Method for evaluating credit offers

Abstract

The present invention describes a computer-implemented method for evaluating credit offers or financial products, comprising steps of identifying, locating, and evaluating potential customers, and improvements related to the assignment of credit or financial products based on credit and financial data associated with the user. The method generates an outcome or recommendation based on an offer for the open market and another for the closed market, in order to provide an improved version. The method uses sources of information disclosed by the user, as well as sources of secondary information, such as credit information reporting agencies (CIRAs), alternative data, business data platforms, such as Syntage, and public data provided by entities such as the central bank (e.g., Banxico in Mexico) and regulatory agencies, such as the National Banking and Securities Commission (CNBV) in Mexico or the Securities and Exchange Commission (SEC) in the United States; and in-house research.

Description

[0001]METHOD FOR EVALUATING CREDIT OFFERS Industrial property rights reserved Part of the description of this application contains material subject to industrial property rights protection. The owner of said rights has no objection to the reproduction by facsimile of the patent document or the description of the application by any person, as it appears in the patent file or records at the Patent and Trademark Office, but otherwise reserves all industrial property rights. INDUSTRIAL FIELD OF APPLICATION The present invention falls within the field of identification, classification, and evaluation of candidates and improvements related to the allocation of financial products based on credit data associated with the user and financial institutions.BACKGROUND There is a wide variety of literature related to credit risk analysis, methodologies that prioritize debt collection and / or debt placement with a debt collector holding debt portfolios in the closed market. STATE OF THE ART BACKGROUND US2007 / 043661 describes “a system for placing debts with a debt collector from multiple debt portfolios. Each debt portfolio includes a plurality of individual debt accounts. The system includes a collections module to maintain collector data for a plurality of debt collectors. The system includes a placement module to match individual debt accounts from the plurality of debt portfolios with selected debt collectors from the plurality of debt collectors. The matching is performed to assign debt accounts to the most suitable collector to collect the debt.”The matching process can be determined based on the historical performance of the collectors and the characteristics of the debt account. Document CN117829980A discloses “a method and system for managing the life cycle of credit and loan orders, and relates to the technical field of credit and loan management, […] establishing a monitoring period to monitor the company's business condition in real time, and predict repayment capacity by calculating the company's expected profit value; at the credit assessment stage, the company's repayment record is counted, the data is transmitted to the database, and the credit grade is calculated according to the number of days of early repayment, overdue repayment, and normal repayment.” Generally speaking, the state of the art is focused on increasing the efficiency of debt collection through perspectives that discard or exclude external information.BRIEF DESCRIPTION OF THE INVENTION Within the technical field of credit or financial product allocation, methods and systems have been described that assess the risks of debt settlement candidates based on credit data associated with those candidates. For example, the application may include parameters such as a debt score generated by rating the user's credit data. This credit data may include, among others, public records, the user's credit score and classification, the credit / debit ratio, available credit, delinquent accounts, negative accounts, instances of negative information on the credit report, average account age, debit, or a combination of these parameters. As noted in the background section, there is a growing need to predict high-risk users in terms of credit in a timely manner.The present invention proposes an improvement to the methodology for evaluating credit or financial product offers for subsequent recommendation by integrating information sources and preferences declared by the user, secondary information sources, alternative data, and public data that are classified or rated using a score with data from both the institution itself (internal) and other institutions (external information). This integration is related to the recommendation of credit or financial products according to the user's profile and preferences, and is not limited solely to risk predictions or collection methods. DESCRIPTION OF THE FIGURES Figure 1. Schematic diagram of the computer-implemented method of the present invention. Figure 2.Graph of average ordinary rate as a function of the total annual cost (CAT) for choosing an offer among groups or clusters of ten different financial products. Figure 3. Diagram of the application of a decision tree technique for product classification using the method of the invention. DETAILED DESCRIPTION OF THE INVENTION The present invention proposes a computer-implemented method for evaluating and assigning credit or financial product offers that produces and displays a result or derivation based on an offer for the open market; another offer for the closed market; said derivation comprises presenting a recommendation or improved version by the method of the invention.The method utilizes various information sources, such as user-declared information, as well as secondary sources like Credit Information Societies (CISs), alternative data, and public data provided by entities such as the Central Bank (e.g., Banxico) and regulatory bodies, such as the National Banking and Securities Commission (CNBV) in Mexico or the Securities and Exchange Commission (SEC) in the United States; as well as compilations from the global computer network, including public and web data. The comprehensive operation of the method of the present invention involves three fundamental layers that describe the complete process; these three layers are described below: a) the layer from the user's perspective, b) the