data analyzed by the sentiment analysis engine, the system suggests the next follow-up actions. Users can then send follow-up emails or schedule meetings as needed.

[0567] input

[0568] Data on opening and viewing status, and analysis data from the sentiment analysis engine.

[0569] output

[0570] Follow-up actions (e.g., sending a follow-up email or scheduling a meeting).

[0571] Specific actions

[0572] Users check the situation on the Domo dashboard and take follow-up actions based on suggestions from the sentiment analysis engine.

[0573] (Application Example 2)

[0574] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0575] Traditional methods of acquiring new customers made it difficult to process large amounts of customer data quickly and efficiently, and to make appropriate proposals to each customer. Furthermore, customizing proposals based on customer emotions and feedback was challenging, resulting in insufficient engagement with customers. This led to decreased customer engagement and difficulties in acquiring new customers.

[0576] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting data from a database, removing confidential information and providing it to the AI; means for acquiring untapped company information via an API, analyzing the characteristics of the companies and providing it to the AI; means for the AI ​​to anticipate potential challenges for each company and select and generate a proposal template based on those challenges; means for having the AI ​​generate a proposal email based on the generated proposal and sending it together with the proposal; means for tracking the open and viewing status of sent emails to enable follow-up; means for collecting data such as past purchase history and feedback and analyzing sentiment data; means for customizing the content of the proposal and proposal email based on the sentiment data; and means for generating proposal emails and coupons for individual customers using a generation AI model. This makes it possible to efficiently process large amounts of data and make customized proposals based on customer sentiment.

[0577] A "database" is an information system that stores and manages data such as past documents and feedback.

[0578] "Confidential information" refers to important information that must not be leaked to outsiders.

[0579] "AI" refers to artificial intelligence, a technology that processes and analyzes large amounts of data to automatically generate suggestions.

[0580] "API" stands for Application Programming Interface, which is a gateway for different software systems to exchange data and functions.

[0581] "Company characteristics" refer to the characteristics of a company, such as its industry, size, and profitability.

[0582] A "proposal" is a document that outlines the proposed content for a specific client.

[0583] A "template" is a format that serves as a guide for efficiently creating proposals, emails, and other documents.

[0584] A "proposal email" is an email sent to a client that includes an overview of the proposal and the reasons for the proposal.

[0585] "Tracking" is the process of tracking and recording things like whether an email was opened or whether links were clicked after it was sent.

[0586] "Follow-up" refers to additional actions taken after sending a proposal email in order to maintain and deepen the relationship with the customer.

[0587] "Purchase history" refers to a record of products that a customer has purchased in the past.

[0588] "Feedback" refers to the opinions and evaluations received from customers.

[0589] "Emotional data" refers to data that analyzes customer emotions and reactions and expresses them as numerical or textual data.

[0590] A "generative AI model" is a model trained to automatically generate documents and suggestions using artificial intelligence.

[0591] A "coupon" is an electronic or paper voucher used to offer discounts or benefits to customers.

[0592] This invention relates to a new customer acquisition system comprising a database, server, AI, API, emotion engine, and user terminal. This system automatically generates proposals and proposal emails and customizes them based on user emotion data through the following steps.

[0593] The server collects data such as historical documents and feedback from the database, and preprocesses it by removing confidential information. It uses the Python pandas library to retrieve data and filter out confidential information. It also uses a sentiment engine to analyze user responses in past emails and generate sentiment data. IBM Watson® Tone Analyzer is used as the sentiment engine.

[0594] Using an API, the server retrieves new customer information from an integrated data platform (e.g., Domo). It sends an HTTP request to obtain company information (in JSON format) and analyzes the company's characteristics. Clustering analysis is performed using the scikit-learn library. This characteristic data is used as reference when an AI (e.g., OpenAI® GPT-4®) generates a proposal.

[0595] Using the GPT-4 generative AI model, the AI ​​anticipates potential challenges specific to each company and selects and generates the optimal proposal template based on those challenges. It then customizes the template based on sentiment data while incorporating company-specific information (such as the number of employees and sales). This process generates more effective proposals that reflect the user's emotions.

[0596] Next, the AI ​​creates a proposal email based on the generated proposal. Using sentiment data provided by the sentiment engine, it adjusts the email content to maximize customer response. This email includes a summary of the proposal and the reasons for the proposal, and is sent all at once. The server uses the SMTP protocol to send the email to all recipients simultaneously.

[0597] Tracking pixels are embedded in emails to track email open rates and link click rates. The collected tracking data is reflected in a dashboard (e.g., Domo Dashboard) in real time, allowing users to monitor the situation. Based on the data analyzed by the sentiment engine, follow-up actions can be suggested, thereby improving the effectiveness of new customer acquisition.

[0598] Specific example 1:

[0599] The server collects customer data from the database and preprocesses it using the Python pandas library. For example, it scans the "Proposals" folder and filters out unnecessary information. It also uses IBM Watson Tone Analyzer to generate sentiment data from past email feedback.

[0600] Specific example 2:

[0601] The server sends a request to the integrated data platform's API to retrieve company information. Then, clustering analysis is performed using scikit-learn to identify the new company A as belonging to the "IT industry." This data is provided to OpenAI GPT-4, where a generative AI model selects a template related to "IT security" and embeds specific information about company A into the generated proposal.

[0602] Example of a prompt message 1:

[0603] "Customer Information: Detailed Information on Company A"

[0604] "Emotion data: [{\"tone_id\": \"joy\", \"score\": 0.8}]"

[0605] "Please generate a customized proposal email based on this."

[0606] Example of a prompt message 2:

[0607] "Customer Information: Customer X's purchase history and feedback"

[0608] "Emotion data: [{\"tone_id\": \"anger\", \"score\": 0.6}]"

[0609] "Please generate a follow-up suggestion email based on this."

[0610] As a result, this system can efficiently process large amounts of data and provide customized suggestions based on customer sentiment, thereby maximizing the effectiveness of acquiring new customers.

[0611] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0612] Step 1:

[0613] The server collects data such as historical documents and feedback from the database. It uses the Python pandas library for this purpose. The input is customer data stored in the database, which is retrieved, sensitive information is removed, and it is transformed into a clean dataset. The output is pre-processed data, formatted into a format that the AI ​​can learn from.

[0614] Step 2:

[0615] The server uses a sentiment engine (e.g., IBM Watson Tone Analyzer) to analyze the collected feedback data. The input is feedback text, which the sentiment engine analyzes to generate sentiment data (e.g., joy 80%). The output is the analyzed sentiment data, which is used for subsequent customization.

[0616] Step 3:

[0617] The server uses an API to retrieve new customer information from an integrated data platform (e.g., Domo). The input is an API request, and the output is company information in JSON format. The server receives this data and analyzes the characteristics of the company.

[0618] Step 4:

[0619] The server uses the scikit-learn library to perform clustering analysis on company characteristic data. The input is the company information obtained in step 3, which is then analyzed and classified using clustering methods. The output is cluster (segment) information for each company.

[0620] Step 5:

[0621] AI (e.g., OpenAI GPT-4) uses provided company characteristics data to anticipate potential challenges for each company and select the optimal proposal template. The input is company characteristics data and historical templates, and the system generates a proposal. The output is a customized proposal template.

[0622] Step 6:

[0623] The emotion engine adjusts the content of proposals and proposal emails based on past emotion data. The input is a customized proposal template and emotion data, which is used to fine-tune the content. The output is a proposal and proposal email customized according to the emotion.

[0624] Step 7:

[0625] The server sends the generated proposal and proposal email using the SMTP protocol. The input is a customized proposal and proposal email, intended to be sent to the specified email address. The output is the transmission log.

[0626] Step 8:

[0627] The server embeds tracking pixels into emails to track open rates and link clicks. The input is the sent email, to which the tracking pixels are added. The output is recorded as a tracking log and displayed on the dashboard.

[0628] Step 9:

[0629] Users check email open rates and link click rates on a dashboard (e.g., Domo). Input is real-time updated tracking data. Users refer to this data to consider their next follow-up action. Output is the decision on the follow-up action.

[0630] Step 10:

[0631] The user takes follow-up actions based on suggestions from the emotion engine. The input is the suggested content, which the user may use to send additional emails or schedule meetings. The output is the execution of the follow-up action.

[0632] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0633] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0634] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0635] [Second Embodiment]

[0636] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0637] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0638] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0639] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0640] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0641] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0642] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0643] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0644] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0645] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0646] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0647] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0648] Modes for carrying out the invention

[0649] This invention relates to a system that utilizes AI and an integrated data platform to automatically generate proposals for acquiring new customers and to send proposal emails and materials in bulk. The embodiments for carrying out the invention will be described in detail below.

[0650] System Configuration

[0651] This system consists of a database, server, AI, API, and user terminals.

[0652] database

[0653] Accumulate past proposals and related documents.

[0654] server

[0655] Data acquisition and preprocessing

[0656] Customer data acquisition and analysis

[0657] AI-powered proposal generation and email creation.

[0658] Mass email sending

[0659] Situation tracking

[0660] AI

[0661] Learning for Proposal Writing

[0662] Anticipated challenges for each company

[0663] Automated proposal and email generation

[0664] API

[0665] Data integration with integrated data platforms (e.g., Compass, Domo)

[0666] User terminal

[0667] Review of proposal and email content

[0668] Execute follow-up actions

[0669] Operating Procedure

[0670] Step 1: Data Collection and Preprocessing

[0671] server

[0672] Past proposals and related documents are collected from the database, confidential information is removed, and then the data is preprocessed for AI training. Specifically, filtering is performed to reduce unnecessary information.

[0673] Specific example

[0674] The server scans the "Proposals" folder from the database and centrally collects all files. Then, it uses the Python pandas library to identify and filter data containing sensitive information.

[0675] Step 2: Gathering and analyzing information on untapped customers

[0676] server

[0677] Use APIs to retrieve information on untapped companies (industry, number of employees, sales, etc.) from an integrated data platform.

[0678] We analyze the characteristics of companies and provide that data to AI.

[0679] Specific example

[0680] The server sends a request to the integrated data platform's API to retrieve information about company A. Then, using the scikit-learn library, the retrieved information is clustered and identified as belonging to the "IT industry".

[0681] Step 3: Proposal Generation

[0682] server

[0683] The AI ​​anticipates the challenges specific to each company and selects a proposal template based on those challenges.

[0684] By embedding company-specific information into templates, customized proposals are automatically generated.

[0685] Specific example

[0686] The AI ​​selects a template related to "IT security," fills in specific information such as "number of employees at company A" and "sales revenue," and generates a proposal tailored to company A.

[0687] Step 4: Generating the proposal email

[0688] server

[0689] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[0690] Specific example

[0691] The AI ​​generates a proposal email containing the message, "We propose security enhancements for company A." This email includes a proposal document attached in PDF format.

[0692] Step 5: Send out proposals and emails to all recipients.

[0693] server

[0694] The proposal and proposal email will be sent via the mail server based on the mass mailing list.

[0695] Record transmission logs and manage status.

[0696] Specific example

[0697] The server sends the proposal and proposal email to "Company A" using the SMTP protocol. Upon successful transmission, the status is logged.

[0698] Step 6: Follow-up

[0699] User

[0700] You can check email open rates and proposal view rates through the Domo dashboard.

[0701] Send follow-up emails or schedule meetings as needed.

[0702] Specific example

[0703] The user confirms in the Domo dashboard that the email sent to "Company A" has been opened, and the next step is to schedule a follow-up meeting.

[0704] conclusion

[0705] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. In addition, tracking of sending status and follow-up functions improve the overall effectiveness and speed of sales activities.

[0706] The following describes the processing flow.

[0707] Step 1: Data Collection and Preprocessing

[0708] server

[0709] Collect past proposals and related documents from the database. Specifically, use the Python pandas library to execute appropriate queries and retrieve the data.

[0710] Remove confidential information from the acquired data. Filter based on specific keywords or patterns to remove personal information and corporate confidential information.

[0711] The pre-processed data is provided to the AI. The data is converted to a standard format such as CSV and saved as training data for the AI.

[0712] Step 2: Gathering and analyzing information on untapped customers

[0713] server

[0714] Obtain information on untapped companies via the integrated data platform's API. For example, send an HTTP request and receive company information (in JSON format).

[0715] Based on the acquired data, we will analyze the characteristics of the companies. We will use the scikit-learn library to perform clustering analysis and feature extraction.

[0716] The analysis results are provided to the AI. The company characteristics data is converted into a format that the AI ​​can use as reference data for generating proposals (e.g., a data frame format).

[0717] Step 3: Select and generate a proposal template

[0718] server

[0719] AI anticipates potential challenges for each company. Based on company characteristic data, it identifies challenges by comparing them with similar past cases.

[0720] Based on the task, a proposal template is selected. Proposal templates are provided in advance, and the AI ​​selects the appropriate one.

[0721] The template is used to embed company-specific information and generate customized proposals. Specific data (e.g., company revenue or challenges) is embedded in the template's placeholders.

[0722] Step 4: Generating the proposal email

[0723] server

[0724] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[0725] Select the necessary email template and fill in your company's specific content. This will generate customized proposal emails tailored to individual companies.

[0726] Combine the proposal email and proposal document into a single package. Attach the proposal document in PDF format to the email and set the email body accordingly.

[0727] Step 5: Send out proposals and emails to all recipients.

[0728] server

[0729] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[0730] Record transmission logs for each recipient and manage their status. Save success / failure results to a log file for later review.

[0731] Step 6: Tracking email open and view rates

[0732] server

[0733] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[0734] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[0735] Step 7: Follow-up Action

[0736] User

[0737] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[0738] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[0739] (Example 1)

[0740] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0741] The traditional process of creating and sending proposals for acquiring new customers is time-consuming and labor-intensive. Furthermore, manual data collection, analysis, and email sending are inefficient and prone to human error. Additionally, the lack of follow-up and tracking after sending results in low effectiveness of sales activities.

[0742] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0743] This invention includes a server that collects data from a database, removes confidential information, and provides it to the AI; a server that acquires untapped company information via an API, analyzes the characteristics of the companies, and provides it to the AI; a server that anticipates potential challenges for each company and selects and generates proposal templates based on those challenges; a server that generates proposal emails based on the generated proposals and sends them together with the proposals; and a server that tracks the open and viewed status of the sent emails to enable follow-up. This enables the automatic generation and mass sending of proposals and effective follow-up.

[0744] (List of definitions)

[0745] A "database" is a storage device used to accumulate past proposals and related documents.

[0746] "AI" refers to an artificial intelligence system used for automated decision-making and proposal generation.

[0747] An "API" stands for Application Programming Interface, which is used to exchange data between external systems.

[0748] "Untapped company information" refers to data about potential new customer companies, including industry, number of employees, and sales revenue.

[0749] Clustering analysis is a process that uses the scikit-learn library to classify, analyze, and group company data.

[0750] A "template" is a pre-defined form used by AI to generate proposals, and it is a format for embedding company-specific information.

[0751] A "proposal" is a document that outlines a customized proposal tailored to a specific company.

[0752] A "proposal email" is a message used to send the contents of a generated proposal as an email.

[0753] An "SMTP server" is a server used to send emails using the Simple Mail Transfer Protocol.

[0754] The "Domo Dashboard" is a visualization tool for tracking email open rates and proposal view rates.

[0755] "Follow-up" refers to the process of taking additional actions or making contact regarding submitted proposals or emails.

[0756] "Filtering" is the process of removing information from data that should not be released externally.

[0757] Modes for carrying out the invention

[0758] This invention relates to a system for automatically generating and efficiently sending proposals and emails for acquiring new customers. In particular, it aims to automate proposal creation, analyze company characteristics, and improve the efficiency of email sending by utilizing AI and an integrated data platform. The specific methods for realizing this system are described below.

[0759] This system consists of a database, servers, AI, APIs, and user terminals.

[0760] database

[0761] A storage device is used to accumulate past proposals and related documents. This allows the AI ​​to learn from past proposal data and generate more accurate proposals.

[0762] server

[0763] The server is responsible for several processes, including the following:

[0764] 1. Data acquisition and preprocessing:

[0765] The server collects proposals and related documents from the database, filters the data using the Python pandas library, and removes confidential information.

[0766] Specific example: The server executes an SQL query against the database to retrieve proposal data. This data is then read using the pandas library, and confidential information is filtered out.

[0767] 2. Collection and analysis of untapped company information:

[0768] The server uses APIs to retrieve information on untapped companies from integrated data platforms (e.g., Compass, Domo). This information includes industry, number of employees, and revenue.

[0769] The server uses the scikit-learn library to perform clustering analysis on company characteristics and provides the results to the AI.

[0770] Specific example: A server sends an API request to retrieve company characteristic data. Clustering analysis is performed using the scikit-learn library to identify company characteristics.

[0771] 3. Proposal generation:

[0772] The AI ​​anticipates the challenges specific to each company and selects a proposal template based on those challenges. It then generates a customized proposal by embedding company-specific information into the template.

[0773] Specific example: The AI ​​selects an appropriate template from a template database and uses the Jinja2 template engine to embed company information.

[0774] 4. Generating and sending proposal emails:

[0775] Based on the proposal document, the AI ​​generates a proposal email. The generated email includes a summary of the proposal and the reasons for the proposal.

[0776] The server uses the SMTP protocol to send proposal emails and proposal documents in bulk. It records transmission logs and manages the status.

[0777] Specific example: AI uses an email template to generate an email with a proposal PDF attached. The server sends the email via SMTP server and records the sending log.

[0778] User terminal

[0779] The user terminal is a device used to review proposals and email content, and to perform follow-up actions. The Domo dashboard allows users to check email open rates and proposal viewing rates, and to send follow-up emails or schedule meetings as needed.

