AI-powered system for supporting business strategy and investment plan formulation.

The AI system addresses the fragmentation of business strategy formulation by integrating multimodal input, data analysis, and interactive proposal generation, offering comprehensive and customizable strategic support.

JP2026175252APending Publication Date: 2026-11-02ONPLANETZ CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2025080807
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2025-04-21
Publication Date
2026-11-02

AI Technical Summary

Technical Problem

Current AI systems lack a comprehensive solution that integrates the functions of receiving and organizing business managers' ideas and concerns in a dialogue format, conducting cross-sectional research of vast amounts of internal and external data, and automatically generating strategic options and presenting materials for comparison and consideration, leading to fragmented processes and reliance on subjective human expertise.

Method used

An AI system that accepts consultation content through multimodal inputs, analyzes and structures issues, collects and analyzes relevant data, and generates multiple strategic proposals, supporting interactive refinement through dialogue, utilizing a knowledge base and learning function to provide customized advice.

Benefits of technology

The system provides high-quality, time-efficient, and accurate strategic planning and decision-making by integrating data collection, analysis, and proposal generation, reducing reliance on human effort and bias, and enhancing flexibility and comprehensiveness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2026175252000001_ABST
    Figure 2026175252000001_ABST
Patent Text Reader

Abstract

We provide a management support system that makes it easy to objectively compare and evaluate proposed solutions and generates useful information that contributes to management decisions. [Solution] The management support method using the management strategy AI system involves receiving consultation content provided by the manager (user) (interface unit), analyzing the consultation content, structuring the issues (language understanding unit), and obtaining insights related to the issues (information gathering unit). Using these analysis results and structured issue information as input, it automatically generates multiple management strategy proposals (strategy generation unit) and can provide feedback (dialogue presentation unit). It is possible to organize and output multiple strategy proposals derived from the manager's consultation content, along with their rationale and conditions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a system that acquires consultation contents on business ideas and management issues from business operators, collects and analyzes related information by artificial intelligence (AI), and automatically proposes management strategies and investment plans. In particular, the present invention relates to a management consulting support system that hears management consultation contents from a plurality of interfaces such as chat interaction, voice input, Web forms, and meeting records, and conducts market research, strategy formulation, issue structuring, and solution presentation based thereon.

Background Art

[0002] In corporate management, business operators are required to quickly and accurately formulate strategies and make decisions in response to issues they face and new business ideas. However, conventionally, it has been necessary to consult human experts such as management consultants, or for business operators themselves to spend a great deal of time conducting market research and analysis. As a result, the process leading to management decisions required time and effort, and relied heavily on subjectivity and experience.

[0003] In recent years, with advances in generative AI and chatbot technologies, their application to automatic business document creation and decision support has become active. For example, AI for automatically creating business plans has emerged, which can generate business plans including market analysis and financial plans in a short time simply by inputting basic information. Tools represented by PlanitAI are reported to reduce the plan creation that conventionally took 30 hours to about 10 minutes, with AI supporting market research, document creation, and financial planning. In addition, services such as Franca AI have also been developed, which comprehensively streamline the entire business plan creation process for subsidy applications (from business research to writing and project management) using generative AI. Franca features consistent support from automatic research to document creation, and a patent application has been filed for this unique technology. However, these are mainly specialized in creating business plans in a predetermined format, and do not cover problem definition through dialogue with business operators or the presentation of alternative solutions.

[0004] Meanwhile, AI-powered meeting support technologies are also advancing. Services that automatically record and summarize online meetings (such as Sembly AI and Otter.ai) have been put into practical use, and attempts are being made to have AI analyze meeting content in real time to extract important decisions, risks, and issues. For example, a patented technology by Microsoft proposes a system that uses AI to analyze what is said during a meeting, determines its importance based on its relevance to the agenda and the emotions of the participants, and then provides notifications and suggestions. This is helping to quickly capture insights that can be gained from meetings. In addition, AI meeting minute assistants are beginning to have advanced conversational functions that can answer questions and perform analyses via chat across the contents of multiple meetings, and are expected to be applied from extracting insights from meetings to generating the next action. However, these are mainly limited to summarizing meeting content and extracting tasks, and do not cover the formulation of management strategies themselves or the consideration of alternative strategies.