layer of data sources used, and c) the layer of the risk and recommendation model.The following describes each layer before detailing its respective criteria. a) User Layer for the present invention. This layer relates to the interactions and stages the user experiences when using the services of this method. From registration and applying for credit or a financial product to selecting the best offers and receiving a recommendation for an improved offer, this layer provides a detailed view of the user's journey and highlights features not found in other credit or financial product issuers or marketplaces. Creating a user profile. In certain embodiments, the user layer comprises the following stages: - Member or User Registration. This begins when an invited guest decides to register as a member or user; - Application for credit or a financial product.Once registered, the user submits a loan or financial product application, as a loan is not necessarily or exclusively offered; b) Data Source Layer. The data source layer describes the various information sources that the present invention uses to evaluate and generate the best loan or financial product offers for users. Some of these sources, including issuers and credit marketplaces, are in the prior art; the present invention incorporates additional sources, such as data from a Central Bank (such as Banxico) and proprietary research on entities that do not report to the Central Bank and financial authorities such as the SEC in the United States or CONDUSEF, as well as entities that do report to these institutions, providing a competitive technical advantage and a more accurate evaluation. The data sources used are listed below: Data source for individuals.- Contact Information. Basic user contact information, such as name, email address, and phone number. - Personal Data. User's personal information, including full name, date of birth, and gender. - Address. Information about the user's residential address. - Password. Authentication and security data for accessing the platform of this invention. - Credit Report / Rating. Credit bureau information, including the user's credit score and credit history. Institutions such as Credit Bureau or Credit Circle can be consulted to obtain the user's financial details and credit behavior. - Credit Information. Includes details of the credit products the user has used, including amounts, terms, and conditions. - Additional Information. Information provided by the user, such as employment, education, and other relevant data. - Alternative Data.Alternative information such as weeks of contributions to social security, location, and other details. This may include employment and personal references, tax records, and social security records, such as data from the Mexican Social Security Institute (IMSS) in Mexico or the Social Security Administration (SSA) in the United States. - Information from the Central Bank and regulatory bodies. Public data provided by the Central Bank and regulatory bodies, such as the CNBV in Mexico or the SEC in the United States, including statistics and special credit reports. - Additional information about the user's credit history. Information obtained directly from the user about their experience and preferences with credit products. - Information from additional investigations into entities not reported to the Central Bank (such as Banxico).These data are obtained through research on financial entities that do not report to the Central Bank, including, but not limited to, SOFOMES (non-bank financial institutions), FINTECH companies, and others; as well as through our own research on entities that do report to these institutions. This information is updated before the activation of the invention's method or whenever necessary to ensure complete market coverage. This novel aspect allows the method of the present invention to offer a more comprehensive evaluation. It is important to note that, in the prior art, current comparison websites or marketplaces do not incorporate this information, as they only have access to information from the institutions with which they have agreements. - Information from the Central Bank (such as Banxico) and from the present invention.It provides details on data provided by the Central Bank and by the present invention, such as proprietary research on entities that do not report to the Central Bank and financial authorities like the SEC in the United States or CONDUSEF, as well as on entities that do report to these institutions, including characteristics and benefits. - User tenure and products with the present invention. Data on the user's tenure as a platform user and the products they have used over time. Data source for legal entities. For this type of entity, the following information sources are included: Annual Tax Returns and Invoices. These allow for estimating income and evaluating business relationships with other companies by identifying the Federal Taxpayer Registry (RFC) in Mexico or the Taxpayer Identification Number (TIN) in the United States. Business Information Sources.Information from platforms like Sintage and Cobalto, extracted from sources such as the Tax Administration Service (SAT in Mexico) or the Internal Revenue Service (IRS in the United States), the Public Commercial Registry (RPC), and the Single Registry of Guarantees (RUC), can be consulted to validate the company's incorporation and risk profile. Criminal Records and Profiling. Through tools such as Nufi, Google News APIs (Application Programming Interfaces), and credit reports from Círculo de Crédito or Dun & Bradstreet, criminal records and financial information can be accessed. Shareholder Report. To comply with anti-money laundering (AML) regulations, information on shareholders with at least 25% ownership in the company is required. Digitization of Financial Statements.Tools like those from Blue Cognition allow for the digitization of financial statements, facilitating the analysis of indicators such as the Altman Z-score, which measures the probability of a company's bankruptcy. Economic Classification (NAICS): As a non-exhaustive example, the North American Industry Classification System (NAICS) code from the National Institute of Statistics and Geography (INEGI in Mexico) is used to classify a company's economic activity and determine its risk, particularly in manufacturing. c) Risk and Recommendation Model Layer. The risk