[0780] Example prompt statements

[0781] Data collection:

[0782] "Write Python code to scan the 'Proposals' folder in the database and collect all files. Then, use pandas to identify and remove columns that contain sensitive information."

[0783] Collection and analysis of corporate information:

[0784] "Write Python code that retrieves information from company A's API and performs clustering analysis using scikit-learn."

[0785] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. In addition, tracking of sending status and follow-up functions improve the overall effectiveness and speed of sales activities.

[0786] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0787] Program processing flow

[0788] Step 1: Data Collection and Preprocessing

[0789] Input: Database containing past proposals and related documents.

[0790] Output: Pre-processed data with sensitive information removed, ready to be provided to the AI.

[0791] Specific actions:

[0792] 1. The server connects to the database and uses SQL queries to collect proposal data.

[0793] 2. After obtaining the proposal data, the data is filtered using the Python pandas library to remove confidential information.

[0794] 3. Save the filtered data to provide it to the AI ​​model.

[0795] Specific data processing:

[0796] The server extracts proposal data from the database using SQL queries. This data is then read into pandas, and specific columns or values ​​are filtered to remove sensitive information. Finally, the pre-processed data is saved as a file.

[0797] Specific example:

[0798] The server downloads all files in the "Proposal" folder at once. The pandas drop function is used to remove columns containing confidential information.

[0799] Step 2: Gathering and analyzing information on untapped companies

[0800] Input: Company information obtained via API

[0801] Output: Analyzed company characteristics data

[0802] Specific actions:

[0803] 1. The server sends a request to the integrated data platform's API to retrieve data on untapped companies.

[0804] 2. Perform clustering analysis on the acquired data using the scikit-learn library.

[0805] 3. Save the analysis results to the database and prepare them for provision to the AI.

[0806] Specific data processing:

[0807] The server uses an API to retrieve company data (e.g., industry, number of employees, sales). This data is then analyzed using scikit-learn's clustering algorithm to group company characteristics. The analysis results are saved for use with the AI.

[0808] Specific example:

[0809] The server retrieves data from company A via an API request, performs clustering analysis using the scikit-learn KMeans algorithm, and identifies company A's industry.

[0810] Step 3: Proposal Generation

[0811] Input: Company characteristics data and past proposal templates

[0812] Output: Customized proposal

[0813] Specific actions:

[0814] 1. The server uses AI to anticipate the challenges specific to each company and selects a proposal template based on those challenges.

[0815] 2. Create customized proposals by embedding company-specific information into proposal templates.

[0816] Specific data processing:

[0817] The AI ​​selects an appropriate proposal template based on company characteristics data and uses the Jinja2 template engine to embed company information.

[0818] Specific example:

[0819] The AI ​​selects a template related to "IT security" from a template database and fills in specific information about company A (number of employees, sales revenue).

[0820] Step 4: Generating the proposal email

[0821] Input: Generated proposal

[0822] Output: Proposal email

[0823] Specific actions:

[0824] 1. The AI ​​analyzes the generated proposal and creates a proposal email.

[0825] 2. Attach the proposal PDF file to the email you created.

[0826] Specific data processing:

[0827] The AI ​​extracts necessary information from the generated proposal and inserts it into an email template. The proposal PDF is then attached to the email to complete the process.

[0828] Specific example:

[0829] The AI ​​generates an email stating, "We propose security enhancements for company A," and attaches a PDF of the proposal.

[0830] Step 5: Send out proposals and emails to all recipients.

[0831] Input: Proposal email and proposal PDF

[0832] Output: Recording of transmission logs and status information

[0833] Specific actions:

[0834] 1. The server uses the SMTP protocol to send proposal emails and proposal PDFs simultaneously.

[0835] 2. Log the status and manage it when the transmission is complete.

[0836] Specific data processing:

[0837] Send emails via SMTP server. The sending status and details of the email are recorded in the sending log.

[0838] Specific example:

[0839] The server uses SMTP to send the proposal and proposal email to company A and records it in the transmission log.

[0840] Step 6: Follow-up

[0841] Input: Email open status and proposal viewing status

[0842] Output: Follow-up actions (e.g., sending an email, scheduling a meeting)

[0843] Specific actions:

[0844] 1. Users can check email open rates and proposal view rates through the Domo dashboard.

[0845] 2. Send follow-up emails or schedule meetings as needed.

[0846] Specific data processing:

[0847] Check the open and view rates on the Domo dashboard and take the next action (send a follow-up email, schedule a meeting).

[0848] Specific example:

[0849] The user confirms on the Domo dashboard that "the email sent to company A has been opened" and sets up a follow-up meeting.

[0850] (Application Example 1)

[0851] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0852] Traditional customer proposal creation systems required manual proposal creation and customer information analysis, which was time-consuming and labor-intensive. Furthermore, the process of generating personalized proposals was cumbersome, hindering the efficiency of new customer acquisition. Additionally, there was a lack of management functions to track the effectiveness of generated proposals and provide appropriate follow-up.

[0853] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0854] This invention includes a server that provides data to an AI after collecting information from a database and removing confidential information; a server that provides data to an AI after obtaining untapped company information via an API, analyzing the characteristics of the companies, and providing this information to the AI; a server that allows the AI ​​to anticipate potential challenges for each company and select and generate proposal templates based on those challenges; a server that allows the AI ​​to generate proposal emails based on the generated proposals and sends them together with the proposals; a server that tracks the open and viewed status of sent emails to enable follow-up; a server that collects customer information from a sales site, analyzes it based on past purchase history and visit data, and generates customized product proposals as part of the proposals; and a server that generates personalized proposal emails based on the customized product proposals, sends them all at once via a mail server, and provides in-app notifications. This makes it possible to automatically and efficiently generate personalized proposals and proposal emails for new customers and send them all at once. In addition, it is possible to track the open and viewed status of emails and perform effective follow-up.

[0855] A "database" is a data storage system that centrally stores information such as past documents and proposals, and allows data to be retrieved as needed.

[0856] "Confidential information" refers to non-public information about a company or its customers that should not be leaked to the outside.

[0857] "AI" is an abbreviation for artificial intelligence, which refers to technologies that automate or optimize specific tasks through data analysis and machine learning.

[0858] "API" stands for Application Programming Interface, which is an interface for exchanging data and functions between different software programs.

[0859] "Untapped company information" refers to information about companies with which no business transactions have yet been concluded, including industry, number of employees, and sales figures.

[0860] A "template" is a basic model used when creating documents such as proposals or emails, and it includes a specific format and content.

[0861] A "proposal" is a document that formalizes a proposal made to a company or customer, and it includes details of the proposal, the reasons for it, and specific benefits.

[0862] A "proposal email" is an email sent to a customer to which a proposal and related information are presented. It is an email that outlines the proposal and its benefits.

[0863] A "sales site" refers to a website or online platform used for selling products.

[0864] "Customer information" refers to all information about a customer, such as purchase history and website visit history.

[0865] "Customized product recommendations" refer to suggestions tailored to the customer's needs, based on their purchase history and visit data, to propose the most suitable products for that customer.

[0866] A "personalized suggestion email" is a suggestion email that contains content that is individually customized for a specific customer.

[0867] "Follow-up" refers to ongoing support and tracking activities conducted after the initial proposal, primarily to check email open rates and the effectiveness of the proposed content.

[0868] This invention relates to a system for automatically generating personalized product proposals and sending them out via email and accompanying materials to customers on a sales website in order to acquire new customers. The embodiments for carrying out the invention will be described in detail below.

[0869] System Configuration

[0870] This system consists of a database, server, AI, API, and user terminals.

[0871] database

[0872] The system stores information such as past proposals, purchase history, and visit data.

[0873] server

[0874] Data acquisition and preprocessing

[0875] Customer data acquisition and analysis

[0876] AI-powered proposal generation and email creation.

[0877] Mass email sending and in-app notifications

[0878] Situation tracking

[0879] AI

[0880] Learning for Proposal Writing

[0881] Anticipating challenges for each customer

[0882] Automated proposal and email generation

[0883] API

[0884] Data integration with the integrated data platform

[0885] User terminal

[0886] Review of proposal and email content

[0887] Execute follow-up actions

[0888] Program processing

[0889] The server first collects past purchase history and visit data from the database and performs preprocessing. Specifically, it uses the Python pandas library to format the data and filters out sensitive information. Next, it retrieves untapped customer information via the integrated data platform's API. The collected customer information is classified using clustering analysis with the scikit-learn library.

[0890] The AI ​​selects a suitable product suggestion template for each categorized customer and generates customized product suggestions based on their purchase history and interests. During this process, a generation AI model is used to generate prompts and create specific suggestion content. The generated suggestion emails are sent to customers simultaneously via a mail server, along with in-app notifications.

[0891] Hardware and software

[0892] Hardware:

[0893] server

[0894] User terminal

[0895] software:

[0896] Python

[0897] pandas

[0898] scikit-learn

[0899] smtplib

[0900] request

[0901] Specific example

[0902] For example, when using this system to provide personalized product recommendations to a new customer G1, the server retrieves information on the untapped customer G1 via the integrated data platform's API and classifies the customer's characteristics using clustering analysis. The AI ​​then generates a "health and wellness-related" product recommendation document based on G1's past purchase history and visit data, and creates a personalized recommendation email for G1. The generated email is sent via a mail server, and in-app notifications are also sent.

[0903] Example of a prompt:

[0904] "Please generate an email suggesting health and wellness products based on the products the customer has purchased and the pages they have viewed. The email should include explanations of why the suggested products are suitable for the customer's needs."

[0905] This makes it possible to automatically and efficiently generate and send personalized product proposals and promotional emails to new customers on the sales site. Furthermore, it allows for tracking email open rates and viewing activity, enabling effective follow-up.

[0906] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0907] Step 1:

[0908] The server collects past purchase history and visit data from the database. Specifically, it reads and formats this data using the Python pandas library. Since the collected data may contain sensitive information, it is filtered to remove any confidential information.

[0909] Input: Purchase history and visit data from the database

[0910] Output: Formatted and sensitive data removed dataset

[0911] Step 2:

[0912] The server retrieves untapped customer information via the integrated data platform's API. Specifically, it sends API requests to obtain information such as the customer's industry, number of employees, and sales revenue. The retrieved data is then read using the pandas library and converted into the appropriate format.

[0913] Input: API request for the integrated data platform

[0914] Output: Dataset of characteristics information for untapped customers

[0915] Step 3:

[0916] The server performs clustering analysis on the acquired customer information using the scikit-learn library. This clustering analysis classifies the customers into clusters based on their characteristics.

[0917] Input: Dataset of characteristics information for untapped customers

[0918] Output: Customer data with cluster labels assigned.

[0919] Step 4:

[0920] The AI ​​selects the optimal proposal template for each customer based on cluster labels. After selecting a template, it automatically generates a customized proposal by embedding customer characteristic information.

[0921] Input: Customer data with cluster labels assigned.

[0922] Output: Customized proposal

[0923] Step 5:

[0924] The AI ​​creates a personalized proposal email based on the generated proposal. The proposal email includes a summary of the proposal and the reasons for the proposal.

[0925] Input: Customized proposal

[0926] Output: Personalized suggestion email

[0927] Step 6:

[0928] The server sends proposals and proposal emails via the mail server based on a mass mailing list. It also simultaneously sends in-app notifications. It records sending logs and manages the status.

[0929] Input: Personalized suggestion email

[0930] Output: Sending log, in-app notification status

[0931] Step 7:

[0932] Users can check email open rates and proposal viewing rates through the dashboard. They can also send follow-up emails or schedule meetings as needed.

[0933] Input: Sending log, in-app notification status

[0934] Output: Follow-up actions, such as follow-up emails or meeting scheduling.

[0935] Through the above processing steps, this system can automatically and efficiently generate personalized proposals and emails for new customers and send them in bulk. Furthermore, it can track email open and view rates, enabling effective follow-up.

[0936] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0937] Modes for carrying out the invention

[0938] This invention relates to a system that utilizes AI, an integrated data platform, and an emotion engine to automatically generate proposals and emails for acquiring new customers, and further customizes them based on the user's emotions. The embodiments for carrying out the invention will be described in detail below.

[0939] System Configuration

[0940] This system consists of a database, server, AI, API, emotion engine, and user terminal.

[0941] database

[0942] It stores past proposals, related documents, and user sentiment data.

[0943] server

[0944] Data acquisition and preprocessing

[0945] Customer data acquisition and analysis

[0946] AI-powered proposal generation and email creation.

[0947] Emotional engine integration

[0948] Mass email sending

[0949] Situation tracking

[0950] AI

[0951] Learning for Proposal Writing

[0952] Anticipated challenges for each company

[0953] Automated proposal and email generation

[0954] API

[0955] Data integration with integrated data platforms (e.g., Compass, Domo)

[0956] Emotional Engine

[0957] We analyze user emotions and provide emotion data in real time. Based on this emotion data, we customize proposals and email content.

[0958] User terminal

[0959] Review of proposal and email content

[0960] Execute follow-up actions

[0961] Operating Procedure

[0962] Step 1: Data Collection and Preprocessing

[0963] server

[0964] Past proposals and related documents are collected from the database, confidential information is removed, and then the data is preprocessed for AI training. For example, filtering is performed to reduce unnecessary information.

[0965] The emotion engine is used to analyze user sentiment data regarding past suggestion emails and store it in a database.

[0966] Specific example

[0967] The server scans the "Proposals" folder from the database and retrieves the data using the Python pandas library. It filters out sensitive information and provides the preprocessed data to the AI. Additionally, an emotion engine analyzes user responses in past emails and stores the emotion data in the database.

[0968] Step 2: Gathering and analyzing information on untapped customers

[0969] server

[0970] We will use an API to retrieve information on untapped companies from an integrated data platform. Specifically, we will send an HTTP request and receive company information (in JSON format).

[0971] Based on the acquired data, we analyze the characteristics of the company. For example, we perform clustering analysis using the scikit-learn library.

[0972] The analysis results are provided to the AI. The company characteristics data is used by the AI ​​as reference data for generating proposals.

[0973] Specific example

[0974] The server sends a request to the integrated data platform's API to retrieve information about company A. Then, using scikit-learn, it identifies that company A belongs to the "IT industry".

[0975] Step 3: Select and generate a proposal template

[0976] server

[0977] The AI ​​anticipates potential challenges for each company and selects a proposal template based on those challenges. The optimal template is chosen based on company characteristic data.

[0978] The system automatically generates customized proposals by embedding company-specific information into proposal templates.

[0979] Using an emotion engine, template selection and proposal content are adjusted based on the user's past emotional data.

[0980] Specific example

[0981] The AI ​​selects a template related to "IT security" and fills in specific information about company A, such as "number of employees" and "sales revenue." At the same time, the emotion engine refers to past emotion data and adjusts the content to elicit a more positive response.

[0982] Step 4: Generating the proposal email

[0983] server

[0984] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[0985] Select the necessary email template, embed company-specific content, and customize the email content by reflecting sentiment data analyzed by the sentiment engine.

[0986] Combine the proposal email and proposal document into a single package.

[0987] Specific example

[0988] The AI ​​generates a proposal email containing the message, "We propose strengthening security for company A," and attaches the proposal as a PDF document. The emotion engine adjusts the wording and tone of the email to suit the user's preferences.

[0989] Step 5: Send out proposals and emails to all recipients.

[0990] server

[0991] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[0992] Record transmission logs for each recipient and manage their status.

[0993] Specific example

[0994] The server uses the SMTP protocol to send the proposal and proposal email to "Company A," and logs the status upon successful transmission.

[0995] Step 6: Tracking email open and view rates

[0996] server

[0997] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[0998] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[0999] Specific example

[1000] The server embeds tracking pixels in emails to detect open rates, which are then displayed in real time on the Domo dashboard.

[1001] Step 7: Follow-up Action

[1002] User

[1003] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[1004] Based on the data analyzed by the emotion engine, the next follow-up action will be suggested.

[1005] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[1006] Specific example

[1007] The user confirms in the Domo dashboard that the email sent to "Company A" was opened, and based on the sentiment engine's suggestions, sets up a follow-up meeting as the next step.

[1008] conclusion

[1009] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. Tracking and follow-up features improve the overall effectiveness and speed of sales activities. In addition, the integration of an emotion engine enables customization based on user emotions, further enhancing the accuracy and effectiveness of proposals.

[1010] The following describes the processing flow.

[1011] Step 1: Data Collection and Preprocessing

[1012] server

[1013] Collect past proposals and related documents from the database. Specifically, use the Python pandas library to execute appropriate queries and retrieve the data.

[1014] Remove confidential information from the acquired data. For example, use regular expressions based on specific keywords or patterns to filter out confidential information and remove personal and corporate confidential information.

[1015] The pre-processed data is provided to the AI. The data is converted to a standard format such as CSV and saved as a training dataset for the AI.

[1016] In addition, an emotion engine is used to analyze users' emotional data regarding past suggestion emails and store it in a database. The emotion engine uses natural language processing technology to analyze the content of emails and assigns emotional labels such as positive, negative, and neutral.

[1017] Step 2: Gathering and analyzing information on untapped customers

[1018] server

[1019] We will use an API to retrieve information on untapped companies from an integrated data platform. Specifically, we will send an HTTP request and receive company information (in JSON format).

[1020] Based on the acquired data, we analyze the characteristics of companies. For example, we use the scikit-learn library to perform clustering analysis and feature extraction to reveal characteristics such as the company's industry, size, and sales revenue.

[1021] The analysis results are provided to the AI. The company characteristics data is converted into a format that the AI ​​can use as reference data for generating proposals (e.g., a data frame format).

[1022] Step 3: Select and generate a proposal template

[1023] server

[1024] AI anticipates potential challenges for each company. Based on company characteristic data, it compares it with similar past cases and industry trends to identify the most relevant issues.