[0005] Furthermore, an example of an AI assistant specializing in a particular field is a virtual advisor in the accounting and tax sector. For instance, a service called SoVa has developed and already obtained a patent for its "AI Tax Accountant" technology, which automatically imports accounting and operational data for each company and provides accurate advice tailored to the company's situation via chat. While personalized consultations utilizing internal data are becoming possible in this way, they are currently limited to back-office operations such as accounting and tax. In sophisticated and comprehensive decision-making areas such as business strategy and investment, AI solutions that can advise users from the vague idea stage and provide comprehensive proposals based on external market data and the latest news have not yet been fully realized.

[0006] Thus, conventional technology lacked a system that could integrate the three functions of (1) receiving and organizing managers' ideas and concerns in a dialogue format, (2) conducting cross-sectional research of vast amounts of internal and external data according to the content of the consultation, and (3) automatically generating strategic options and presenting materials for comparison and consideration. Existing AI tools only provide partial functionality for each of these functions and have not yet reached the point of automating the comprehensive support (from interviews to problem analysis and solution formulation) that management consultants provide. [Prior art documents] [Patent Documents]

[0007] [Patent Document 1] U.S. Patent No. 11,736,310, "Conversational AI for intelligent meeting service" (Filing date: March 19, 2021; Publication date: August 22, 2023) Description: This invention uses AI to analyze online meetings with multiple participants, identify relevant agenda items from the content of the discussions, and notify and propose them to the participants. The technology concerns the analysis of dialogue content during meetings and real-time proposals, and does not cover the formulation of business strategies itself.

[0008] [Patent Document 2] (Reference) Explanation of IBM's patent application for "Scenario Planning Advisor" (filed in 2017): This is a scenario planning support system jointly developed with IBM Research, which is reported to utilize cognitive technology and analysis to identify and manage emerging risks. While it is an example of AI-based risk scenario analysis, it is unclear whether it generates proposals using dialogue input from specific business consultations.

[0009] [Non-Patent Document 1] PR Times News Release "Vividir Implements AI Business Plan Proposal Function and Applies for Patent - Providing New Strategic Support to Business Leaders" (August 20, 2024) Explanation: This SaaS tool uses AI to automatically propose the necessary figures for business plans by analyzing past data and market trends. While it dramatically improves the efficiency of business plan formulation, it does not have a function to formulate the strategy itself by analyzing the content of interviews with users.

[0010] [Non-Patent Document 2] PR Times News Release: "Franca AI Releases Service to Reduce Business Plan Creation Time by More Than 50% with Generating AI" (January 8, 2025) Explanation: This service comprehensively supports the process of creating business plans for subsidy applications, from research to writing, using AI. It is equipped with proprietary, patent-pending generating AI technology, but its intended use is to assist in creating documents in a specific format and does not cover general conversational consulting on management issues.

[0011] [Non-Patent Document 3] PR Times News Release "Virtual Accounting Firm SoVa Acquires Patent for Basic Technology of 'AI x Tax Accountant'!" (December 30, 2023) Explanation: An AI chat service that automatically imports accounting data from user companies and provides personalized advice tailored to the company's situation. It is specialized in its own area (accounting and tax) and does not cover general business strategy consultations.

[0012] [Non-Patent Document 4] Note article "How to Create the Fastest and Highest Quality Business Plan with PlanetAI" (February 20, 2025) Explanation: This tool automatically generates a business plan, including market analysis and financial planning, using AI based on basic business information input. It boasts a reduction of over 90% in work time, allowing for the creation of a plan in a short time. However, user input is limited to a fixed form and does not support highly flexible business consultations. [Overview of the project] [Problems that the invention aims to solve]

[0013] As is clear from the background described above, a system is still not readily available in which business owners can easily consult about ideas and concerns, and in which AI automatically provides multifaceted management support based on that information. The specific challenges are as follows:

[0014] Issue ▲1▼: Burden and bias in information gathering - The sources of information necessary for management decisions are diverse, including market statistics, opinions on social media, news articles, and internal company data. Managers themselves and traditional consultants have to collect and analyze these manually, which is either incomplete or time-consuming. Furthermore, it relies on individual experience, which carries the risk of oversights and biases.