and recommendation model layer analyzes the collected data to generate personalized credit or financial product recommendations. This model includes product classification and filtering, the application of selection guidelines, and the generation of optimized offers for the user.The methodology evaluates loans or financial products in the open and closed markets and generates a customized offer. I) Ranking of loan or financial products according to user preferences. The method for evaluating loan or financial product offers of the present invention generates a different result or outcome for each user, based on an offer for the open market, an offer for the closed market, and provides or recommends an improved version. Classification of parameters to define a personalization model.The computer-implemented method of the invention integrates user-reported information sources, as well as secondary information sources such as Credit Reporting Agencies (CRAs), alternative data, and public data provided by entities such as the Central Bank (e.g., Banxico in Mexico) and regulatory bodies, such as the National Banking and Securities Commission (CNBV) in Mexico or the Securities and Exchange Commission (SEC) in the United States. Furthermore, the present invention allows for the addition of more financial products that are not part of publicly available information, providing a technical advantage when evaluating and improving a credit offer or financial product, not limited to a single collection method.For the purposes of this application, a financial product refers, but is not limited to, consumer loans, mortgages, auto loans, personal loans, payroll loans, credit cards, lines of credit, business financing, leasing, factoring, trusts, working capital financing, bridge loans, financial leases, syndicated loans, investment products such as investment funds, certificates of deposit, life insurance, property insurance, pension plans, and savings products.Financial products for both individuals and businesses are evaluated based on the speed and efficiency of the application and approval process, considering the following factors: • Simplicity of the process, • Digitization and automation, • Access to mobile platforms, • Eligibility criteria, • Speed ​​of disbursement, • Transparency in costs and conditions, • Higher returns, • Security and protection of capital, • Risk management, • Performance, • Volatility, • Diversification, • Flexibility in terms and payments, • Support and advice, • Fees, • Impact on credit history, and • Opportunities for refinancing or restructuring. Details of customization.The method of the present invention incorporates, as relevant variables, the user's choices, preferences, and needs, offering the following selection options: - Collection of user preferences by selecting from one or more of the following preference criteria: Lowest interest rate: Information on the interest rates of all available credit or financial products is collected and ranked from lowest to highest. Longest repayment term: Credit or financial products are evaluated according to the maximum repayment term they offer and ranked from longest to shortest. Shortest repayment term: Products are ranked according to the minimum repayment term, from shortest to longest. Highest credit line: Products offering the highest credit lines are considered and ranked from highest to lowest amount.Cashback: Products offering cashback are identified and ranked based on the percentage offered. No Annual Fee: Products with no annual fee are listed. Rewards and Points: Credit products are ranked according to the amount and value of rewards and points offered. Interest-Free Installment Promotions: Products offering interest-free installment promotions are ranked. Low Fees and Commissions: Products are evaluated and ranked according to their fees and commissions, from lowest to highest. Additional Benefits (Insurance, Assistance, etc.): Products offering additional benefits, such as insurance, assistance, performance, and low cost, are identified and ranked. Efficient Customer Service: Customer service ratings and reviews are considered, and products are ranked according to reported efficiency.Access to cash advances: Products that allow cash advances are ranked, considering the terms and limits. Fraud protection: Products are evaluated and ranked according to the fraud protection measures they offer. Intuitive and easy-to-use mobile application: Products are ranked based on the usability and functionality of their mobile applications. Fast application and approval process: Products are evaluated based on the speed of the application and approval process. The preference criteria listed for this description are not exhaustive; therefore, other specific preference criteria declared by the user may be included. The method also allows for combinations of preference criteria. - Product Sorting and Ranking: With the collected information, available products are sorted and ranked according to user preferences.This classification is performed using algorithms that weight each preference according to the importance assigned by the user. Comparison of parameters to segment a risk model. - Comparative Analysis: A comparative analysis is carried out between information and data from the Central Bank (such as Banxico), and regulatory bodies, such as the National Banking and Securities Commission, CNBV in Mexico or the SEC (Securities and Exchange Commission) in the United States, the present invention, and others available on the market. While the prior art is limited to comparing products, the method of the present invention performs a detailed product ranking process, highlighting the advantages and disadvantages of each based on user preferences. This process not only compares characteristics but also comprehensively evaluates both the quantitative and qualitative benefits of each credit or financial product.When users are given the option to choose and indicate their preferences regarding various aspects of a product, the method performs an analysis, classification, and comparison of parameters to segment a risk model that ranks products according to a personalized quantitative scheme based on those preferences. For example, if a user selects and / or prefers cashback as their primary benefit, the system can quantitatively compare the different cashback plans offered. Through purchase simulations, as a learning