[1025] Based on the selected issue, a proposal template is chosen. Proposal templates are prepared in advance, and the AI ​​selects the appropriate one.

[1026] The template is used to embed company-specific information and generate customized proposals. Specific data (e.g., company revenue or challenges) is embedded in the template's placeholders.

[1027] Using an emotion engine, template selection and proposal content are adjusted based on the user's past emotional data. For example, expressions and designs that received many positive responses are prioritized.

[1028] Step 4: Generating the proposal email

[1029] server

[1030] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[1031] Select the necessary email template and fill in your company's specific content. The sentiment engine analyzes emotional data to customize the email content for greater effectiveness. For example, it adopts phrasing and tones that elicit positive responses from users.

[1032] The proposal email and proposal document will be packaged together as a single package, with the proposal document attached to the email in PDF format.

[1033] Step 5: Send out proposals and emails to all recipients.

[1034] server

[1035] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[1036] Record transmission logs for each recipient and manage their status. Save success / failure results to a log file for later review.

[1037] Step 6: Tracking email open and view rates

[1038] server

[1039] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[1040] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[1041] Step 7: Follow-up Action

[1042] User

[1043] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[1044] Based on the data analyzed by the emotion engine, the next follow-up action is suggested. For example, if the user showed a positive response to the previous email, a more assertive approach is suggested.

[1045] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[1046] (Example 2)

[1047] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[1048] This invention relates to a system that significantly reduces the time and effort required for acquiring new customers in a company's sales activities and makes proposals to customers more effective. Conventional systems require a great deal of time and effort to create proposals and generate proposal emails to customers, and they are not sufficiently customized based on customer sentiment. Therefore, there has been a need for a system that supports new customer acquisition in an efficient and effective way.

[1049] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1050] This invention includes a server that provides data to an AI after collecting information from a database and removing confidential information; a server that provides data to an AI after obtaining untapped company information via an API and analyzing the characteristics of the companies and providing this information to the AI; a server that allows the AI ​​to anticipate potential challenges for each company and select and generate proposal templates based on those challenges; a server that allows the AI ​​to generate proposal emails based on the generated proposals and sends them together with the proposals; a server that tracks the open and viewed status of sent emails and enables follow-up; a server that uses an emotion analysis engine to accumulate past emotional data of users and customize the content of proposals and proposal emails; a server that sends proposal emails and proposals all at once; a server that uses a Domo dashboard to display the sending status in real time; and a server that suggests follow-up actions. This enables the efficient and effective generation and sending of proposals and proposal emails, and makes it easier to realize optimal proposals based on the user's emotions.

[1051] "Document collection" refers to the act of obtaining past proposals and related documents from a database.

[1052] "Removal of confidential information" is the process of filtering and removing information from collected materials that should not be released to the public.

[1053] "Means of providing data to AI" refers to the process of inputting pre-processed data into a machine learning system.

[1054] "Acquiring company information" means collecting data on untapped companies via APIs.

[1055] "Company characteristic analysis" refers to evaluating the characteristics of each company based on acquired company information using methods such as clustering analysis.

[1056] "Identifying challenges" refers to AI predicting potential problems and areas for improvement specific to each company.

[1057] "Selecting a proposal template" is a method of choosing the most suitable proposal format based on the anticipated challenges.

[1058] "Proposal generation" refers to the process of creating a customized proposal by embedding company-specific information into a template.

[1059] "Proposal email generation" refers to the process where AI creates an email based on the generated proposal document, outlining the proposal's summary and reasons.

[1060] "Mass email sending" refers to the process of simultaneously sending a generated proposal document and proposal email to multiple recipients based on a list.

[1061] "Open tracking" refers to tracking and recording whether or not an email that has been sent has been opened.

[1062] "Follow-up action" refers to the act of conducting follow-up surveys or making additional contact with users after sending a proposal email, depending on their situation.

[1063] A "sentiment analysis engine" is a system that analyzes a user's past emotional data and optimizes the suggested content based on that data.

[1064] The "Domo Dashboard" is a tool for visualizing data and displaying real-time tracking information.

[1065] This invention relates to a system that utilizes AI, an integrated data platform, and an emotion analysis engine to automatically generate proposals and emails for acquiring new customers, and further customizes them based on the user's emotions. The detailed configuration and operation of the system are described below.

[1066] System Configuration

[1067] This system consists of a database, server, AI, API, sentiment analysis engine, and user terminal.

[1068] database

[1069] It stores past proposals, related documents, and user sentiment data.

[1070] server

[1071] It performs data collection and preprocessing, customer data acquisition and analysis, AI-powered proposal generation and email creation, sentiment analysis engine integration, mass email sending, and status tracking.

[1072] AI

[1073] This system provides learning opportunities for proposal creation, anticipates challenges specific to each company, and automatically generates proposals and emails.

[1074] API

[1075] Establish data integration with integrated data platforms (e.g., Compass, Domo).

[1076] Emotion analysis engine

[1077] We analyze user emotions and provide emotion data in real time. Based on this emotion data, we customize proposals and email content.

[1078] User terminal

[1079] Review the proposal and email content, and take follow-up actions.

[1080] Operating Procedure

[1081] The following explains how the system works using specific examples for each step.

[1082] Step 1: Data Acquisition and Preprocessing

[1083] server

[1084] The server collects past proposals and related documents from the database and removes confidential information. This process uses the Python pandas library to load, filter, and clean the data. It also uses a sentiment analysis engine to analyze user sentiment data regarding past proposal emails and stores it in the database.

[1085] Specific example

[1086] The server scans the "Proposal" folder from the database, retrieves the data using the pandas library, filters it, and provides the preprocessed data to the AI. The sentiment analysis engine analyzes the user's reactions contained in past emails and stores the sentiment data in the database.

[1087] Step 2: Gathering and analyzing information on untapped customers

[1088] server

[1089] The server uses an API to retrieve information on untapped companies from the integrated data platform. Specifically, it sends an HTTP request and receives company information in JSON format. Then, it uses the scikit-learn library to perform clustering analysis, extract company characteristics, and provide them to the AI.

[1090] Specific example

[1091] The server sends an HTTP request to the integrated data platform's API to retrieve information about company A, and then uses scikit-learn to identify that company A belongs to the "IT industry".

[1092] Step 3: Select and generate a proposal template.

[1093] server

[1094] The AI ​​anticipates potential challenges for each company and selects proposal templates based on those challenges. Based on company characteristic data, it selects the optimal template, embeds company-specific information into the template, and automatically generates a customized proposal. In addition, it uses an emotion analysis engine to adjust template selection and proposal content based on the user's past emotion data.

[1095] Specific example

[1096] The AI ​​selects a template related to "IT security" and fills in information such as company A's "number of employees" and "sales revenue." Simultaneously, an emotion analysis engine refers to past emotion data and adjusts the content to elicit a positive response from the user.

[1097] Step 4: Generate the proposal email

[1098] server

[1099] Based on the generated proposal document, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its proposal, and its content is customized using a sentiment analysis engine. The proposal email and proposal document are then combined into a single package.

[1100] Specific example

[1101] The AI ​​generates a proposal email containing the message, "We propose strengthening security for company A," and attaches the proposal as a PDF document. An emotion analysis engine adjusts the wording and tone of the email to suit the user's preferences.

[1102] Step 5: Send proposals and emails to all recipients.

[1103] server

[1104] The server uses the SMTP protocol to send proposals and proposal emails in bulk. Specifically, it uses the smtplib library to send emails based on a mailing list. It also records sending logs and manages the sending status.

[1105] Specific example

[1106] The server uses the SMTP protocol to send the proposal and proposal email to "Company A," and logs the status upon successful transmission.

[1107] Step 6: Track email open and view rates

[1108] server

[1109] The server uses embedded tracking pixels to track email open rates and proposal viewing rates. The collected tracking data is reflected in the Domo dashboard in real time.

[1110] Specific example

[1111] The server embeds tracking pixels in emails to detect open rates, which are then displayed in real time on the Domo dashboard.

[1112] Step 7: Follow-up Action

[1113] User

[1114] Users can check email open rates and proposal view rates through the Domo dashboard. Based on the data analyzed by the sentiment analysis engine, the system suggests the next follow-up actions. Users can then take additional follow-up actions as needed, such as sending follow-up emails or scheduling meetings.

[1115] Specific example

[1116] The user confirms on the Domo dashboard that "Company A" has opened the proposal and sets up a follow-up meeting based on the sentiment analysis engine's recommendations.

[1117] Examples of prompt statements

[1118] Select a proposal template and generate a proposal tailored to Company A. Adjust the content based on past sentiment data and output the optimal proposal in PDF format. Also, send the proposal and proposal email based on the mass email mailing list and track the open rates.

[1119] As described above, this system utilizes AI and emotion analysis engines to enable the automated generation and customization of efficient and effective proposals and emails for new customers, and further optimizes sales activities through tracking of sending status and suggesting follow-up actions.

[1120] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1121] Step 1: Data Collection and Preprocessing

[1122] server

[1123] The server collects past proposals and related documents from the database. It executes SQL queries against the database to read the relevant data. Then, it removes confidential information from the collected data. Specifically, it uses the Python pandas library to filter and clean the data. It also uses a sentiment analysis engine to analyze past proposal email data, extract user sentiment data, and store it in the database.

[1124] input

[1125] Proposals, related documents, and past proposal email data collected from the database.

[1126] output

[1127] Confidential information has been removed, and the proposal data has been pre-processed and analyzed for sentiment.

[1128] Specific actions

[1129] The server retrieves data from the database, filters the necessary data using the pandas library, removes sensitive information, and performs preprocessing.

[1130] Step 2: Gathering and analyzing information on untapped customers

[1131] server

[1132] The server uses an API to retrieve information on untapped companies from an integrated data platform. It sends an HTTP request and receives company information in JSON format. Then, it performs clustering analysis using the scikit-learn library to extract company characteristics. The characteristic data is analyzed and provided to the AI.

[1133] input

[1134] Information on untapped companies obtained via API (in JSON format).

[1135] output

[1136] Company characteristic data extracted through clustering analysis.

[1137] Specific actions

[1138] The server sends an API request to retrieve information on untapped companies. This data is then clustered and analyzed using scikit-learn to extract company characteristics.

[1139] Step 3: Select and generate a proposal template

[1140] server

[1141] The AI ​​anticipates potential challenges for each company and selects a proposal template based on those challenges. It uses company characteristic data to choose the optimal proposal template and embeds company-specific information into it. Furthermore, it uses an emotion analysis engine to adjust the proposal content based on the user's past emotional data.

[1142] input

[1143] Company characteristic data and historical sentiment data.

[1144] output

[1145] A customized proposal tailored to each company.

[1146] Specific actions

[1147] AI analyzes the data and selects the optimal proposal template. An emotion analysis engine then adjusts the content.

[1148] Step 4: Generating the proposal email

[1149] server

[1150] Based on the generated proposal, the AI ​​creates a proposal email. Using the Jinja2 template engine, it generates an email containing an overview of the proposal and the reasons for the proposal. Using an emotion analysis engine, the email content is customized based on the user's emotion data.

[1151] input

[1152] Generated proposals, past sentiment data.

[1153] output

[1154] Customized proposal email.

[1155] Specific actions

[1156] The AI ​​generates emails based on the proposal, and the sentiment analysis engine adjusts the content.

[1157] Step 5: Send out proposals and emails to all recipients.

[1158] server

[1159] The server uses the SMTP protocol to send proposals and proposal emails in bulk. It uses the smtplib library to send emails based on a designated mailing list. It records sending logs and manages the sending status.

[1160] input

[1161] Customized proposal emails, proposal documents, and mass mailing lists.

[1162] output

[1163] Log and transmission status after transmission is complete.

[1164] Specific actions

[1165] The server sends an email using the SMTP protocol and logs the status.

[1166] Step 6: Tracking email open and view rates

[1167] server

[1168] The server uses embedded tracking pixels to track email open rates and proposal view rates. The collected data is reflected in the Domo dashboard in real time.

[1169] input

[1170] A proposal email with embedded tracking pixels.

[1171] output

[1172] Tracking data on opening and browsing activity.

[1173] Specific actions

[1174] The server uses tracking pixels to detect open and viewed status and displays it on the Domo dashboard.

[1175] Step 7: Follow-up Action

[1176] User

[1177] Through the Domo dashboard, users can check email open rates and proposal view rates. Based on the data analyzed by the sentiment analysis engine, the system suggests the next follow-up actions. Users can then send follow-up emails or schedule meetings as needed.

[1178] input

[1179] Data on opening and viewing status, and analysis data from the sentiment analysis engine.

[1180] output

[1181] Follow-up actions (e.g., sending a follow-up email or scheduling a meeting).

[1182] Specific actions

[1183] Users check the situation on the Domo dashboard and take follow-up actions based on suggestions from the sentiment analysis engine.

[1184] (Application Example 2)

[1185] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[1186] Traditional methods of acquiring new customers made it difficult to process large amounts of customer data quickly and efficiently, and to make appropriate proposals to each customer. Furthermore, customizing proposals based on customer emotions and feedback was challenging, resulting in insufficient engagement with customers. This led to decreased customer engagement and difficulties in acquiring new customers.

[1187] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting data from a database, removing confidential information and providing it to the AI; means for acquiring untapped company information via an API, analyzing the characteristics of the companies and providing it to the AI; means for the AI ​​to anticipate potential challenges for each company and select and generate a proposal template based on those challenges; means for having the AI ​​generate a proposal email based on the generated proposal and sending it together with the proposal; means for tracking the open and viewing status of sent emails to enable follow-up; means for collecting data such as past purchase history and feedback and analyzing sentiment data; means for customizing the content of the proposal and proposal email based on the sentiment data; and means for generating proposal emails and coupons for individual customers using a generation AI model. This makes it possible to efficiently process large amounts of data and make customized proposals based on customer sentiment.

[1188] A "database" is an information system that stores and manages data such as past documents and feedback.

[1189] "Confidential information" refers to important information that must not be leaked to outsiders.

[1190] "AI" refers to artificial intelligence, a technology that processes and analyzes large amounts of data to automatically generate suggestions.

[1191] "API" stands for Application Programming Interface, which is a gateway for different software systems to exchange data and functions.

[1192] "Company characteristics" refer to the characteristics of a company, such as its industry, size, and profitability.

[1193] A "proposal" is a document that outlines the proposed content for a specific client.

[1194] A "template" is a format that serves as a guide for efficiently creating proposals, emails, and other documents.

[1195] A "proposal email" is an email sent to a client that includes an overview of the proposal and the reasons for the proposal.

[1196] "Tracking" is the process of tracking and recording things like whether an email was opened or whether links were clicked after it was sent.

[1197] "Follow-up" refers to additional actions taken after sending a proposal email in order to maintain and deepen the relationship with the customer.

[1198] "Purchase history" refers to a record of products that a customer has purchased in the past.

[1199] "Feedback" refers to the opinions and evaluations received from customers.

[1200] "Emotional data" refers to data that analyzes customer emotions and reactions and expresses them as numerical or textual data.

[1201] A "generative AI model" is a model trained to automatically generate documents and suggestions using artificial intelligence.

[1202] A "coupon" is an electronic or paper voucher used to offer discounts or benefits to customers.

[1203] This invention relates to a new customer acquisition system comprising a database, server, AI, API, emotion engine, and user terminal. This system automatically generates proposals and proposal emails and customizes them based on user emotion data through the following steps.

[1204] The server collects data such as historical documents and feedback from the database, and preprocesses it by removing sensitive information. It uses the Python pandas library to retrieve data and filter out sensitive information. It also uses a sentiment engine to analyze user responses in past emails and generate sentiment data. IBM Watson Tone Analyzer is used as the sentiment engine.

[1205] Using an API, the server retrieves new customer information from an integrated data platform (e.g., Domo). It sends HTTP requests to obtain company information (in JSON format) and analyzes the company's characteristics. Clustering analysis is performed using the scikit-learn library. This characteristic data is used as reference when an AI (e.g., OpenAI GPT-4) generates proposals.

[1206] Using the GPT-4 generative AI model, the AI ​​anticipates potential challenges specific to each company and selects and generates the optimal proposal template based on those challenges. It then customizes the template based on sentiment data while incorporating company-specific information (such as the number of employees and sales). This process generates more effective proposals that reflect the user's emotions.

[1207] Next, the AI ​​creates a proposal email based on the generated proposal. Using sentiment data provided by the sentiment engine, it adjusts the email content to maximize customer response. This email includes a summary of the proposal and the reasons for the proposal, and is sent all at once. The server uses the SMTP protocol to send the email to all recipients simultaneously.

[1208] Tracking pixels are embedded in emails to track email open rates and link click rates. The collected tracking data is reflected in a dashboard (e.g., Domo Dashboard) in real time, allowing users to monitor the situation. Based on the data analyzed by the sentiment engine, follow-up actions can be suggested, thereby improving the effectiveness of new customer acquisition.

[1209] Specific example 1:

[1210] The server collects customer data from the database and preprocesses it using the Python pandas library. For example, it scans the "Proposals" folder and filters out unnecessary information. It also uses IBM Watson Tone Analyzer to generate sentiment data from past email feedback.

[1211] Specific example 2:

[1212] The server sends a request to the integrated data platform's API to retrieve company information. Then, clustering analysis is performed using scikit-learn to identify the new company A as belonging to the "IT industry." This data is provided to OpenAI GPT-4, where a generative AI model selects a template related to "IT security" and embeds specific information about company A into the generated proposal.

[1213] Example of a prompt message 1:

[1214] "Customer Information: Detailed Information on Company A"

[1215] "Emotion data: [{\"tone_id\": \"joy\", \"score\": 0.8}]"

[1216] "Please generate a customized proposal email based on this."