[0015] Challenge ▲2▼: The Personality of Strategic Planning - Formulating business strategies and investment plans requires advanced expertise and experience, and tends to be highly dependent on individuals. While AI text generation technology has made it possible to create templates for business plans, current automation tools lack the flexibility to present multiple strategic options devised within appropriate frameworks for open issues.

[0016] Challenge ▲3▼: Fragmentation of the Problem-Solving Process - Currently available AI tools have fragmented functions, such as chat consultation, data analysis, and text generation. For example, there are separate AI tools for summarizing meeting minutes and AI tools for generating business plans, but there is no integrated solution that seamlessly connects the entire process of **"understanding the consultation content" → "investigating relevant data" → "formulating a strategy" → "presenting a proposal"**. As a result, managers have to go through individual tools and experts, which prolongs the decision-making process.

[0017] The present invention aims to solve these problems and provide a comprehensive management strategy support platform that allows managers to instantly obtain useful feedback from AI on ideas they have conceived and challenges they face. [Means for solving the problem]

[0018] According to the present invention, to solve the above problems, there is provided an AI system that analyzes the content of hearings obtained through dialogue with business managers, automatically collects and analyzes related information, and generates strategic proposals. Specifically, it includes the following components.

[0019] Multimodal input means: Accepts business consultation content input in various forms, such as chat UIs, voice input (including voice recognition), web forms, and even recordings / minutes of offline meetings. From voice and meeting records, text conversion and summarization are performed via automatic speech recognition (ASR) and language analysis to extract the user's intent and issues.

[0020] Issue understanding and structuring unit: Analyzes the input consultation content (e.g., questions such as "Should we enter a new market?" and "What is the cause of stagnant sales?") through natural language processing, and structures business issues. For example, it internally generates an issue tree that breaks down problems into elements (decomposition based on business strategy frameworks), and organizes sub-issues to be considered (market environment, competitive landscape, internal company resources, risk factors, etc.). This allows the AI to clarify the points at issue even if the user's question is vague.

[0021] Information collection and analysis unit: A module that automatically searches for and collects relevant internal data (internal company sales performance data, financial data, customer data, etc.) and external data (public statistical data, SNS posting data, news articles, academic papers, industry reports, etc.) in accordance with the above issue decomposition. Internal data is securely imported into the system as needed (via API linkage or database connection), and external data is acquired through web crawling, APIs, queries to third-party databases, and the like. After acquisition, statistical analysis, sentiment analysis, and trend analysis are performed to extract important facts and suggestions related to the issues. At this time, the AI can, for example, grasp brand reputation from customer voices on SNS, and detect industry trends and risk factors from the latest news.

[0022] Strategy and Investment Plan Generation Unit: This is a module that inputs the insights obtained from the problem understanding unit and the information analysis unit to generate multiple management strategy plans and investment plan plans. Here, a model in which knowledge of management strategy frameworks is incorporated into generative AI (large language models, etc.) is used, characterized by generating multiple alternatives. Each plan is configured to be a different strategic option corresponding to the user's consultation content, such as "Scenario A: Deepening the Current Market", "Scenario B: Entering a New Market", "Scenario C: Product Diversification", etc. For each strategy, expected effects and merits, preconditions, risks and points to note, etc. are also output in a format that the AI can explain. Unlike conventional chatbots that simply output a single answer, this supports comparative examination for decision-making.

[0023] Interactive Proposal and Feedback Unit: This is an interface that presents the generated strategy plans and solutions to the user, and responds to questions and answers and additional requests in a chat format. The user can follow up with questions such as "What if we use more conservative estimates?" and "Are there other risk mitigation measures?" regarding the presented plans, and the AI calls the information collection unit or the generation unit again to update the response. As a result, an interactive sparring (brainstorming) process with the user is realized, and the strategy is refined through the collaboration between the AI and the user.

[0024] Knowledge Base and Learning Data: As domain knowledge supporting this system, a database is provided that accumulates management theories, past cases, success and failure patterns, best practices of various industries, and the like. The generative AI refers to this knowledge base to enhance the consistency and persuasiveness of responses. Furthermore, by accumulating dialogues and results for each user company and using them for machine learning feedback within the permitted scope, it becomes possible to provide advice customized for that company as consultations are repeated (learning function). Thereby, for example, continuous strategic advice is realized, in which progress and results of previously examined measures are reflected in new proposals based on the consultation history of a certain company.