process, it is possible to determine which of these plans offers the best monetary value for a standardized expenditure, thus highlighting the offers that maximize the benefit based on the user's preferences. This approach ensures that the advantages and disadvantages of each option are presented clearly and in accordance with the established priorities.This analysis allows for the identification of products that best align with the user's needs and preferences, providing a technical advantage over other credit or financial product comparison methods. The method of the present invention involves combining the results or derivations it generates. By way of definition, it is stated that: Closed market products are those offered by institutions with which the prospect / user already has a relationship. For example, if it is detected that the prospect / user has accounts at a financial institution (bank), the closed market will include the products that these institutions have available specifically for that user. These products are usually customized or have special conditions due to the existing relationship between the user and the institution.Open market products are those available to the user from institutions with which they have no prior relationship. This includes all credit offers or financial products available in the financial market that are not restricted to current users of a specific institution. These products provide a broader view of credit options, allowing the user to compare and choose from a wider range of offers. Ultimately, the result or derivation of the improved version of the present invention is an optimized proposal based on the best product available to the user, combining the options provided by the open and closed markets, respectively. This offer takes into account the user's preferences and the product they value most, improving the most important aspects according to predefined parameters within their choice preferences and risk profile.Optimization may include improvements in the interest rate, additional benefits, reduced fees, and other factors crucial to the user, such as lower rates and reduced prices. II) Eliminating promotional, expired, or zero-interest products. The method for evaluating and allocating credit offers or financial products of the present invention includes an essential step to ensure the relevance and validity of the offers presented to the user. In this step, after classification and comparison, promotional products are eliminated using separation criteria, such as those that are part of expired promotions or special discounts, products with expired terms, and zero-interest offers that could distort the objective comparison.This process not only ensures that the offers presented to the user are current and applicable, but also guarantees that the featured options are truly competitive in terms of financial terms and benefits offered. By eliminating these products, the method of the present invention prevents the user from receiving outdated or biased information, thus optimizing the accuracy and usefulness of the generated recommendations. Product elimination using separation criteria. This stage is carried out by incorporating the actions indicated below. - Identify promotional products. All products collected in the previous stage are reviewed to identify those that are on temporary promotion. Promotional products are identified based on the start and end dates of the promotion provided by the financial institutions. - Verify expiration.The expiration dates of the products are compared to the current date. Any product whose expiration period has expired is removed from the list of offers. - Remove products with Zero Interest Rates: Products offering a 0% interest rate are identified. These products are removed from the set of offers; only under certain circumstances are they not removed unless they are an integral part of a specific strategy of the invention, such as exclusive internal promotions. - Update the Offer List: After removing products from the offers, including expired or zero-interest products, the product list is updated. This revised list ensures that all offers presented are current, valid, and have competitive rates. - Final Validation: A final validation is performed to ensure that the updated list meets the user preference criteria established in the parameter classification stage to define the personalization model.The relevance and suitability of the products are reviewed again according to user preferences to ensure the best possible experience. This stage is fundamental to maintaining the integrity and validity of the offers presented, ensuring that users, through the method of the present invention, receive updated and relevant credit or financial product proposals optimized for their specific needs and preferences. Product filtering is based on requirements reported by institutions to the Central Bank (such as Banxico) and financial authorities such as the National Banking and Securities Commission (CNBV) and the National Commission for the Protection and Defense of Financial Services Users (CONDUSEF) in Mexico; or the SEC (Securities and Exchange Commission) in the United States.In this stage, credit or financial products are filtered based on the specific requirements reported by financial institutions to the Central Bank, as well as on internal research. These filters ensure that the offers presented to the user are suitable and accessible according to their financial and personal profile. It is important to note that these filters are specific to the application of the present invention, since each issuer typically has its own approval criteria. Basic filters are applied to the products that institutions report for their internal products and, alternatively, for external products. This substage is an inherent support for the risk model segmentation. Parameters to consider: - First Loan or Financial Product. Identify the products available to users seeking their first loan, whether individuals or legal entities.If the user is a first-time loan applicant, filter for products that do not require prior credit history. - Minimum Credit Score. Review the minimum credit score requirements for each product. Remove from the list any products that require a higher credit score than the user. - Monthly Income. Compare the minimum monthly income requirements of the products with the monthly income reported by the user. Remove from the list any products that require a higher monthly income than the user. Verify if the user requires greater liquidity or repayment capacity. - Minimum Employment Length. Evaluate the products based on the minimum employment length requirements. Remove from the list any products that require a longer employment length than the user. - Minimum and Maximum Age. Verify the minimum and maximum age requirements for each product. Remove from the list any products that do not meet the user's age range.