[1217] Example of a prompt message 2:

[1218] "Customer Information: Customer X's purchase history and feedback"

[1219] "Emotion data: [{\"tone_id\": \"anger\", \"score\": 0.6}]"

[1220] "Please generate a follow-up suggestion email based on this."

[1221] As a result, this system can efficiently process large amounts of data and provide customized suggestions based on customer sentiment, thereby maximizing the effectiveness of acquiring new customers.

[1222] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1223] Step 1:

[1224] The server collects data such as historical documents and feedback from the database. It uses the Python pandas library for this purpose. The input is customer data stored in the database, which is retrieved, sensitive information is removed, and it is transformed into a clean dataset. The output is pre-processed data, formatted into a format that the AI ​​can learn from.

[1225] Step 2:

[1226] The server uses a sentiment engine (e.g., IBM Watson Tone Analyzer) to analyze the collected feedback data. The input is feedback text, which the sentiment engine analyzes to generate sentiment data (e.g., joy 80%). The output is the analyzed sentiment data, which is used for subsequent customization.

[1227] Step 3:

[1228] The server uses an API to retrieve new customer information from an integrated data platform (e.g., Domo). The input is an API request, and the output is company information in JSON format. The server receives this data and analyzes the characteristics of the company.

[1229] Step 4:

[1230] The server uses the scikit-learn library to perform clustering analysis on company characteristic data. The input is the company information obtained in step 3, which is then analyzed and classified using a clustering method. The output is cluster (segment) information for each company.

[1231] Step 5:

[1232] AI (e.g., OpenAI GPT-4) uses provided company characteristics data to anticipate potential challenges for each company and select the optimal proposal template. The input is company characteristics data and historical templates, and the system generates a proposal. The output is a customized proposal template.

[1233] Step 6:

[1234] The emotion engine adjusts the content of proposals and proposal emails based on past emotion data. The input is a customized proposal template and emotion data, which is used to fine-tune the content. The output is a proposal and proposal email customized according to the emotion.

[1235] Step 7:

[1236] The server sends the generated proposal and proposal email using the SMTP protocol. The input is a customized proposal and proposal email, intended to be sent to the specified email address. The output is the transmission log.

[1237] Step 8:

[1238] The server embeds tracking pixels into emails to track open rates and link clicks. The input is the sent email, to which the tracking pixels are added. The output is recorded as a tracking log and displayed on the dashboard.

[1239] Step 9:

[1240] Users check email open rates and link click rates on a dashboard (e.g., Domo). Input is real-time updated tracking data. Users refer to this data to consider their next follow-up action. Output is the decision on the follow-up action.

[1241] Step 10:

[1242] The user takes follow-up actions based on suggestions from the emotion engine. The input is the suggestion itself, which the user may use to send additional emails or schedule meetings. The output is the execution of the follow-up action.

[1243] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1244] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1245] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[1246] [Third Embodiment]

[1247] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[1248] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1249] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1250] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[1251] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1252] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1253] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1254] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1255] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1256] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1257] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1258] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[1259] Modes for carrying out the invention

[1260] This invention relates to a system that utilizes AI and an integrated data platform to automatically generate proposals for acquiring new customers and to send proposal emails and materials in bulk. The embodiments for carrying out the invention will be described in detail below.

[1261] System Configuration

[1262] This system consists of a database, server, AI, API, and user terminals.

[1263] database

[1264] Accumulate past proposals and related documents.

[1265] server

[1266] Data acquisition and preprocessing

[1267] Customer data acquisition and analysis

[1268] AI-powered proposal generation and email creation.

[1269] Mass email sending

[1270] Situation tracking

[1271] AI

[1272] Learning for Proposal Writing

[1273] Anticipated challenges for each company

[1274] Automated proposal and email generation

[1275] API

[1276] Data integration with integrated data platforms (e.g., Compass, Domo)

[1277] User terminal

[1278] Review of proposal and email content

[1279] Execute follow-up actions

[1280] Operating Procedure

[1281] Step 1: Data Collection and Preprocessing

[1282] server

[1283] Past proposals and related documents are collected from the database, confidential information is removed, and then the data is preprocessed for AI training. Specifically, filtering is performed to reduce unnecessary information.

[1284] Specific example

[1285] The server scans the "Proposals" folder from the database and centrally collects all files. Then, it uses the Python pandas library to identify and filter data containing sensitive information.

[1286] Step 2: Gathering and analyzing information on untapped customers

[1287] server

[1288] Use APIs to retrieve information on untapped companies (industry, number of employees, sales, etc.) from an integrated data platform.

[1289] We analyze the characteristics of companies and provide that data to AI.

[1290] Specific example

[1291] The server sends a request to the integrated data platform's API to retrieve information about company A. Then, using the scikit-learn library, the retrieved information is clustered and identified as belonging to the "IT industry".

[1292] Step 3: Proposal Generation

[1293] server

[1294] The AI ​​anticipates the challenges specific to each company and selects a proposal template based on those challenges.

[1295] By embedding company-specific information into templates, customized proposals are automatically generated.

[1296] Specific example

[1297] The AI ​​selects a template related to "IT security," fills in specific information such as "number of employees at company A" and "sales revenue," and generates a proposal tailored to company A.

[1298] Step 4: Generating the proposal email

[1299] server

[1300] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[1301] Specific example

[1302] The AI ​​generates a proposal email containing the message, "We propose security enhancements for company A." This email includes a proposal document attached in PDF format.

[1303] Step 5: Send out proposals and emails to all recipients.

[1304] server

[1305] The proposal and proposal email will be sent via the mail server based on the mass mailing list.

[1306] Record transmission logs and manage status.

[1307] Specific example

[1308] The server sends the proposal and proposal email to "Company A" using the SMTP protocol. Upon successful transmission, the status is logged.

[1309] Step 6: Follow-up

[1310] User

[1311] You can check email open rates and proposal view rates through the Domo dashboard.

[1312] Send follow-up emails or schedule meetings as needed.

[1313] Specific example

[1314] The user confirms in the Domo dashboard that the email sent to "Company A" has been opened, and the next step is to schedule a follow-up meeting.

[1315] conclusion

[1316] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. In addition, tracking of sending status and follow-up functions improve the overall effectiveness and speed of sales activities.

[1317] The following describes the processing flow.

[1318] Step 1: Data Collection and Preprocessing

[1319] server

[1320] Collect past proposals and related documents from the database. Specifically, use the Python pandas library to execute appropriate queries and retrieve the data.

[1321] Remove confidential information from the acquired data. Filter based on specific keywords or patterns to remove personal information and corporate confidential information.

[1322] The pre-processed data is provided to the AI. The data is converted to a standard format such as CSV and saved as training data for the AI.

[1323] Step 2: Gathering and analyzing information on untapped customers

[1324] server

[1325] Obtain information on untapped companies via the integrated data platform's API. For example, send an HTTP request and receive company information (in JSON format).

[1326] Based on the acquired data, we will analyze the characteristics of the companies. We will use the scikit-learn library to perform clustering analysis and feature extraction.

[1327] The analysis results are provided to the AI. The company characteristics data is converted into a format that the AI ​​can use as reference data for generating proposals (e.g., a data frame format).

[1328] Step 3: Select and generate a proposal template

[1329] server

[1330] AI anticipates potential challenges for each company. Based on company characteristic data, it identifies challenges by comparing them with similar past cases.

[1331] Based on the task, a proposal template is selected. Proposal templates are provided in advance, and the AI ​​selects the appropriate one.

[1332] The template is used to embed company-specific information and generate customized proposals. Specific data (e.g., company revenue or challenges) is embedded in the template's placeholders.

[1333] Step 4: Generating the proposal email

[1334] server

[1335] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[1336] Select the necessary email template and fill in your company's specific content. This will generate customized proposal emails tailored to individual companies.

[1337] Combine the proposal email and proposal document into a single package. Attach the proposal document in PDF format to the email and set the email body accordingly.

[1338] Step 5: Send out proposals and emails to all recipients.

[1339] server

[1340] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[1341] Record transmission logs for each recipient and manage their status. Save success / failure results to a log file for later review.

[1342] Step 6: Tracking email open and view rates

[1343] server

[1344] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[1345] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[1346] Step 7: Follow-up Action

[1347] User

[1348] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[1349] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[1350] (Example 1)

[1351] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1352] The traditional process of creating and sending proposals for acquiring new customers is time-consuming and labor-intensive. Furthermore, manual data collection, analysis, and email sending are inefficient and prone to human error. Additionally, the lack of follow-up and tracking after sending results in low effectiveness of sales activities.

[1353] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1354] This invention includes a server that collects data from a database, removes confidential information, and provides it to the AI; a server that acquires untapped company information via an API, analyzes the characteristics of the companies, and provides it to the AI; a server that anticipates potential challenges for each company and selects and generates proposal templates based on those challenges; a server that generates proposal emails based on the generated proposals and sends them together with the proposals; and a server that tracks the open and viewed status of the sent emails to enable follow-up. This enables the automatic generation and mass sending of proposals and effective follow-up.

[1355] (List of definitions)

[1356] A "database" is a storage device used to accumulate past proposals and related documents.

[1357] "AI" refers to an artificial intelligence system used for automated decision-making and proposal generation.

[1358] An "API" stands for Application Programming Interface, which is used to exchange data between external systems.

[1359] "Untapped company information" refers to data about potential new customer companies, including industry, number of employees, and sales revenue.

[1360] Clustering analysis is a process that uses the scikit-learn library to classify, analyze, and group company data.

[1361] A "template" is a pre-defined form used by AI to generate proposals, and it is a format for embedding company-specific information.

[1362] A "proposal" is a document that outlines a customized proposal tailored to a specific company.

[1363] A "proposal email" is a message used to send the contents of a generated proposal as an email.

[1364] An "SMTP server" is a server used to send emails using the Simple Mail Transfer Protocol.

[1365] The "Domo Dashboard" is a visualization tool for tracking email open rates and proposal view rates.

[1366] "Follow-up" refers to the process of taking additional actions or making contact regarding submitted proposals or emails.

[1367] "Filtering" is the process of removing information from data that should not be released externally.

[1368] Modes for carrying out the invention

[1369] This invention relates to a system for automatically generating and efficiently sending proposals and emails for acquiring new customers. In particular, it aims to automate proposal creation, analyze company characteristics, and improve the efficiency of email sending by utilizing AI and an integrated data platform. The specific methods for realizing this system are described below.

[1370] This system consists of a database, servers, AI, APIs, and user terminals.

[1371] database

[1372] A storage device is used to accumulate past proposals and related documents. This allows the AI ​​to learn from past proposal data and generate more accurate proposals.

[1373] server

[1374] The server is responsible for several processes, including the following:

[1375] 1. Data acquisition and preprocessing:

[1376] The server collects proposals and related documents from the database, filters the data using the Python pandas library, and removes confidential information.

[1377] Specific example: The server executes an SQL query against the database to retrieve proposal data. This data is then read using the pandas library, and confidential information is filtered out.

[1378] 2. Collection and analysis of untapped company information:

[1379] The server uses APIs to retrieve information on untapped companies from integrated data platforms (e.g., Compass, Domo). This information includes industry, number of employees, and revenue.

[1380] The server uses the scikit-learn library to perform clustering analysis on company characteristics and provides the results to the AI.

[1381] Specific example: A server sends an API request to retrieve company characteristic data. Clustering analysis is performed using the scikit-learn library to identify company characteristics.

[1382] 3. Proposal generation:

[1383] The AI ​​anticipates the challenges specific to each company and selects a proposal template based on those challenges. It then generates a customized proposal by embedding company-specific information into the template.

[1384] Specific example: The AI ​​selects an appropriate template from a template database and uses the Jinja2 template engine to embed company information.

[1385] 4. Generating and sending proposal emails:

[1386] Based on the proposal document, the AI ​​generates a proposal email. The generated email includes a summary of the proposal and the reasons for the proposal.

[1387] The server uses the SMTP protocol to send proposal emails and proposal documents in bulk. It records transmission logs and manages the status.

[1388] Specific example: AI uses an email template to generate an email with a proposal PDF attached. The server sends the email via SMTP server and records the sending log.

[1389] User terminal

[1390] The user terminal is a device used to review proposals and email content, and to perform follow-up actions. The Domo dashboard allows users to check email open rates and proposal viewing rates, and to send follow-up emails or schedule meetings as needed.

[1391] Example prompt statements

[1392] Data collection:

[1393] "Write Python code to scan the 'Proposals' folder in the database and collect all files. Then, use pandas to identify and remove columns that contain sensitive information."

[1394] Collection and analysis of corporate information:

[1395] "Write Python code that retrieves information from company A's API and performs clustering analysis using scikit-learn."

[1396] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. In addition, tracking of sending status and follow-up functions improve the overall effectiveness and speed of sales activities.

[1397] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1398] Program processing flow

[1399] Step 1: Data Collection and Preprocessing

[1400] Input: Database containing past proposals and related documents.

[1401] Output: Pre-processed data with sensitive information removed, ready to be provided to the AI.

[1402] Specific actions:

[1403] 1. The server connects to the database and uses SQL queries to collect proposal data.

[1404] 2. After obtaining the proposal data, the data is filtered using the Python pandas library to remove confidential information.

[1405] 3. Save the filtered data to provide it to the AI ​​model.

[1406] Specific data processing:

[1407] The server extracts proposal data from the database using SQL queries. This data is then read into pandas, and specific columns or values ​​are filtered to remove sensitive information. Finally, the pre-processed data is saved as a file.

[1408] Specific example:

[1409] The server downloads all files in the "Proposal" folder at once. The pandas drop function is used to remove columns containing confidential information.

[1410] Step 2: Gathering and analyzing information on untapped companies

[1411] Input: Company information obtained via API

[1412] Output: Analyzed company characteristics data

[1413] Specific actions:

[1414] 1. The server sends a request to the integrated data platform's API to retrieve data on untapped companies.

[1415] 2. Perform clustering analysis on the acquired data using the scikit-learn library.

[1416] 3. Save the analysis results to the database and prepare them for provision to the AI.

[1417] Specific data processing:

[1418] The server uses an API to retrieve company data (e.g., industry, number of employees, sales). This data is then analyzed using scikit-learn's clustering algorithm to group company characteristics. The analysis results are saved for use with the AI.

[1419] Specific example:

[1420] The server retrieves data from company A via an API request, performs clustering analysis using the scikit-learn KMeans algorithm, and identifies company A's industry.

[1421] Step 3: Proposal Generation

[1422] Input: Company characteristics data and past proposal templates

[1423] Output: Customized proposal

[1424] Specific actions:

[1425] 1. The server uses AI to anticipate the challenges specific to each company and selects a proposal template based on those challenges.

[1426] 2. Create customized proposals by embedding company-specific information into proposal templates.

[1427] Specific data processing:

[1428] The AI ​​selects an appropriate proposal template based on company characteristics data and uses the Jinja2 template engine to embed company information.

[1429] Specific example:

[1430] The AI ​​selects a template related to "IT security" from a template database and fills in specific information about company A (number of employees, sales revenue).

[1431] Step 4: Generating the proposal email

[1432] Input: Generated proposal

[1433] Output: Proposal email

[1434] Specific actions:

[1435] 1. The AI ​​analyzes the generated proposal and creates a proposal email.

[1436] 2. Attach the proposal PDF file to the email you created.

[1437] Specific data processing:

[1438] The AI ​​extracts necessary information from the generated proposal and inserts it into an email template. The proposal PDF is then attached to the email to complete the process.

[1439] Specific example:

[1440] The AI ​​generates an email stating, "We propose security enhancements for company A," and attaches a PDF of the proposal.

[1441] Step 5: Send out proposals and emails to all recipients.

[1442] Input: Proposal email and proposal PDF

[1443] Output: Recording of transmission logs and status information

[1444] Specific actions:

[1445] 1. The server uses the SMTP protocol to send proposal emails and proposal PDFs simultaneously.

[1446] 2. Log the status and manage it when the transmission is complete.

[1447] Specific data processing:

[1448] Send emails via SMTP server. The sending status and details of the email are recorded in the sending log.

[1449] Specific example:

[1450] The server uses SMTP to send the proposal and proposal email to company A and records it in the transmission log.

[1451] Step 6: Follow-up

[1452] Input: Email open status and proposal viewing status

[1453] Output: Follow-up actions (e.g., sending an email, scheduling a meeting)

[1454] Specific actions:

[1455] 1. Users can check email open rates and proposal view rates through the Domo dashboard.

[1456] 2. Send follow-up emails or schedule meetings as needed.

[1457] Specific data processing:

[1458] Check the open and view rates on the Domo dashboard and take the next action (send a follow-up email, schedule a meeting).

[1459] Specific example:

[1460] The user confirms on the Domo dashboard that "the email sent to company A has been opened" and sets up a follow-up meeting.

[1461] (Application Example 1)

[1462] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1463] Traditional customer proposal creation systems required manual proposal creation and customer information analysis, which was time-consuming and labor-intensive. Furthermore, the process of generating personalized proposals was cumbersome, hindering the efficiency of new customer acquisition. Additionally, there was a lack of management functions to track the effectiveness of generated proposals and provide appropriate follow-up.

[1464] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1465] This invention includes a server that provides data to an AI after collecting information from a database and removing confidential information; a server that provides data to an AI after obtaining untapped company information via an API, analyzing the characteristics of the companies, and providing this information to the AI; a server that allows the AI ​​to anticipate potential challenges for each company and select and generate proposal templates based on those challenges; a server that allows the AI ​​to generate proposal emails based on the generated proposals and sends them together with the proposals; a server that tracks the open and viewed status of sent emails to enable follow-up; a server that collects customer information from a sales site, analyzes it based on past purchase history and visit data, and generates customized product proposals as part of the proposals; and a server that generates personalized proposal emails based on the customized product proposals, sends them all at once via a mail server, and provides in-app notifications. This makes it possible to automatically and efficiently generate personalized proposals and proposal emails for new customers and send them all at once. In addition, it is possible to track the open and viewed status of emails and perform effective follow-up.