[0025] Figure 1 shows an example of the system configuration of the present invention. Figure 1 shows how an interface unit (11) that receives voice and text input from the user (manager), a language understanding unit (12) that analyzes the input content, an information gathering unit (13) that collects relevant data, a strategy generation unit (14) that generates a strategy proposal, and a dialogue presentation unit (15) that presents the generation results and interacts with the user work together. Figure 2 is a flowchart of the method of the present invention, showing the flow from obtaining the user's management consultation in step S201, analyzing and structuring the issue in S202, collecting and analyzing relevant information in S203, generating a strategy proposal in S204, presenting it to the user and receiving feedback in S205, modifying the strategy by looping back as necessary, and finally outputting a proposal report in S206. [Effects of the Invention]

[0026] According to this invention, it becomes possible for managers to obtain high-quality analysis and proposals while significantly reducing time and costs in the strategic planning and decision-making process. The specific effects and differentiating points of each component are as follows.

[0027] Comprehensive one-stop support: Because AI handles everything from initial consultation to proposal, it integrates processes that were previously fragmented. While other companies' solutions offer separate services such as chat support and document generation, this invention covers the entire management consulting process, allowing users to enjoy the convenience of completing everything on a single platform.

[0028] Multimodal information acquisition: It accepts input in all formats, including voice, text, and meeting minutes, allowing users to consult using the most suitable method for their situation. In particular, its ability to automatically extract and analyze the content of offline meetings is a unique feature that builds upon Microsoft's meeting AI patents and commercially available meeting summarization tools, extending even to strategic proposals after meetings.

[0029] Problem structuring and comprehensive issue coverage: The AI ​​function, which breaks down problems and creates analysis plans like a consultant, can uncover potential issues from vague inquiries from users. This is progressive because, unlike PlanetAI and similar systems which relied on user input, the AI ​​can autonomously propose areas for consideration.

[0030] Integrated Analysis of Internal and External Data: Advanced evidence collection and analysis combining proprietary company information with external market data can be performed with a single click. While AI for utilizing internal data, such as SoVa, already exists, this solution provides broader intelligence by cross-searching dynamic data from social media and news, as well as statistical materials, and reflecting that information in strategy.

[0031] Presentation of Multiple Strategic Options: Unlike conventional AI that always provides only one answer, this system simultaneously presents multiple plans with different axes, such as risk tolerance, allowing managers to compare and consider options. This has value similar to human brainstorming and supports users' creative decision-making. This effect prevents tunnel vision caused by reliance on AI suggestions and instead works to broaden perspectives.

[0032] Interactive Improvement and Learning: Even after a proposal is made, the AI ​​can engage in dialogue with the user to refine the details and make revised proposals under new conditions, accompanying them until a satisfactory solution is reached. This interactivity cannot be obtained with simple automated plan creation tools, enabling detailed responses similar to those in actual consulting. Furthermore, the more it is used, the more the accuracy of the answers, which are better suited to the user organization, improves, and value accumulates with continued use.

[0033] As described above, the present invention system possesses an integration and flexibility not found in existing technologies, bringing a qualitative improvement to the business strategy planning and decision-making process. For example, just as Franca AI reduced the man-hours required to create business plans by up to 72%, the present invention is expected to dramatically reduce the research and analysis man-hours required of managers and strategists. At the same time, the accuracy and comprehensiveness of the proposed content may surpass that of human effort. Due to these effects, companies can formulate strategies quickly and accurately even in a rapidly changing business environment, which will greatly contribute to strengthening their competitive advantage. [Brief explanation of the drawing]

[0034] [Figure 1] This block diagram shows the overall configuration of the present invention system. It illustrates the module for inputting user interview content, the AI-powered modules for problem analysis, information gathering, and strategy generation, and their interconnections. [Figure 2] Flowchart of the present invention method. It shows a series of processing steps from receiving user inquiries to problem analysis, information gathering, strategy planning, proposal presentation, and feedback incorporation. [Modes for carrying out the invention]

[0035] [Overall Overview] As shown in Figure 1, the business strategy support system (10) of the present invention includes an input interface unit (11) that receives business consultations from users, a problem understanding and structuring unit (12) that analyzes the input content, an information gathering and analysis unit (13) that collects data related to the problem, a strategy generation unit (14) that generates strategies and solutions, and an interactive proposal unit (15) that presents and interacts with the user. Furthermore, a group of servers on the cloud is provided as a platform for operating the system (10), and it can also be linked with on-premise environments and edge devices (such as conference microphone terminals) as needed. Each component is implemented by a software module or AI model and operates while exchanging data with each other.