[1466] A "database" is a data storage system that centrally stores information such as past documents and proposals, and allows data to be retrieved as needed.

[1467] "Confidential information" refers to non-public information about a company or its customers that should not be leaked to the outside.

[1468] "AI" is an abbreviation for artificial intelligence, which refers to technologies that automate or optimize specific tasks through data analysis and machine learning.

[1469] "API" stands for Application Programming Interface, which is an interface for exchanging data and functions between different software programs.

[1470] "Untapped company information" refers to information about companies with which no business transactions have yet been concluded, including industry, number of employees, and sales figures.

[1471] A "template" is a basic model used when creating documents such as proposals or emails, and it includes a specific format and content.

[1472] A "proposal" is a document that formalizes a proposal made to a company or customer, and it includes details of the proposal, the reasons for it, and specific benefits.

[1473] A "proposal email" is an email sent to a customer to which a proposal and related information are presented. It is an email that outlines the proposal and its benefits.

[1474] A "sales site" refers to a website or online platform used for selling products.

[1475] "Customer information" refers to all information about a customer, such as purchase history and website visit history.

[1476] "Customized product recommendations" refer to suggestions tailored to the customer's needs, based on their purchase history and visit data, to propose the most suitable products for that customer.

[1477] A "personalized suggestion email" is a suggestion email that contains content that is individually customized for a specific customer.

[1478] "Follow-up" refers to ongoing support and tracking activities conducted after the initial proposal, primarily to check email open rates and the effectiveness of the proposed content.

[1479] This invention relates to a system for automatically generating personalized product proposals and sending them out via email and accompanying materials to customers on a sales website in order to acquire new customers. The embodiments for carrying out the invention will be described in detail below.

[1480] System Configuration

[1481] This system consists of a database, server, AI, API, and user terminals.

[1482] database

[1483] The system stores information such as past proposals, purchase history, and visit data.

[1484] server

[1485] Data acquisition and preprocessing

[1486] Customer data acquisition and analysis

[1487] AI-powered proposal generation and email creation.

[1488] Mass email sending and in-app notifications

[1489] Situation tracking

[1490] AI

[1491] Learning for Proposal Writing

[1492] Anticipating challenges for each customer

[1493] Automated proposal and email generation

[1494] API

[1495] Data integration with the integrated data platform

[1496] User terminal

[1497] Review of proposal and email content

[1498] Execute follow-up actions

[1499] Program processing

[1500] The server first collects past purchase history and visit data from the database and performs preprocessing. Specifically, it uses the Python pandas library to format the data and filters out sensitive information. Next, it retrieves untapped customer information via the integrated data platform's API. The collected customer information is classified using clustering analysis with the scikit-learn library.

[1501] The AI ​​selects a suitable product suggestion template for each categorized customer and generates customized product suggestions based on their purchase history and interests. During this process, a generation AI model is used to generate prompts and create specific suggestion content. The generated suggestion emails are sent to customers simultaneously via a mail server, along with in-app notifications.

[1502] Hardware and software

[1503] Hardware:

[1504] server

[1505] User terminal

[1506] software:

[1507] Python

[1508] pandas

[1509] scikit-learn

[1510] smtplib

[1511] request

[1512] Specific example

[1513] For example, when using this system to provide personalized product recommendations to a new customer G1, the server retrieves information on the untapped customer G1 via the integrated data platform's API and classifies the customer's characteristics using clustering analysis. The AI ​​then generates a "health and wellness-related" product recommendation document based on G1's past purchase history and visit data, and creates a personalized recommendation email for G1. The generated email is sent via a mail server, and in-app notifications are also sent.

[1514] Example of a prompt:

[1515] "Please generate an email suggesting health and wellness products based on the products the customer has purchased and the pages they have viewed. The email should include explanations of why the suggested products are suitable for the customer's needs."

[1516] This makes it possible to automatically and efficiently generate and send personalized product proposals and promotional emails to new customers on the sales site. Furthermore, it allows for tracking email open rates and viewing activity, enabling effective follow-up.

[1517] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1518] Step 1:

[1519] The server collects past purchase history and visit data from the database. Specifically, it reads and formats this data using the Python pandas library. Since the collected data may contain sensitive information, it is filtered to remove any confidential information.

[1520] Input: Purchase history and visit data from the database

[1521] Output: Formatted and sensitive data removed dataset

[1522] Step 2:

[1523] The server retrieves untapped customer information via the integrated data platform's API. Specifically, it sends API requests to obtain information such as the customer's industry, number of employees, and sales revenue. The retrieved data is then read using the pandas library and converted into the appropriate format.

[1524] Input: API request for the integrated data platform

[1525] Output: Dataset of characteristics information for untapped customers

[1526] Step 3:

[1527] The server performs clustering analysis on the acquired customer information using the scikit-learn library. This clustering analysis classifies the customers into clusters based on their characteristics.

[1528] Input: Dataset of characteristics information for untapped customers

[1529] Output: Customer data with cluster labels assigned.

[1530] Step 4:

[1531] The AI ​​selects the optimal proposal template for each customer based on cluster labels. After selecting a template, it automatically generates a customized proposal by embedding customer characteristic information.

[1532] Input: Customer data with cluster labels assigned.

[1533] Output: Customized proposal

[1534] Step 5:

[1535] The AI ​​creates a personalized proposal email based on the generated proposal. The proposal email includes a summary of the proposal and the reasons for the proposal.

[1536] Input: Customized proposal

[1537] Output: Personalized suggestion email

[1538] Step 6:

[1539] The server sends proposals and proposal emails via the mail server based on a mass mailing list. It also simultaneously sends in-app notifications. It records sending logs and manages the status.

[1540] Input: Personalized suggestion email

[1541] Output: Sending log, in-app notification status

[1542] Step 7:

[1543] Users can check email open rates and proposal viewing rates through the dashboard. They can also send follow-up emails or schedule meetings as needed.

[1544] Input: Sending log, in-app notification status

[1545] Output: Follow-up actions, such as follow-up emails or meeting scheduling.

[1546] Through the above processing steps, this system can automatically and efficiently generate personalized proposals and emails for new customers and send them in bulk. Furthermore, it can track email open and view rates, enabling effective follow-up.

[1547] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1548] Modes for carrying out the invention

[1549] This invention relates to a system that utilizes AI, an integrated data platform, and an emotion engine to automatically generate proposals and emails for acquiring new customers, and further customizes them based on the user's emotions. The embodiments for carrying out the invention will be described in detail below.

[1550] System Configuration

[1551] This system consists of a database, server, AI, API, emotion engine, and user terminal.

[1552] database

[1553] It stores past proposals, related documents, and user sentiment data.

[1554] server

[1555] Data acquisition and preprocessing

[1556] Customer data acquisition and analysis

[1557] AI-powered proposal generation and email creation.

[1558] Emotional engine integration

[1559] Mass email sending

[1560] Situation tracking

[1561] AI

[1562] Learning for Proposal Writing

[1563] Anticipated challenges for each company

[1564] Automated proposal and email generation

[1565] API

[1566] Data integration with integrated data platforms (e.g., Compass, Domo)

[1567] Emotional Engine

[1568] We analyze user emotions and provide emotion data in real time. Based on this emotion data, we customize proposals and email content.

[1569] User terminal

[1570] Review of proposal and email content

[1571] Execute follow-up actions

[1572] Operating Procedure

[1573] Step 1: Data Collection and Preprocessing

[1574] server

[1575] Past proposals and related documents are collected from the database, confidential information is removed, and then the data is preprocessed for AI training. For example, filtering is performed to reduce unnecessary information.

[1576] The emotion engine is used to analyze user sentiment data regarding past suggestion emails and store it in a database.

[1577] Specific example

[1578] The server scans the "Proposals" folder from the database and retrieves the data using the Python pandas library. It filters out sensitive information and provides the preprocessed data to the AI. Additionally, an emotion engine analyzes user responses in past emails and stores the emotion data in the database.

[1579] Step 2: Gathering and analyzing information on untapped customers

[1580] server

[1581] We will use an API to retrieve information on untapped companies from an integrated data platform. Specifically, we will send an HTTP request and receive company information (in JSON format).

[1582] Based on the acquired data, we analyze the characteristics of the company. For example, we perform clustering analysis using the scikit-learn library.

[1583] The analysis results are provided to the AI. The company characteristics data is used by the AI ​​as reference data for generating proposals.

[1584] Specific example

[1585] The server sends a request to the integrated data platform's API to retrieve information about company A. Then, using scikit-learn, it identifies that company A belongs to the "IT industry".

[1586] Step 3: Select and generate a proposal template

[1587] server

[1588] The AI ​​anticipates potential challenges for each company and selects a proposal template based on those challenges. The optimal template is chosen based on company characteristic data.

[1589] The system automatically generates customized proposals by embedding company-specific information into proposal templates.

[1590] Using an emotion engine, template selection and proposal content are adjusted based on the user's past emotional data.

[1591] Specific example

[1592] The AI ​​selects a template related to "IT security" and fills in specific information about company A, such as "number of employees" and "sales revenue." At the same time, the emotion engine refers to past emotion data and adjusts the content to elicit a more positive response.

[1593] Step 4: Generating the proposal email

[1594] server

[1595] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[1596] Select the necessary email template, embed company-specific content, and customize the email content by reflecting sentiment data analyzed by the sentiment engine.

[1597] Combine the proposal email and proposal document into a single package.

[1598] Specific example

[1599] The AI ​​generates a proposal email containing the message, "We propose strengthening security for company A," and attaches the proposal as a PDF document. The emotion engine adjusts the wording and tone of the email to suit the user's preferences.

[1600] Step 5: Send out proposals and emails to all recipients.

[1601] server

[1602] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[1603] Record transmission logs for each recipient and manage their status.

[1604] Specific example

[1605] The server uses the SMTP protocol to send the proposal and proposal email to "Company A," and logs the status upon successful transmission.

[1606] Step 6: Tracking email open and view rates

[1607] server

[1608] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[1609] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[1610] Specific example

[1611] The server embeds tracking pixels in emails to detect open rates, which are then displayed in real time on the Domo dashboard.

[1612] Step 7: Follow-up Action

[1613] User

[1614] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[1615] Based on the data analyzed by the emotion engine, the next follow-up action will be suggested.

[1616] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[1617] Specific example

[1618] The user confirms in the Domo dashboard that the email sent to "Company A" was opened, and based on the sentiment engine's suggestions, sets up a follow-up meeting as the next step.

[1619] conclusion

[1620] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. Tracking and follow-up features improve the overall effectiveness and speed of sales activities. In addition, the integration of an emotion engine enables customization based on user emotions, further enhancing the accuracy and effectiveness of proposals.

[1621] The following describes the processing flow.

[1622] Step 1: Data Collection and Preprocessing

[1623] server

[1624] Collect past proposals and related documents from the database. Specifically, use the Python pandas library to execute appropriate queries and retrieve the data.

[1625] Remove confidential information from the acquired data. For example, use regular expressions based on specific keywords or patterns to filter out confidential information and remove personal and corporate confidential information.

[1626] The pre-processed data is provided to the AI. The data is converted to a standard format such as CSV and saved as a training dataset for the AI.

[1627] In addition, an emotion engine is used to analyze users' emotional data regarding past suggestion emails and store it in a database. The emotion engine uses natural language processing technology to analyze the content of emails and assigns emotional labels such as positive, negative, and neutral.

[1628] Step 2: Gathering and analyzing information on untapped customers

[1629] server

[1630] We will use an API to retrieve information on untapped companies from an integrated data platform. Specifically, we will send an HTTP request and receive company information (in JSON format).

[1631] Based on the acquired data, we analyze the characteristics of companies. For example, we use the scikit-learn library to perform clustering analysis and feature extraction to reveal characteristics such as the company's industry, size, and sales revenue.

[1632] The analysis results are provided to the AI. The company characteristics data is converted into a format that the AI ​​can use as reference data for generating proposals (e.g., a data frame format).

[1633] Step 3: Select and generate a proposal template

[1634] server

[1635] AI anticipates potential challenges for each company. Based on company characteristic data, it compares it with similar past cases and industry trends to identify the most relevant issues.

[1636] Based on the selected issue, a proposal template is chosen. Proposal templates are prepared in advance, and the AI ​​selects the appropriate one.

[1637] The template is used to embed company-specific information and generate customized proposals. Specific data (e.g., company revenue or challenges) is embedded in the template's placeholders.

[1638] Using an emotion engine, template selection and proposal content are adjusted based on the user's past emotional data. For example, expressions and designs that received many positive responses are prioritized.

[1639] Step 4: Generating the proposal email

[1640] server

[1641] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[1642] Select the necessary email template and fill in your company's specific content. The sentiment engine analyzes emotional data to customize the email content for greater effectiveness. For example, it adopts phrasing and tones that elicit positive responses from users.

[1643] The proposal email and proposal document will be packaged together as a single package, with the proposal document attached to the email in PDF format.

[1644] Step 5: Send out proposals and emails to all recipients.

[1645] server

[1646] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[1647] Record transmission logs for each recipient and manage their status. Save success / failure results to a log file for later review.

[1648] Step 6: Tracking email open and view rates

[1649] server

[1650] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[1651] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[1652] Step 7: Follow-up Action

[1653] User

[1654] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[1655] Based on the data analyzed by the emotion engine, the next follow-up action is suggested. For example, if the user showed a positive response to the previous email, a more assertive approach is suggested.

[1656] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[1657] (Example 2)

[1658] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1659] This invention relates to a system that significantly reduces the time and effort required for acquiring new customers in a company's sales activities and makes proposals to customers more effective. Conventional systems require a great deal of time and effort to create proposals and generate proposal emails to customers, and they are not sufficiently customized based on customer sentiment. Therefore, there has been a need for a system that supports new customer acquisition in an efficient and effective way.

[1660] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1661] This invention includes a server that provides data to an AI after collecting information from a database and removing confidential information; a server that provides data to an AI after obtaining untapped company information via an API and analyzing the characteristics of the companies and providing this information to the AI; a server that allows the AI ​​to anticipate potential challenges for each company and select and generate proposal templates based on those challenges; a server that allows the AI ​​to generate proposal emails based on the generated proposals and sends them together with the proposals; a server that tracks the open and viewed status of sent emails and enables follow-up; a server that uses an emotion analysis engine to accumulate past emotional data of users and customize the content of proposals and proposal emails; a server that sends proposal emails and proposals all at once; a server that uses a Domo dashboard to display the sending status in real time; and a server that suggests follow-up actions. This enables the efficient and effective generation and sending of proposals and proposal emails, and makes it easier to realize optimal proposals based on the user's emotions.

[1662] "Document collection" refers to the act of obtaining past proposals and related documents from a database.

[1663] "Removal of confidential information" is the process of filtering and removing information from collected materials that should not be released to the public.

[1664] "Means of providing data to AI" refers to the process of inputting pre-processed data into a machine learning system.

[1665] "Acquiring company information" means collecting data on untapped companies via APIs.

[1666] "Company characteristic analysis" refers to evaluating the characteristics of each company based on acquired company information using methods such as clustering analysis.

[1667] "Identifying challenges" refers to AI predicting potential problems and areas for improvement specific to each company.

[1668] "Selecting a proposal template" is a method of choosing the most suitable proposal format based on the anticipated challenges.

[1669] "Proposal generation" refers to the process of creating a customized proposal by embedding company-specific information into a template.

[1670] "Proposal email generation" refers to the process where AI creates an email based on the generated proposal document, outlining the proposal's summary and reasons.

[1671] "Mass email sending" refers to the process of simultaneously sending a generated proposal document and proposal email to multiple recipients based on a list.

[1672] "Open tracking" refers to tracking and recording whether or not an email that has been sent has been opened.

[1673] "Follow-up action" refers to the act of conducting follow-up surveys or making additional contact with users after sending a proposal email, depending on their situation.

[1674] A "sentiment analysis engine" is a system that analyzes a user's past emotional data and optimizes the suggested content based on that data.

[1675] The "Domo Dashboard" is a tool for visualizing data and displaying real-time tracking information.

[1676] This invention relates to a system that utilizes AI, an integrated data platform, and an emotion analysis engine to automatically generate proposals and emails for acquiring new customers, and further customizes them based on the user's emotions. The detailed configuration and operation of the system are described below.

[1677] System Configuration

[1678] This system consists of a database, server, AI, API, sentiment analysis engine, and user terminal.

[1679] database

[1680] It stores past proposals, related documents, and user sentiment data.

[1681] server

[1682] It performs data collection and preprocessing, customer data acquisition and analysis, AI-powered proposal generation and email creation, sentiment analysis engine integration, mass email sending, and status tracking.

[1683] AI

[1684] This system provides learning opportunities for proposal creation, anticipates challenges specific to each company, and automatically generates proposals and emails.

[1685] API

[1686] Establish data integration with integrated data platforms (e.g., Compass, Domo).

[1687] Emotion analysis engine

[1688] We analyze user emotions and provide emotion data in real time. Based on this emotion data, we customize proposals and email content.

[1689] User terminal

[1690] Review the proposal and email content, and take follow-up actions.

[1691] Operating Procedure

[1692] The following explains how the system works using specific examples for each step.

[1693] Step 1: Data Acquisition and Preprocessing

[1694] server

[1695] The server collects past proposals and related documents from the database and removes confidential information. This process uses the Python pandas library to load, filter, and clean the data. It also uses a sentiment analysis engine to analyze user sentiment data regarding past proposal emails and stores it in the database.

[1696] Specific example

[1697] The server scans the "Proposal" folder from the database, retrieves the data using the pandas library, filters it, and provides the preprocessed data to the AI. The sentiment analysis engine analyzes the user's reactions contained in past emails and stores the sentiment data in the database.

[1698] Step 2: Gathering and analyzing information on untapped customers

[1699] server

[1700] The server uses an API to retrieve information on untapped companies from the integrated data platform. Specifically, it sends an HTTP request and receives company information in JSON format. Then, it uses the scikit-learn library to perform clustering analysis, extract company characteristics, and provide them to the AI.

[1701] Specific example

[1702] The server sends an HTTP request to the integrated data platform's API to retrieve information about company A, and then uses scikit-learn to identify that company A belongs to the "IT industry".

[1703] Step 3: Select and generate a proposal template.

[1704] server

[1705] The AI ​​anticipates potential challenges for each company and selects proposal templates based on those challenges. Based on company characteristic data, it selects the optimal template, embeds company-specific information into the template, and automatically generates a customized proposal. In addition, it uses an emotion analysis engine to adjust template selection and proposal content based on the user's past emotion data.

[1706] Specific example

[1707] The AI ​​selects a template related to "IT security" and fills in information such as company A's "number of employees" and "sales revenue." Simultaneously, an emotion analysis engine refers to past emotion data and adjusts the content to elicit a positive response from the user.

[1708] Step 4: Generate the proposal email

[1709] server

[1710] Based on the generated proposal document, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its proposal, and its content is customized using a sentiment analysis engine. The proposal email and proposal document are then combined into a single package.

[1711] Specific example

[1712] The AI ​​generates a proposal email containing the message, "We propose strengthening security for company A," and attaches the proposal as a PDF document. An emotion analysis engine adjusts the wording and tone of the email to suit the user's preferences.

[1713] Step 5: Send proposals and emails to all recipients.

[1714] server

[1715] The server uses the SMTP protocol to send proposals and proposal emails in bulk. Specifically, it uses the smtplib library to send emails based on a mailing list. It also records sending logs and manages the sending status.

[1716] Specific example

[1717] The server uses the SMTP protocol to send the proposal and proposal email to "Company A," and logs the status upon successful transmission.

[1718] Step 6: Track email open and view rates

[1719] server

[1720] The server uses embedded tracking pixels to track email open rates and proposal viewing rates. The collected tracking data is reflected in the Domo dashboard in real time.

[1721] Specific example

[1722] The server embeds tracking pixels in emails to detect open rates, which are then displayed in real time on the Domo dashboard.

[1723] Step 7: Follow-up Action

[1724] User

[1725] Users can check email open rates and proposal view rates through the Domo dashboard. Based on the data analyzed by the sentiment analysis engine, the system suggests the next follow-up actions. Users can then take additional follow-up actions as needed, such as sending follow-up emails or scheduling meetings.

[1726] Specific example

[1727] The user confirms on the Domo dashboard that "Company A" has opened the proposal and sets up a follow-up meeting based on the sentiment analysis engine's recommendations.

[1728] Examples of prompt statements

[1729] Select a proposal template and generate a proposal tailored to Company A. Adjust the content based on past sentiment data and output the optimal proposal in PDF format. Also, send the proposal and proposal email based on the mass email mailing list and track the open rates.

[1730] As described above, this system utilizes AI and emotion analysis engines to enable the automated generation and customization of efficient and effective proposals and emails for new customers, and further optimizes sales activities through tracking of sending status and suggesting follow-up actions.

[1731] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1732] Step 1: Data Collection and Preprocessing

[1733] server

[1734] The server collects past proposals and related documents from the database. It executes SQL queries against the database to read the relevant data. Then, it removes confidential information from the collected data. Specifically, it uses the Python pandas library to filter and clean the data. It also uses a sentiment analysis engine to analyze past proposal email data, extract user sentiment data, and store it in the database.

[1735] input

[1736] Proposals, related documents, and past proposal email data collected from the database.

[1737] output

[1738] Confidential information has been removed, and the proposal data has been pre-processed and analyzed for sentiment.

[1739] Specific actions

[1740] The server retrieves data from the database, filters the necessary data using the pandas library, removes sensitive information, and performs preprocessing.

[1741] Step 2: Gathering and analyzing information on untapped customers

[1742] server

[1743] The server uses an API to retrieve information on untapped companies from an integrated data platform. It sends an HTTP request and receives company information in JSON format. Then, it performs clustering analysis using the scikit-learn library to extract company characteristics. The characteristic data is analyzed and provided to the AI.

[1744] input

[1745] Information on untapped companies obtained via API (in JSON format).

[1746] output

[1747] Company characteristic data extracted through clustering analysis.

[1748] Specific actions

[1749] The server sends an API request to retrieve information on untapped companies. This data is then clustered and analyzed using scikit-learn to extract company characteristics.

[1750] Step 3: Select and generate a proposal template

[1751] server

[1752] The AI ​​anticipates potential challenges for each company and selects a proposal template based on those challenges. It uses company characteristic data to choose the optimal proposal template and embeds company-specific information into it. Furthermore, it uses an emotion analysis engine to adjust the proposal content based on the user's past emotional data.

[1753] input

[1754] Company characteristic data and historical sentiment data.

[1755] output

[1756] A customized proposal tailored to each company.

[1757] Specific actions

[1758] AI analyzes the data and selects the optimal proposal template. An emotion analysis engine then adjusts the content.

[1759] Step 4: Generating the proposal email

[1760] server

[1761] Based on the generated proposal, the AI ​​creates a proposal email. Using the Jinja2 template engine, it generates an email containing an overview of the proposal and the reasons for the proposal. Using an emotion analysis engine, the email content is customized based on the user's emotion data.

[1762] input

[1763] Generated proposals, past sentiment data.

[1764] output

[1765] Customized proposal email.

[1766] Specific actions

[1767] The AI ​​generates emails based on the proposal, and the sentiment analysis engine adjusts the content.

[1768] Step 5: Send out proposals and emails to all recipients.

[1769] server

[1770] The server uses the SMTP protocol to send proposals and proposal emails in bulk. It uses the smtplib library to send emails based on a designated mailing list. It records sending logs and manages the sending status.

[1771] input

[1772] Customized proposal emails, proposal documents, and mass mailing lists.

[1773] output

[1774] Log and transmission status after transmission is complete.

[1775] Specific actions

[1776] The server sends an email using the SMTP protocol and logs the status.

[1777] Step 6: Tracking email open and view rates

[1778] server

[1779] The server uses embedded tracking pixels to track email open rates and proposal view rates. The collected data is reflected in the Domo dashboard in real time.

[1780] input

[1781] A proposal email with embedded tracking pixels.

[1782] output

[1783] Tracking data on opening and browsing activity.

[1784] Specific actions

[1785] The server uses tracking pixels to detect open and viewed status and displays it on the Domo dashboard.

[1786] Step 7: Follow-up Action

[1787] User

[1788] Through the Domo dashboard, users can check email open rates and proposal view rates. Based on the data analyzed by the sentiment analysis engine, the system suggests the next follow-up actions. Users can then send follow-up emails or schedule meetings as needed.

[1789] input

[1790] Data on opening and viewing status, and analysis data from the sentiment analysis engine.

[1791] output

[1792] Follow-up actions (e.g., sending a follow-up email or scheduling a meeting).

[1793] Specific actions

[1794] Users check the situation on the Domo dashboard and take follow-up actions based on suggestions from the sentiment analysis engine.

[1795] (Application Example 2)

[1796] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[1797] Traditional methods of acquiring new customers made it difficult to process large amounts of customer data quickly and efficiently, and to make appropriate proposals to each customer. Furthermore, customizing proposals based on customer emotions and feedback was challenging, resulting in insufficient engagement with customers. This led to decreased customer engagement and difficulties in acquiring new customers.

[1798] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for collecting data from a database, removing confidential information and providing it to the AI; means for acquiring untapped company information via an API, analyzing the characteristics of the companies and providing it to the AI; means for the AI ​​to anticipate potential challenges for each company and select and generate a proposal template based on those challenges; means for having the AI ​​generate a proposal email based on the generated proposal and sending it together with the proposal; means for tracking the open and viewing status of sent emails to enable follow-up; means for collecting data such as past purchase history and feedback and analyzing sentiment data; means for customizing the content of the proposal and proposal email based on the sentiment data; and means for generating proposal emails and coupons for individual customers using a generation AI model. This makes it possible to efficiently process large amounts of data and make customized proposals based on customer sentiment.

[1799] A "database" is an information system that stores and manages data such as past documents and feedback.

[1800] "Confidential information" refers to important information that must not be leaked to outsiders.

[1801] "AI" refers to artificial intelligence, a technology that processes and analyzes large amounts of data to automatically generate suggestions.

[1802] "API" stands for Application Programming Interface, which is a gateway for different software systems to exchange data and functions.

[1803] "Company characteristics" refer to the characteristics of a company, such as its industry, size, and profitability.

[1804] A "proposal" is a document that outlines the proposed content for a specific client.

[1805] A "template" is a format that serves as a guide for efficiently creating proposals, emails, and other documents.

[1806] A "proposal email" is an email sent to a client that includes an overview of the proposal and the reasons for the proposal.

[1807] "Tracking" is the process of tracking and recording things like whether an email was opened or whether links were clicked after it was sent.

[1808] "Follow-up" refers to additional actions taken after sending a proposal email in order to maintain and deepen the relationship with the customer.

[1809] "Purchase history" refers to a record of products that a customer has purchased in the past.

[1810] "Feedback" refers to the opinions and evaluations received from customers.

[1811] "Emotional data" refers to data that analyzes customer emotions and reactions and expresses them as numerical or textual data.

[1812] A "generative AI model" is a model trained to automatically generate documents and suggestions using artificial intelligence.

[1813] A "coupon" is an electronic or paper voucher used to offer discounts or benefits to customers.

[1814] This invention relates to a new customer acquisition system comprising a database, server, AI, API, emotion engine, and user terminal. This system automatically generates proposals and proposal emails and customizes them based on user emotion data through the following steps.

[1815] The server collects data such as historical documents and feedback from the database, and preprocesses it by removing sensitive information. It uses the Python pandas library to retrieve data and filter out sensitive information. It also uses a sentiment engine to analyze user responses in past emails and generate sentiment data. IBM Watson Tone Analyzer is used as the sentiment engine.

[1816] Using an API, the server retrieves new customer information from an integrated data platform (e.g., Domo). It sends HTTP requests to obtain company information (in JSON format) and analyzes the company's characteristics. Clustering analysis is performed using the scikit-learn library. This characteristic data is used as reference when an AI (e.g., OpenAI GPT-4) generates proposals.

[1817] Using the GPT-4 generative AI model, the AI ​​anticipates potential challenges specific to each company and selects and generates the optimal proposal template based on those challenges. It then customizes the template based on sentiment data while incorporating company-specific information (such as the number of employees and sales). This process generates more effective proposals that reflect the user's emotions.

[1818] Next, the AI ​​creates a proposal email based on the generated proposal. Using sentiment data provided by the sentiment engine, it adjusts the email content to maximize customer response. This email includes a summary of the proposal and the reasons for the proposal, and is sent all at once. The server uses the SMTP protocol to send the email to all recipients simultaneously.

[1819] Tracking pixels are embedded in emails to track email open rates and link click rates. The collected tracking data is reflected in a dashboard (e.g., Domo Dashboard) in real time, allowing users to monitor the situation. Based on the data analyzed by the sentiment engine, follow-up actions can be suggested, thereby improving the effectiveness of new customer acquisition.

[1820] Specific example 1:

[1821] The server collects customer data from the database and preprocesses it using the Python pandas library. For example, it scans the "Proposals" folder and filters out unnecessary information. It also uses IBM Watson Tone Analyzer to generate sentiment data from past email feedback.

[1822] Specific example 2:

[1823] The server sends a request to the integrated data platform's API to retrieve company information. Then, clustering analysis is performed using scikit-learn to identify the new company A as belonging to the "IT industry." This data is provided to OpenAI GPT-4, where a generative AI model selects a template related to "IT security" and embeds specific information about company A into the generated proposal.

[1824] Example of a prompt message 1:

[1825] "Customer Information: Detailed Information on Company A"

[1826] "Emotion data: [{\"tone_id\": \"joy\", \"score\": 0.8}]"

[1827] "Please generate a customized proposal email based on this."

[1828] Example of a prompt message 2:

[1829] "Customer Information: Customer X's purchase history and feedback"

[1830] "Emotion data: [{\"tone_id\": \"anger\", \"score\": 0.6}]"

[1831] "Please generate a follow-up suggestion email based on this."

[1832] As a result, this system can efficiently process large amounts of data and provide customized suggestions based on customer sentiment, thereby maximizing the effectiveness of acquiring new customers.

[1833] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1834] Step 1:

[1835] The server collects data such as historical documents and feedback from the database. It uses the Python pandas library for this purpose. The input is customer data stored in the database, which is retrieved, sensitive information is removed, and it is transformed into a clean dataset. The output is pre-processed data, formatted into a format that the AI ​​can learn from.

[1836] Step 2:

[1837] The server uses a sentiment engine (e.g., IBM Watson Tone Analyzer) to analyze the collected feedback data. The input is feedback text, which the sentiment engine analyzes to generate sentiment data (e.g., joy 80%). The output is the analyzed sentiment data, which is used for subsequent customization.

[1838] Step 3:

[1839] The server uses an API to retrieve new customer information from an integrated data platform (e.g., Domo). The input is an API request, and the output is company information in JSON format. The server receives this data and analyzes the characteristics of the company.

[1840] Step 4:

[1841] The server uses the scikit-learn library to perform clustering analysis on company characteristic data. The input is the company information obtained in step 3, which is then analyzed and classified using a clustering method. The output is cluster (segment) information for each company.

[1842] Step 5:

[1843] AI (e.g., OpenAI GPT-4) uses provided company characteristics data to anticipate potential challenges for each company and select the optimal proposal template. The input is company characteristics data and historical templates, and the system generates a proposal. The output is a customized proposal template.

[1844] Step 6:

[1845] The emotion engine adjusts the content of proposals and proposal emails based on past emotion data. The input is a customized proposal template and emotion data, which is used to fine-tune the content. The output is a proposal and proposal email customized according to the emotion.

[1846] Step 7:

[1847] The server sends the generated proposal and proposal email using the SMTP protocol. The input is a customized proposal and proposal email, intended to be sent to the specified email address. The output is the transmission log.

[1848] Step 8:

[1849] The server embeds tracking pixels into emails to track open rates and link clicks. The input is the sent email, to which the tracking pixels are added. The output is recorded as a tracking log and displayed on the dashboard.

[1850] Step 9:

[1851] Users check email open rates and link click rates on a dashboard (e.g., Domo). Input is real-time updated tracking data. Users refer to this data to consider their next follow-up action. Output is the decision on the follow-up action.

[1852] Step 10:

[1853] The user takes follow-up actions based on suggestions from the emotion engine. The input is the suggestion itself, which the user may use to send additional emails or schedule meetings. The output is the execution of the follow-up action.

[1854] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1855] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1856] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1857] [Fourth Embodiment]

[1858] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1859] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1860] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1861] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1862] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1863] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1864] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1865] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1866] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1867] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1868] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1869] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1870] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1871] Modes for carrying out the invention

[1872] This invention relates to a system that automatically generates proposals for acquiring new customers and sends out proposal emails and materials in bulk, utilizing AI and an integrated data platform. The embodiments for carrying out the invention will be described in detail below.

[1873] System Configuration

[1874] This system consists of a database, server, AI, API, and user terminals.

[1875] database

[1876] Accumulate past proposals and related documents.

[1877] server

[1878] Data acquisition and preprocessing

[1879] Customer data acquisition and analysis

[1880] AI-powered proposal generation and email creation.

[1881] Mass email sending

[1882] Situation tracking

[1883] AI

[1884] Learning for Proposal Writing

[1885] Anticipated challenges for each company

[1886] Automated proposal and email generation

[1887] API

[1888] Data integration with integrated data platforms (e.g., Compass, Domo)

[1889] User terminal

[1890] Review of proposal and email content

[1891] Execute follow-up actions

[1892] Operating Procedure

[1893] Step 1: Data Collection and Preprocessing

[1894] server

[1895] Past proposals and related documents are collected from the database, confidential information is removed, and then the data is preprocessed for AI training. Specifically, filtering is performed to reduce unnecessary information.

[1896] Specific example

[1897] The server scans the "Proposals" folder from the database and centrally collects all files. Then, it uses the Python pandas library to identify and filter data containing sensitive information.

[1898] Step 2: Gathering and analyzing information on untapped customers

[1899] server

[1900] Use APIs to retrieve information on untapped companies (industry, number of employees, sales, etc.) from an integrated data platform.

[1901] We analyze the characteristics of companies and provide that data to AI.

[1902] Specific example

[1903] The server sends a request to the integrated data platform's API to retrieve information about company A. Then, using the scikit-learn library, the retrieved information is clustered and identified as belonging to the "IT industry".

[1904] Step 3: Proposal Generation

[1905] server

[1906] The AI ​​anticipates the challenges specific to each company and selects a proposal template based on those challenges.

[1907] By embedding company-specific information into templates, customized proposals are automatically generated.

[1908] Specific example

[1909] The AI ​​selects a template related to "IT security," fills in specific information such as "number of employees at company A" and "sales revenue," and generates a proposal tailored to company A.

[1910] Step 4: Generating the proposal email

[1911] server

[1912] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[1913] Specific example

[1914] The AI ​​generates a proposal email containing the message, "We propose security enhancements for company A." This email includes a proposal document attached in PDF format.

[1915] Step 5: Send out the proposal and email to everyone.

[1916] server

[1917] The proposal and proposal email will be sent via the mail server based on the mass mailing list.