[0036] [Input Interface Unit (11)] This unit is responsible for receiving various inputs from the user. In this embodiment, a web application or mobile application equipped with a chatbot screen is provided, and the user can send their inquiry in text. They can also ask questions by voice via a microphone, and the voice is first converted into text by the ASR (Automatic Speech Recognition) engine (11a). In the case of offline meetings, it is also possible to upload meeting records (recorded data or transcribed text). For example, if a user provides the system with the minutes of an internal company meeting, the contents can also be used as input data for subsequent analysis. The input interface unit (11) passes the obtained text data to the problem understanding unit (12) while retaining speaker identification and timestamp information as needed.

[0037] The key feature of this embodiment is its ability to aggregate information without loss at this input stage. In other words, it enables the unified analysis of diverse information formats, such as the content of verbal explanations in conversations with human consultants, notes written on a whiteboard, or checkboxes on questionnaire forms. While conventional AI business planning support tools only anticipated users entering text into predetermined forms, this system accepts input including free-form conversations, allowing managers to naturally communicate ideas without interrupting their thought process.

[0038] [Problem Understanding and Structuring Unit (12)] This unit reads the user's problems and intentions from the input content and organizes the points of discussion for consulting. Using natural language processing (NLP) technology, it extracts keywords from the input text, analyzes dependencies, classifies the intent of the questions, and maps the consultation content to a machine-readable problem model. For example, in response to the user's statement, "I want to know the strategy for success in launching a new product into the market," the unit first understands the structure, such as that the theme is "launching a new product into the market" and that a "strategy for success" is being sought. Next, it uses general-purpose business analysis frameworks (such as SWOT analysis, 4P analysis, and 3C analysis) to list sub-points related to this theme. In the case of launching a new product, possible points of discussion include "Is there a market need?", "What is the competitive landscape?", "What are our company's technological advantages?", and "What is the required investment scale and return forecast?". The Problem Understanding Unit (12) structures these points of discussion in a tree or graph format, generating a data structure that can be referenced by the subsequent Information Gathering Unit (13) and Strategy Generation Unit (14).

[0039] In this embodiment, an analysis plan is formulated within the system based on this problem structure. That is, it includes an algorithm that determines which information sources should be consulted for each issue and which analytical methods (statistical analysis, case-based reasoning, scenario simulation, etc.) should be applied. For example, for "market needs," publicly available statistical data and customer reviews on social media are collected, and for "competitive situation," news articles and industry reports are searched. This provides instructions (search plan) to the information gathering unit (13) described later, enabling efficient and comprehensive data collection.

[0040] One of the effects of the problem understanding and structuring unit (12) is that even if the user's question is vague, the AI ​​can automatically trace the consultant's thought process, enabling a thorough examination with fewer omissions. Conventionally, automated business plan generation only involved writing based on items provided by the user, without identifying new perspectives. In this invention, the AI ​​actively discovers and supplements issues, allowing it to raise and analyze problem areas that the user themselves was not aware of (for example, pointing out potential risk factors). This is a unique intelligent function not seen in the background technologies of chat AI or plan generation AI.

[0041] [Information Gathering and Analysis Unit (13)] This module collects relevant information from internal and external sources in accordance with the issues defined in the problem structure. In this embodiment, the Information Gathering Unit (13) consists of multiple submodules: (a) Internal Data Access Module, (b) External API / Crawler Module, and (c) Data Analysis Module. First, (a) links with databases and business systems within the user company to acquire internal indicators related to management decisions, such as sales history, customer attributes, inventory status, and personnel data. If necessary, privacy and security are taken into consideration, and the data is anonymized or aggregated before being used for analysis. Next, (b) collects data from external information sources on the internet and in the cloud. Examples include querying public statistical databases (government statistics, official survey reports, etc.), scraping from news sites and social media, and acquiring market data from commercial data providers. The AI ​​automatically generates keywords and performs searches according to the problem, so the research proceeds automatically without the user having to use a search engine individually. The data to be collected may include not only text data but also numerical data, images, slides, etc.