[1918] Record transmission logs and manage status.

[1919] Specific example

[1920] The server sends the proposal and proposal email to "Company A" using the SMTP protocol. Upon successful transmission, the status is logged.

[1921] Step 6: Follow-up

[1922] User

[1923] You can check email open rates and proposal view rates through the Domo dashboard.

[1924] Send follow-up emails or schedule meetings as needed.

[1925] Specific example

[1926] The user confirms in the Domo dashboard that the email sent to "Company A" has been opened, and the next step is to schedule a follow-up meeting.

[1927] conclusion

[1928] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. In addition, tracking of sending status and follow-up functions improve the overall effectiveness and speed of sales activities.

[1929] The following describes the processing flow.

[1930] Step 1: Data Collection and Preprocessing

[1931] server

[1932] Collect past proposals and related documents from the database. Specifically, use the Python pandas library to execute appropriate queries and retrieve the data.

[1933] Remove confidential information from the acquired data. Filter based on specific keywords or patterns to remove personal information and corporate confidential information.

[1934] The pre-processed data is provided to the AI. The data is converted to a standard format such as CSV and saved as training data for the AI.

[1935] Step 2: Gathering and analyzing information on untapped customers

[1936] server

[1937] Obtain information on untapped companies via the integrated data platform's API. For example, send an HTTP request and receive company information (in JSON format).

[1938] Based on the acquired data, we will analyze the characteristics of the companies. We will use the scikit-learn library to perform clustering analysis and feature extraction.

[1939] The analysis results are provided to the AI. The company characteristics data is converted into a format that the AI ​​can use as reference data for generating proposals (e.g., a data frame format).

[1940] Step 3: Select and generate a proposal template

[1941] server

[1942] AI anticipates potential challenges for each company. Based on company characteristic data, it identifies challenges by comparing them with similar past cases.

[1943] Based on the task, a proposal template is selected. Proposal templates are provided in advance, and the AI ​​selects the appropriate one.

[1944] The template is used to embed company-specific information and generate customized proposals. Specific data (e.g., company revenue or challenges) is embedded in the template's placeholders.

[1945] Step 4: Generating the proposal email

[1946] server

[1947] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[1948] Select the necessary email template and fill in your company's specific content. This will generate customized proposal emails tailored to individual companies.

[1949] Combine the proposal email and proposal document into a single package. Attach the proposal document in PDF format to the email and set the email body accordingly.

[1950] Step 5: Send out the proposal and email to everyone.

[1951] server

[1952] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[1953] Record transmission logs for each recipient and manage their status. Save success / failure results to a log file for later review.

[1954] Step 6: Tracking email open and view rates

[1955] server

[1956] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[1957] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[1958] Step 7: Follow-up Action

[1959] User

[1960] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[1961] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[1962] (Example 1)

[1963] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1964] The traditional process of creating and sending proposals for acquiring new customers is time-consuming and labor-intensive. Furthermore, manual data collection, analysis, and email sending are inefficient and prone to human error. Additionally, the lack of follow-up and tracking after sending results in low effectiveness of sales activities.

[1965] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1966] This invention includes a server that collects data from a database, removes confidential information, and provides it to the AI; a server that acquires untapped company information via an API, analyzes the characteristics of the companies, and provides it to the AI; a server that anticipates potential challenges for each company and selects and generates proposal templates based on those challenges; a server that generates proposal emails based on the generated proposals and sends them together with the proposals; and a server that tracks the open and viewed status of the sent emails to enable follow-up. This enables the automatic generation and mass sending of proposals and effective follow-up.

[1967] (List of definitions)

[1968] A "database" is a storage device used to accumulate past proposals and related documents.

[1969] "AI" refers to an artificial intelligence system used for automated decision-making and proposal generation.

[1970] An "API" stands for Application Programming Interface, which is used to exchange data between external systems.

[1971] "Untapped company information" refers to data about potential new customer companies, including industry, number of employees, and sales revenue.

[1972] Clustering analysis is a process that uses the scikit-learn library to classify, analyze, and group company data.

[1973] A "template" is a pre-defined form used by AI to generate proposals, and it is a format for embedding company-specific information.

[1974] A "proposal" is a document that outlines a customized proposal tailored to a specific company.

[1975] A "proposal email" is a message used to send the contents of a generated proposal as an email.

[1976] An "SMTP server" is a server used to send emails using the Simple Mail Transfer Protocol.

[1977] The "Domo Dashboard" is a visualization tool for tracking email open rates and proposal view rates.

[1978] "Follow-up" refers to the process of taking additional actions or making contact regarding submitted proposals or emails.

[1979] "Filtering" is the process of removing information from data that should not be released externally.

[1980] Modes for carrying out the invention

[1981] This invention relates to a system for automatically generating and efficiently sending proposals and emails for acquiring new customers. In particular, it aims to automate proposal creation, analyze company characteristics, and improve the efficiency of email sending by utilizing AI and an integrated data platform. The specific methods for realizing this system are described below.

[1982] This system consists of a database, servers, AI, APIs, and user terminals.

[1983] database

[1984] A storage device is used to accumulate past proposals and related documents. This allows the AI ​​to learn from past proposal data and generate more accurate proposals.

[1985] server

[1986] The server is responsible for several processes, including the following:

[1987] 1. Data acquisition and preprocessing:

[1988] The server collects proposals and related documents from the database, filters the data using the Python pandas library, and removes confidential information.

[1989] Specific example: The server executes an SQL query against the database to retrieve proposal data. This data is then read using the pandas library, and confidential information is filtered out.

[1990] 2. Collection and analysis of untapped company information:

[1991] The server uses an API to retrieve information on untapped companies from an integrated data platform (e.g., Compass, Domo). This information includes industry, number of employees, and revenue.

[1992] The server uses the scikit-learn library to perform clustering analysis on company characteristics and provides the results to the AI.

[1993] Specific example: A server sends an API request to retrieve company characteristic data. Clustering analysis is performed using the scikit-learn library to identify company characteristics.

[1994] 3. Proposal generation:

[1995] The AI ​​anticipates the challenges specific to each company and selects a proposal template based on those challenges. It then embeds company-specific information into the template to generate a customized proposal.

[1996] Specific example: The AI ​​selects an appropriate template from a template database and uses the Jinja2 template engine to embed company information.

[1997] 4. Generating and sending proposal emails:

[1998] Based on the proposal document, the AI ​​generates a proposal email. The generated email includes a summary of the proposal and the reasons for the proposal.

[1999] The server uses the SMTP protocol to send proposal emails and proposal documents in bulk. It records transmission logs and manages the status.

[2000] Specific example: AI uses an email template to generate an email with a proposal PDF attached. The server sends the email via SMTP and records the sending log.

[2001] User terminal

[2002] The user terminal is a device used to review proposals and email content, and to perform follow-up actions. The Domo dashboard allows users to check email open rates and proposal viewing rates, and to send follow-up emails or schedule meetings as needed.

[2003] Example prompt statements

[2004] Data collection:

[2005] Write Python code to scan the 'Proposals' folder in the database and collect all files. Then, use pandas to identify and remove columns containing sensitive information.

[2006] Collection and analysis of corporate information:

[2007] "Write Python code that retrieves information from company A's API and performs clustering analysis using scikit-learn."

[2008] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. In addition, tracking of sending status and follow-up functions improve the overall effectiveness and speed of sales activities.

[2009] The flow of the specific processing in Example 1 will be explained using Figure 11.

[2010] Program processing flow

[2011] Step 1: Data Collection and Preprocessing

[2012] Input: Database containing past proposals and related documents.

[2013] Output: Pre-processed data with sensitive information removed, ready to be provided to the AI.

[2014] Specific actions:

[2015] 1. The server connects to the database and uses SQL queries to collect proposal data.

[2016] 2. After obtaining the proposal data, the data is filtered using the Python pandas library to remove confidential information.

[2017] 3. Save the filtered data to provide it to the AI ​​model.

[2018] Specific data processing:

[2019] The server extracts proposal data from the database using SQL queries. This data is then read using pandas, and specific columns or values ​​are filtered to remove sensitive information. Finally, the pre-processed data is saved as a file.

[2020] Specific example:

[2021] The server downloads all files in the "Proposal" folder at once. The pandas drop function is used to remove columns containing confidential information.

[2022] Step 2: Gathering and analyzing information on untapped companies

[2023] Input: Company information obtained via API

[2024] Output: Analyzed company characteristics data

[2025] Specific actions:

[2026] 1. The server sends a request to the integrated data platform's API to retrieve data on untapped companies.

[2027] 2. Perform clustering analysis on the acquired data using the scikit-learn library.

[2028] 3. Save the analysis results to the database and prepare them for provision to the AI.

[2029] Specific data processing:

[2030] The server uses an API to retrieve company data (e.g., industry, number of employees, sales). This data is then analyzed using scikit-learn's clustering algorithm to group company characteristics. The analysis results are saved for use with the AI.

[2031] Specific example:

[2032] The server retrieves data from company A via an API request, performs clustering analysis using the scikit-learn KMeans algorithm, and identifies company A's industry.

[2033] Step 3: Proposal Generation

[2034] Input: Company characteristics data and past proposal templates

[2035] Output: Customized proposal

[2036] Specific actions:

[2037] 1. The server uses AI to anticipate the challenges specific to each company and selects a proposal template based on those challenges.

[2038] 2. Create customized proposals by embedding company-specific information into proposal templates.

[2039] Specific data processing:

[2040] The AI ​​selects an appropriate proposal template based on company characteristics data and uses the Jinja2 template engine to embed company information.

[2041] Specific example:

[2042] The AI ​​selects an "IT security" template from the template database and fills in specific information about company A (number of employees, sales revenue).

[2043] Step 4: Generating the proposal email

[2044] Input: Generated proposal

[2045] Output: Proposal email

[2046] Specific actions:

[2047] 1. The AI ​​analyzes the generated proposal and creates a proposal email.

[2048] 2. Attach the proposal PDF file to the email you created.

[2049] Specific data processing:

[2050] The AI ​​extracts necessary information from the generated proposal and inserts it into an email template. The proposal PDF is then attached to the email to complete the process.

[2051] Specific example:

[2052] The AI ​​generates an email stating, "We propose security enhancements for company A," and attaches a PDF of the proposal.

[2053] Step 5: Send out the proposal and email to everyone.

[2054] Input: Proposal email and proposal PDF

[2055] Output: Recording of transmission logs and status information

[2056] Specific actions:

[2057] 1. The server uses the SMTP protocol to send proposal emails and proposal PDFs simultaneously.

[2058] 2. Log the status and manage it when the transmission is complete.

[2059] Specific data processing:

[2060] Send emails via SMTP server. The sending status and details of the email are recorded in the sending log.

[2061] Specific example:

[2062] The server uses SMTP to send the proposal and proposal email to company A and records it in the transmission log.

[2063] Step 6: Follow-up

[2064] Input: Email open status and proposal viewing status

[2065] Output: Follow-up actions (e.g., sending an email, scheduling a meeting)

[2066] Specific actions:

[2067] 1. Users can check email open rates and proposal view rates through the Domo dashboard.

[2068] 2. Send follow-up emails or schedule meetings as needed.

[2069] Specific data processing:

[2070] Check the open and view rates on the Domo dashboard and take the next action (send a follow-up email, schedule a meeting).

[2071] Specific example:

[2072] The user confirms on the Domo dashboard that "the email sent to company A has been opened" and sets up a follow-up meeting.

[2073] (Application Example 1)

[2074] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2075] Traditional customer proposal creation systems required manual proposal creation and customer information analysis, which was time-consuming and labor-intensive. Furthermore, the process of generating personalized proposals was cumbersome, hindering the efficiency of new customer acquisition. Additionally, there was a lack of management functions to track the effectiveness of generated proposals and provide appropriate follow-up.

[2076] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[2077] This invention includes a server that provides data to an AI after collecting information from a database and removing confidential information; a server that provides data to an AI after obtaining untapped company information via an API, analyzing the characteristics of the companies, and providing this information to the AI; a server that allows the AI ​​to anticipate potential challenges for each company and select and generate proposal templates based on those challenges; a server that allows the AI ​​to generate proposal emails based on the generated proposals and sends them together with the proposals; a server that tracks the open and viewed status of sent emails to enable follow-up; a server that collects customer information from a sales site, analyzes it based on past purchase history and visit data, and generates customized product proposals as part of the proposals; and a server that generates personalized proposal emails based on the customized product proposals, sends them all at once via a mail server, and provides in-app notifications. This makes it possible to automatically and efficiently generate personalized proposals and proposal emails for new customers and send them all at once. In addition, it is possible to track the open and viewed status of emails and perform effective follow-up.

[2078] A "database" is a data storage system that centrally stores information such as past documents and proposals, and allows data to be retrieved as needed.

[2079] "Confidential information" refers to non-public information about a company or its customers that should not be leaked to the outside.

[2080] "AI" is an abbreviation for artificial intelligence, which refers to technologies that automate or optimize specific tasks through data analysis and machine learning.

[2081] "API" stands for Application Programming Interface, which is an interface for exchanging data and functions between different software programs.

[2082] "Untapped company information" refers to information about companies with which no business transactions have yet been concluded, including industry, number of employees, and sales figures.

[2083] A "template" is a basic model used when creating documents such as proposals or emails, and it includes a specific format and content.

[2084] A "proposal" is a document that formalizes a proposal made to a company or customer, and it includes details of the proposal, the reasons for it, and specific benefits.

[2085] A "proposal email" is an email sent to a customer to which a proposal and related information are presented. It is an email that outlines the proposal and its benefits.

[2086] A "sales site" refers to a website or online platform used for selling products.

[2087] "Customer information" refers to all information about a customer, such as purchase history and website visit history.

[2088] "Customized product recommendations" refer to suggestions tailored to the customer's needs, based on their purchase history and visit data, to propose the most suitable products for that customer.

[2089] A "personalized suggestion email" is a suggestion email that contains content that is individually customized for a specific customer.

[2090] "Follow-up" refers to ongoing support and tracking activities conducted after the initial proposal, primarily to check email open rates and the effectiveness of the proposed content.

[2091] This invention relates to a system for automatically generating personalized product proposals and sending them out via email and accompanying materials to customers on a sales website in order to acquire new customers. The embodiments for carrying out the invention will be described in detail below.

[2092] System Configuration

[2093] This system consists of a database, server, AI, API, and user terminals.

[2094] database

[2095] The system stores information such as past proposals, purchase history, and visit data.

[2096] server

[2097] Data acquisition and preprocessing

[2098] Customer data acquisition and analysis

[2099] AI-powered proposal generation and email creation.

[2100] Mass email sending and in-app notifications

[2101] Situation tracking

[2102] AI

[2103] Learning for Proposal Writing

[2104] Anticipating challenges for each customer

[2105] Automated proposal and email generation

[2106] API

[2107] Data integration with the integrated data platform

[2108] User terminal

[2109] Review of proposal and email content

[2110] Execute follow-up actions

[2111] Program processing

[2112] The server first collects past purchase history and visit data from the database and performs preprocessing. Specifically, it uses the Python pandas library to format the data and filters out sensitive information. Next, it retrieves untapped customer information via the integrated data platform's API. The collected customer information is classified using clustering analysis with the scikit-learn library.

[2113] The AI ​​selects a suitable product suggestion template for each categorized customer and generates customized product suggestions based on their purchase history and interests. During this process, a generation AI model is used to generate prompts and create specific suggestion content. The generated suggestion emails are sent to customers simultaneously via a mail server, along with in-app notifications.

[2114] Hardware and software

[2115] Hardware:

[2116] server

[2117] User terminal

[2118] software:

[2119] Python

[2120] pandas

[2121] scikit-learn

[2122] smtplib

[2123] request

[2124] Specific example

[2125] For example, when using this system to provide personalized product recommendations to a new customer G1, the server retrieves information on the untapped customer G1 via the integrated data platform's API and classifies the customer's characteristics using clustering analysis. The AI ​​then generates a "health and wellness-related" product recommendation document based on G1's past purchase history and visit data, and creates a personalized recommendation email for G1. The generated email is sent via a mail server, and in-app notifications are also sent.

[2126] Example of a prompt:

[2127] "Please generate an email suggesting health and wellness products based on the products the customer has purchased and the pages they have viewed. The email should include explanations of why the suggested products are suitable for the customer's needs."

[2128] This makes it possible to automatically and efficiently generate and send personalized product proposals and promotional emails to new customers on the sales site. Furthermore, it allows for tracking email open rates and viewing activity, enabling effective follow-up.

[2129] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[2130] Step 1:

[2131] The server collects past purchase history and visit data from the database. Specifically, it reads and formats this data using the Python pandas library. Since the collected data may contain sensitive information, it is filtered to remove any confidential information.

[2132] Input: Purchase history and visit data from the database

[2133] Output: Formatted and sensitive data removed dataset

[2134] Step 2:

[2135] The server retrieves untapped customer information via the integrated data platform's API. Specifically, it sends API requests to obtain information such as the customer's industry, number of employees, and sales revenue. The retrieved data is then read using the pandas library and converted into the appropriate format.

[2136] Input: API request for the integrated data platform

[2137] Output: Dataset of characteristics information for untapped customers

[2138] Step 3:

[2139] The server performs clustering analysis on the acquired customer information using the scikit-learn library. This clustering analysis classifies the customers into clusters based on their characteristics.

[2140] Input: Dataset of characteristics information for untapped customers

[2141] Output: Customer data with cluster labels assigned.

[2142] Step 4:

[2143] The AI ​​selects the optimal proposal template for each customer based on cluster labels. After selecting a template, it automatically generates a customized proposal by embedding customer characteristic information.

[2144] Input: Customer data with cluster labels assigned.