[0042] The diverse data collected is then subjected to integrated analysis using the (c) data analysis module. For example, statistical analysis (correlation analysis, regression analysis, time series forecasting, etc.) is applied to numerical data, and text mining (frequently occurring keywords, sentiment analysis, topic modeling, etc.) is applied to text data. It is also possible to analyze public opinion on a new product from a collection of social media posts, or to summarize regulatory trends in the relevant market from the text of industry news. Furthermore, to approach expert analysis, insights are automatically extracted by combining internal and external data. For example, the AI ​​generates textual insights by cross-analyzing multiple sources, such as "Our company's market share is declining, but competitor A has increased its market share by X% in the most recent quarter." These analysis results are temporarily stored in an internal knowledge graph and input into the strategy generation unit (14), and can also be presented to the user as an interim report.

[0043] The Information Gathering and Analysis Department (13) comprehensively collects evidence in a short time, which would require an enormous amount of effort for managers to investigate on their own. For example, existing AI tools such as Franca automate market analysis, but their scope and target were limited. This invention is progressive in that it crosses unstructured data, including SNS, and incorporates highly real-time information (latest news and trends). While Microsoft's meeting AI has a history of analyzing topics and emotions during conversations, this invention takes a different approach by delving into deeper external research after the conversation. Furthermore, although there are examples of incorporating individual user data, such as SoVa, the configuration of linking this to a general-purpose strategy planning AI is novel.

[0044] [Strategy and Investment Proposal Generation Unit (14)] This unit is the core AI of the present invention, and automatically generates management strategy proposals and policy options based on the given problem structure and collected data. In the embodiment, a strategy generation model is used that is fine-tuned in the management strategy domain based on a large-scale language model (LLM). The model has been trained on knowledge of past management strategy cases (both success and failure cases), MBA-style case studies, and various strategic frameworks (SWOT, PEST, value chain analysis, etc.). Then, it takes the current situation analysis results provided by the information collection unit (13) as prompts and generates a strategy proposal that is suitable for the current situation.

[0045] The key point is that it is designed to generate multiple independent strategic options. For example, in this embodiment, three options are presented by default, and the model is adjusted to intentionally output plans with different levels of risk tolerance and resource investment, such as "aggressive expansion," "status quo," and "withdrawal option." Each option automatically generates a detailed explanation that includes items such as (a) Overview: what the strategy is in a nutshell, (b) Benefits: expected benefits and opportunities, (c) Disadvantages and risks: anticipated risk factors, (d) Key measures: a list of specific measures for implementation, (e) Required resources: necessary resources such as funds and personnel, and (f) Predicted KPIs: key indicators that are likely to be achieved if successful. These are formatted according to a template, making them easy for users to compare.

[0046] The strategy generation unit (14) can also perform numerical simulations related to investment plans. For example, statistical models and simulations are separately incorporated and called from the LLM to estimate five-year financial forecasts (sales, profits, return on investment, etc.) for each strategy. It is also possible to incorporate an automated numerical proposal function for business plans, similar to that implemented in Vividir, into this system, so that the revenue and expenditure plan for each strategy can be automatically estimated. In this case, the proposal of a reasonable numerical range by referring to historical data and market benchmarks also applies the methodology of Vividir.

[0047] As described above, the strategy generation unit (14) does not merely summarize text or provide a single solution, but generates a set of options and supporting information necessary for management decision-making. This is where the inventiveness of the present invention lies, setting it apart from conventional general-purpose AIs such as ChatGPT that simply answer user questions one by one. While Franca AI specialized in text generation, the present invention performs comprehensive strategy formulation, including not only text generation but also simulation and calculation. Furthermore, to ensure that the generated results do not contain "illusions" (factual errors) generated by previous AIs, the system is designed to refer to and corroborate evidence data from the information gathering unit (13) at every step. For example, by incorporating specific market statistics and case studies into the explanation of the strategy proposal and presenting their sources to the user, the reliability and accountability of the proposed content are ensured.