[2145] Output: Customized proposal

[2146] Step 5:

[2147] The AI ​​creates a personalized proposal email based on the generated proposal. The proposal email includes a summary of the proposal and the reasons for the proposal.

[2148] Input: Customized proposal

[2149] Output: Personalized suggestion email

[2150] Step 6:

[2151] The server sends proposals and proposal emails via the mail server based on a mass mailing list. It also simultaneously sends in-app notifications. It records sending logs and manages the status.

[2152] Input: Personalized suggestion email

[2153] Output: Sending log, in-app notification status

[2154] Step 7:

[2155] Users can check email open rates and proposal viewing rates through the dashboard. They can also send follow-up emails or schedule meetings as needed.

[2156] Input: Sending log, in-app notification status

[2157] Output: Follow-up actions, such as follow-up emails or meeting scheduling.

[2158] Through the above processing steps, this system can automatically and efficiently generate personalized proposals and emails for new customers and send them in bulk. Furthermore, it can track email open and view rates, enabling effective follow-up.

[2159] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[2160] Modes for carrying out the invention

[2161] This invention relates to a system that utilizes AI, an integrated data platform, and an emotion engine to automatically generate proposals and emails for acquiring new customers, and further customizes them based on the user's emotions. The embodiments for carrying out the invention will be described in detail below.

[2162] System Configuration

[2163] This system consists of a database, server, AI, API, emotion engine, and user terminal.

[2164] database

[2165] It stores past proposals, related documents, and user sentiment data.

[2166] server

[2167] Data acquisition and preprocessing

[2168] Customer data acquisition and analysis

[2169] AI-powered proposal generation and email creation.

[2170] Emotional engine integration

[2171] Mass email sending

[2172] Situation tracking

[2173] AI

[2174] Learning for Proposal Writing

[2175] Anticipated challenges for each company

[2176] Automated proposal and email generation

[2177] API

[2178] Data integration with integrated data platforms (e.g., Compass, Domo)

[2179] Emotional Engine

[2180] We analyze user emotions and provide emotion data in real time. Based on this emotion data, we customize proposals and email content.

[2181] User terminal

[2182] Review of proposal and email content

[2183] Execute follow-up actions

[2184] Operating Procedure

[2185] Step 1: Data Collection and Preprocessing

[2186] server

[2187] Past proposals and related documents are collected from the database, confidential information is removed, and then the data is preprocessed for AI training. For example, filtering is performed to reduce unnecessary information.

[2188] The emotion engine is used to analyze user sentiment data regarding past suggestion emails and store it in a database.

[2189] Specific example

[2190] The server scans the "Proposals" folder from the database and retrieves the data using the Python pandas library. It filters out sensitive information and provides the preprocessed data to the AI. Additionally, an emotion engine analyzes user responses in past emails and stores the emotion data in the database.

[2191] Step 2: Gathering and analyzing information on untapped customers

[2192] server

[2193] We will use an API to retrieve information on untapped companies from an integrated data platform. Specifically, we will send an HTTP request and receive company information (in JSON format).

[2194] Based on the acquired data, we analyze the characteristics of the company. For example, we perform clustering analysis using the scikit-learn library.

[2195] The analysis results are provided to the AI. The company characteristics data is used by the AI ​​as reference data for generating proposals.

[2196] Specific example

[2197] The server sends a request to the integrated data platform's API to retrieve information about company A. Then, using scikit-learn, it identifies that company A belongs to the "IT industry".

[2198] Step 3: Select and generate a proposal template

[2199] server

[2200] The AI ​​anticipates potential challenges for each company and selects a proposal template based on those challenges. The optimal template is chosen based on company characteristic data.

[2201] The system automatically generates customized proposals by embedding company-specific information into proposal templates.

[2202] Using an emotion engine, template selection and proposal content are adjusted based on the user's past emotional data.

[2203] Specific example

[2204] The AI ​​selects a template related to "IT security" and fills in specific information about company A, such as "number of employees" and "sales revenue." At the same time, the emotion engine refers to past emotion data and adjusts the content to elicit a more positive response.

[2205] Step 4: Generating the proposal email

[2206] server

[2207] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[2208] Select the necessary email template, embed company-specific content, and customize the email content by reflecting sentiment data analyzed by the sentiment engine.

[2209] Combine the proposal email and proposal document into a single package.

[2210] Specific example

[2211] The AI ​​generates a proposal email containing the message, "We propose strengthening security for company A," and attaches the proposal as a PDF document. The emotion engine adjusts the wording and tone of the email to suit the user's preferences.

[2212] Step 5: Send out the proposal and email to everyone.

[2213] server

[2214] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[2215] Record transmission logs for each recipient and manage their status.

[2216] Specific example

[2217] The server uses the SMTP protocol to send the proposal and proposal email to "Company A," and logs the status upon successful transmission.

[2218] Step 6: Tracking email open and view rates

[2219] server

[2220] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[2221] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[2222] Specific example

[2223] The server embeds tracking pixels in emails to detect open rates, which are then displayed in real time on the Domo dashboard.

[2224] Step 7: Follow-up Action

[2225] User

[2226] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[2227] Based on the data analyzed by the emotion engine, the next follow-up action will be suggested.

[2228] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[2229] Specific example

[2230] The user confirms in the Domo dashboard that the email sent to "Company A" was opened, and based on the sentiment engine's suggestions, sets up a follow-up meeting as the next step.

[2231] conclusion

[2232] This system allows companies to automatically and efficiently create proposals for new customers and send them out in bulk at the appropriate time. Furthermore, AI-powered problem identification and proposal generation ensure that proposals are tailored to customer needs. Tracking and follow-up features improve the overall effectiveness and speed of sales activities. In addition, the integration of an emotion engine enables customization based on user emotions, further enhancing the accuracy and effectiveness of proposals.

[2233] The following describes the processing flow.

[2234] Step 1: Data Collection and Preprocessing

[2235] server

[2236] Collect past proposals and related documents from the database. Specifically, use the Python pandas library to execute appropriate queries and retrieve the data.

[2237] Remove confidential information from the acquired data. For example, use regular expressions based on specific keywords or patterns to filter out confidential information and remove personal and corporate confidential information.

[2238] The pre-processed data is provided to the AI. The data is converted to a standard format such as CSV and saved as a training dataset for the AI.

[2239] In addition, an emotion engine is used to analyze users' emotional data regarding past suggestion emails and store it in a database. The emotion engine uses natural language processing technology to analyze the content of emails and assigns emotional labels such as positive, negative, and neutral.

[2240] Step 2: Gathering and analyzing information on untapped customers

[2241] server

[2242] We will use an API to retrieve information on untapped companies from an integrated data platform. Specifically, we will send an HTTP request and receive company information (in JSON format).

[2243] Based on the acquired data, we analyze the characteristics of companies. For example, we use the scikit-learn library to perform clustering analysis and feature extraction to reveal characteristics such as the company's industry, size, and sales revenue.

[2244] The analysis results are provided to the AI. The company characteristics data is converted into a format that the AI ​​can use as reference data for generating proposals (e.g., a data frame format).

[2245] Step 3: Select and generate a proposal template

[2246] server

[2247] AI anticipates potential challenges for each company. Based on company characteristic data, it compares it with similar past cases and industry trends to identify the most relevant issues.

[2248] Based on the selected issue, a proposal template is chosen. Proposal templates are prepared in advance, and the AI ​​selects the appropriate one.

[2249] The template is used to embed company-specific information and generate customized proposals. Specific data (e.g., company revenue or challenges) is embedded in the template's placeholders.

[2250] Using an emotion engine, template selection and proposal content are adjusted based on the user's past emotional data. For example, expressions and designs that received many positive responses are prioritized.

[2251] Step 4: Generating the proposal email

[2252] server

[2253] Based on the generated proposal, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its submission.

[2254] Select the necessary email template and fill in your company's specific content. The sentiment engine analyzes emotional data to customize the email content for greater effectiveness. For example, it adopts phrasing and tones that elicit positive responses from users.

[2255] The proposal email and proposal document will be packaged together, with the proposal document attached to the email in PDF format.

[2256] Step 5: Send out the proposal and email to everyone.

[2257] server

[2258] The proposal document and proposal email will be sent via the mail server based on a mass mailing list. The SMTP protocol will be used for email transmission.

[2259] Record transmission logs for each recipient and manage their status. Save success / failure results to a log file for later review.

[2260] Step 6: Tracking email open and view rates

[2261] server

[2262] Track email open rates and proposal viewing rates. For example, use tracking pixels embedded in emails to detect open rates.

[2263] The acquired tracking data is reflected in the dashboard in real time. The Domo dashboard allows users to check the status.

[2264] Step 7: Follow-up Action

[2265] User

[2266] Through the Domo dashboard, you can check email open rates and proposal viewing rates. For example, you might receive information such as, "Company A opened the proposal."

[2267] Based on the data analyzed by the emotion engine, the next follow-up action is suggested. For example, if the user showed a positive response to the previous email, a more assertive approach is suggested.

[2268] If necessary, take additional follow-up actions, such as sending follow-up emails or scheduling meetings.

[2269] (Example 2)

[2270] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[2271] This invention relates to a system that significantly reduces the time and effort required for acquiring new customers in a company's sales activities and makes proposals to customers more effective. Conventional systems require a great deal of time and effort to create proposals and generate proposal emails to customers, and they are not sufficiently customized based on customer sentiment. Therefore, there has been a need for a system that supports new customer acquisition in an efficient and effective way.

[2272] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[2273] This invention includes a server that provides data to an AI after collecting information from a database and removing confidential information; a server that provides data to an AI after obtaining untapped company information via an API and analyzing the characteristics of the companies and providing this information to the AI; a server that allows the AI ​​to anticipate potential challenges for each company and select and generate proposal templates based on those challenges; a server that allows the AI ​​to generate proposal emails based on the generated proposals and sends them together with the proposals; a server that tracks the open and viewed status of sent emails and enables follow-up; a server that uses an emotion analysis engine to accumulate past emotional data of users and customize the content of proposals and proposal emails; a server that sends proposal emails and proposals all at once; a server that uses a Domo dashboard to display the sending status in real time; and a server that suggests follow-up actions. This enables the efficient and effective generation and sending of proposals and proposal emails, and makes it easier to realize optimal proposals based on the user's emotions.

[2274] "Document collection" refers to the act of obtaining past proposals and related documents from a database.

[2275] "Removal of confidential information" is the process of filtering and removing information from collected materials that should not be released to the public.

[2276] "Means of providing data to AI" refers to the process of inputting pre-processed data into a machine learning system.

[2277] "Acquiring company information" means collecting data on untapped companies via APIs.

[2278] "Company characteristic analysis" refers to evaluating the characteristics of each company based on acquired company information using methods such as clustering analysis.

[2279] "Identifying challenges" refers to AI predicting potential problems and areas for improvement specific to each company.

[2280] "Selecting a proposal template" is a method of choosing the most suitable proposal format based on the anticipated challenges.

[2281] "Proposal generation" refers to the process of creating a customized proposal by embedding company-specific information into a template.

[2282] "Proposal email generation" refers to the process where AI creates an email based on the generated proposal document, outlining the proposal's summary and reasons.

[2283] "Mass email sending" refers to the process of simultaneously sending a generated proposal document and proposal email to multiple recipients based on a list.

[2284] "Open tracking" refers to tracking and recording whether or not an email that has been sent has been opened.

[2285] "Follow-up action" refers to the act of conducting follow-up surveys or making additional contact with users after sending a proposal email, depending on their situation.

[2286] A "sentiment analysis engine" is a system that analyzes a user's past emotional data and optimizes the suggested content based on that data.

[2287] The "Domo Dashboard" is a tool for visualizing data and displaying real-time tracking information.

[2288] This invention relates to a system that utilizes AI, an integrated data platform, and an emotion analysis engine to automatically generate proposals and emails for acquiring new customers, and further customizes them based on the user's emotions. The detailed configuration and operation of the system are described below.

[2289] System Configuration

[2290] This system consists of a database, server, AI, API, sentiment analysis engine, and user terminal.

[2291] database

[2292] It stores past proposals, related documents, and user sentiment data.

[2293] server

[2294] It performs data collection and preprocessing, customer data acquisition and analysis, AI-powered proposal generation and email creation, sentiment analysis engine integration, mass email sending, and status tracking.

[2295] AI

[2296] This system provides learning opportunities for proposal creation, anticipates challenges specific to each company, and automatically generates proposals and emails.

[2297] API

[2298] Establish data integration with integrated data platforms (e.g., Compass, Domo).

[2299] Emotion analysis engine

[2300] We analyze user emotions and provide emotion data in real time. Based on this emotion data, we customize proposals and email content.

[2301] User terminal

[2302] Review the proposal and email content, and take follow-up actions.

[2303] Operating Procedure

[2304] The following explains how the system works using specific examples for each step.

[2305] Step 1: Data Acquisition and Preprocessing

[2306] server

[2307] The server collects past proposals and related documents from the database and removes confidential information. This process uses the Python pandas library to load, filter, and clean the data. It also uses a sentiment analysis engine to analyze user sentiment data regarding past proposal emails and stores it in the database.

[2308] Specific example

[2309] The server scans the "Proposal" folder from the database, retrieves data using the pandas library, filters it, and provides the preprocessed data to the AI. The sentiment analysis engine analyzes user responses contained in past emails and stores them in the database as sentiment data.

[2310] Step 2: Gathering and analyzing information on untapped customers

[2311] server

[2312] The server uses an API to retrieve information on untapped companies from the integrated data platform. Specifically, it sends an HTTP request and receives company information in JSON format. Then, it uses the scikit-learn library to perform clustering analysis, extract company characteristics, and provide them to the AI.

[2313] Specific example

[2314] The server sends an HTTP request to the integrated data platform's API to retrieve information about company A, and then uses scikit-learn to identify that company A belongs to the "IT industry".

[2315] Step 3: Select and generate a proposal template.

[2316] server

[2317] The AI ​​anticipates potential challenges for each company and selects proposal templates based on those challenges. Based on company characteristic data, it selects the optimal template, embeds company-specific information into the template, and automatically generates a customized proposal. In addition, it uses an emotion analysis engine to adjust template selection and proposal content based on the user's past emotion data.

[2318] Specific example

[2319] The AI ​​selects a template related to "IT security" and fills in information such as company A's "number of employees" and "sales revenue." Simultaneously, an emotion analysis engine refers to past emotion data and adjusts the content to elicit a positive response from the user.

[2320] Step 4: Generate the proposal email

[2321] server

[2322] Based on the generated proposal document, the AI ​​creates a proposal email. The email includes a summary of the proposal and the reasons for its proposal, and its content is customized using a sentiment analysis engine. The proposal email and proposal document are then combined into a single package.

[2323] Specific example

[2324] The AI ​​generates a proposal email containing the message, "We propose strengthening security for company A," and attaches the proposal as a PDF document. An emotion analysis engine adjusts the wording and tone of the email to suit the user's preferences.

[2325] Step 5: Send proposals and emails to all recipients.

[2326] server

[2327] The server uses the SMTP protocol to send proposals and proposal emails in bulk. Specifically, it uses the smtplib library to send emails based on a mass mailing list. It also records a transmission log and manages the transmission status.

[2328] Specific example

[2329] The server uses the SMTP protocol to send the proposal and proposal email to "Company A," and logs the status upon successful transmission.

[2330] Step 6: Track email open and view rates

[2331] server

[2332] The server uses embedded tracking pixels to track email open rates and proposal viewing rates. The collected tracking data is reflected in the Domo dashboard in real time.

[2333] Specific example

[2334] The server embeds tracking pixels in emails to detect open rates, which are then displayed in real time on the Domo dashboard.

[2335] Step 7: Follow-up Action

[2336] User

[2337] Users can check email open rates and proposal view rates through the Domo dashboard. Based on the data analyzed by the sentiment analysis engine, the system suggests the next follow-up actions. Users can then take additional follow-up actions as needed, such as sending follow-up emails or scheduling meetings.

[2338] Specific example

[2339] The user confirms on the Domo dashboard that "Company A" has opened the proposal, and then sets up a follow-up meeting based on the sentiment analysis engine's recommendations.

[2340] Examples of prompt statements

[2341] Select a proposal template and generate a proposal tailored to Company A. Adjust the content based on past sentiment data and output the optimal proposal in PDF format. Also, send the proposal and proposal email based on the mass email mailing list and track the open rates.

[2342] As described above, this system utilizes AI and emotion analysis engines to enable the automated generation and customization of efficient and effective proposals and emails for new customers, and further optimizes sales activities through tracking of sending status and suggesting follow-up actions.

[2343] The flow of the specific processing in Example 2 will be explained using Figure 13.

[2344] Step 1: Data Collection and Preprocessing

[2345] server

[2346] The server collects past proposals and related documents from the database. It executes SQL queries against the database to read the relevant data. Then, it removes confidential information from the collected data. Specifically, it uses the Python pandas library to filter and clean the data. It also uses a sentiment analysis engine to analyze past proposal email data, extract user sentiment data, and store it in the database.

[2347] input

[2348] Proposals, related documents, and past proposal email data collected from the database.

[2349] output

[2350] Co...

Claims

1. A means of collecting data from a database, removing confidential information, and providing it to an AI, A means of obtaining untapped company information via API, analyzing the characteristics of the companies, and providing this information to AI, A method in which AI anticipates potential challenges for each company and selects and generates proposal templates based on those challenges, A method for having AI generate a proposal email based on the generated proposal document and sending it together with the proposal document, A system that includes means to track the open and viewed status of sent emails, enabling follow-up.

2. The system according to claim 1, comprising means for filtering collected data and removing information that should not be released externally during data collection and preprocessing.

3. The system according to claim 1, which includes means for performing clustering analysis in the analysis of corporate information and providing it to an AI based on corporate characteristic data.

Images