[0048] [Interactive Proposal / Feedback Unit (15)] Multiple strategic proposals output from the Strategy Generation Unit (14) are presented to the user either as a single report or sequentially in the form of a dialogue. The user can view the details of each proposal on the chat screen and ask the AI ​​questions if they have any doubts. For example, if the user asks, "What percentage of market share does Plan B aim to capture?", the AI ​​will refer to the contents of Plan B and provide a numerical answer. In response to requests such as, "Is it possible to combine Plan A and Plan C?", the AI ​​can also instantly generate and propose a new hybrid proposal. In this way, the AI ​​acts like an agent, assisting the user in an interactive decision-making process. If necessary, the Strategy Generation Unit (14) can be re-executed based on the dialogue history, and a revised proposal incorporating user feedback can be presented.

[0049] Furthermore, the interactive proposal section (15) also supports multi-channel output. That is, the final proposal results can be viewed as a summary on the chat screen, exported as a formal report document (PDF or slide presentation), or automatically generated as presentation materials for management meetings. For example, if you instruct it to "create slides for the board of directors with this content," slide materials including key charts and action plans will be output. This function is made possible by linking with existing generation AI tools (slide generation AI such as Gamma). A summary document (executive summary) is also automatically created as the final output, and the user can share it with stakeholders and use it for decision-making.

[0050] [Knowledge Base and Learning Function] This system internally contains a domain knowledge base (16) and a user-specific knowledge store (17). The knowledge base (16) stores industry-specific success strategy patterns and management theories, which are referenced when generating strategies. TIFF2026175252000002.tif14170 is possible. On the other hand, the knowledge store (17) is a database that stores the history of interactions with this system and the results of implemented strategies (for example, the adopted plan and actual performance values) for each user organization. These are managed privately and used for the continuous learning of the AI ​​model with permission. For example, if a user company decides to enter a new market based on a proposal from this system in the past, and subsequent performance data is accumulated, the AI ​​will learn from the results and improve the accuracy of future proposals. This feedback can then be used to improve proposals to other companies in the same industry, deepening the system's overall knowledge of business strategy formulation. As a learning virtual business partner, the system of this invention becomes more reliable the more it is used.

[0051] While embodiments of the present invention have been described above, the present invention is not limited to the above embodiments. For example, other AI services (APIs) may be used as data acquisition targets in the information gathering unit (13), and a configuration in which rule-based AI other than neural networks is used in combination with the algorithm of the strategy generation unit (14) to enhance explainability is also conceivable. Furthermore, if the user so desires, a hybrid operation is possible in which human experts (consultants) are involved in the proposal process to review and modify the AI ​​output. These changes and modifications are within the scope of the concept of the present invention and are included in the claims. [Industrial applicability]

[0052] The business strategy formulation support system of the present invention is widely applicable to the management of companies of all sizes and industries. For example, it will be a useful management support tool in a variety of scenarios, such as new business plan development for startup companies, management improvement consultations for small and medium-sized enterprises, and medium- to long-term strategy formulation projects for large corporations. It can also be used as an in-house tool to streamline the analytical work of consultants in management consulting firms. By automating preliminary market research and document creation with this system, consultants can focus on formulating higher value-added recommendations and consultations with clients. Furthermore, it is expected to be applied to general decision support, such as loan application screening support (automatic evaluation of business plans) in financial institutions and screening of investment targets in venture capital firms. Due to the recent boom in generative AI, there is a great deal of attention on AI tools for business support, and the market demand is large. The present invention is a novel and inventive approach among them, and if properly implemented and deployed, it can become a core technology that drives business DX (Digital Transformation).

[0053] The above describes the detailed explanation and usefulness of the present invention. The present invention combines the expertise of human consultants with the latest AI technology to provide a reliable AI advisor for business leaders. This will lead to innovation and sophistication of the business strategy planning process. The features and differentiating points described for each component support patentability from the perspective of novelty and inventiveness, and it is clear that the present invention is an industrially useful invention.

Claims

1. An AI-powered management support method that includes a series of processing steps: receiving inquiries from business owners using multimodal input methods, structuring the issues using natural language processing, collecting and analyzing relevant internal and external data, automatically generating multiple management strategy proposals, and updating and presenting the strategies through repeated proposals and feedback in a dialogue format.

2. An AI-based management support method according to claim 1, characterized in that each generated strategy proposal is presented in a structured output format that includes expected effects, assumptions, risks, required resources, and assumed KPIs.

Citation Information

Patent Citations

  • US11,736,310