- Update the offer list. Update the product list after applying these filters. This revised and updated list ensures that all presented offers are accessible and suitable according to the user's profile. - Final validation. Perform a final validation to ensure that the updated list meets the user's preference criteria and the requirements of the financial institutions. Application filtering. Hard rejections. At this stage, strict filters known as "hard rejections" are applied to remove from the product list those that do not meet critical risk and viability criteria. These filters guarantee that the credit or financial product offers presented to the user are secure and have a high probability of approval.The variables to consider are listed below: - Minimum Income (Hard_decline_g_000): Verify that the user's monthly or annual income meets the minimum income required for the credit products. Any product requiring an income higher than that reported by the user is rejected. - Credit Score Ranges by Clusters (Hard_decline_g_001): Review the specific credit score requirements for different clusters or user groups. Products requiring a credit score outside the range of the cluster to which the user belongs are rejected. - Accounts with Fraud (Hard_decline_g_002): Identify users with accounts associated with fraud. Products that do not accept users with a history of fraud are rejected. - Severe Debt. A value is established, for example, a debt of more than $4,500 (Hard_decline_g_003): Review the user's outstanding debts.Products that do not accept users with significant debt, such as debts exceeding $4,500, are rejected. - Percentage of accounts 90 days or more past due (Hard_decline_g_004). Evaluate the user's payment history, especially accounts 90 days or more past due. Products that do not accept users with a high percentage of significantly past-due accounts are rejected. - Number of inquiries in the last 6 months (Hard_decline_g_005). Verify the number of credit inquiries made in the last 6 months. Products that do not accept users with an excessive number of recent inquiries are rejected. - Risky utilization levels (Hard_decline_g_006). Review the user's credit utilization level. Products that do not accept users with utilization levels considered risky are rejected. - Over-indebtedness levels (Hard_decline_g_007). Analyze the user's overall debt level.Products that do not accept users with high debt levels are rejected. - Current delinquency level (Hard_decline_g_008). Review the user's current payment status. Products that do not accept users with current delinquencies on their accounts are rejected. - Fraud watch codes in the credit bureau (Hard_decline_g_009). Check for watch codes in the credit bureau that indicate potential fraud. Products that do not accept users with these codes are rejected. - Amounts requested and products offered (Hard_decline_g_010). Compare the amounts requested by the user with the amounts offered by the products. Products that cannot offer the requested amount or that are not suitable for the user's needs are rejected. Update and Validation • After applying the hard rejections, update the list of eligible products.• The final list obtained guarantees that all submitted offers are viable and meet the strict risk and viability criteria established by this methodology. Selection Guidelines At this stage, specific selection guidelines are applied to refine the list of credit or financial products, ensuring that the offers are attractive, suitable, and competitive for the user. These guidelines highlight the ability of the invention's methodology to classify products according to their characteristics, a feature not found in credit comparison tools or issuers.This stage integrates the risk and recommendation model, and upon completion, generates the result or derivation by displaying: best closed market offers, best open market offers, and an enhanced recommendation using the method of the present invention, weighting the option considered most attractive to the user's needs according to their profile. The selection guidelines are indicated below. - Classify by product type (Step_000). Classify each product based on the current maximum credit line, the time on credit (MOBS) of the oldest account, and a segmentation based on a decision tree. Product types include: Secured, Entry level, Classic, Gold, Platinum, and Elite. See Tables 1 and 2, and Figure 3. Based on this classification, offer products personalized to the user's credit profile and experience. - Attractive products based on credit history (Step_001).Select products that are attractive in terms of amounts and credit lines, based on the user's credit bureau experience. This ensures that offers are competitive and appropriate for the user's credit history. - Eliminate duplicate products (Step_002): Remove multiple products from the same financial institution, keeping only the best from each. Criteria for selecting the best products include the average interest rate, the financial institution, and the name of the product. This avoids redundancies and simplifies the choice for the user. - Divide into open and closed market categories (Step_003): Divide the products into open and closed market categories. If it is identified that the user already has products from certain financial institutions, classify those products as closed market.Products from institutions with which the user has no prior relationship are classified as the open market. - Limit products by Institution (Step_004): Limit the view to 1 to 3 products from different institutions for the open market and 1 to 3 products for the closed market. These results are derived from training the selection guidelines using clustering techniques and decision trees that segment product and user groups. This facilitates decision-making for the user by presenting a manageable set of options; it also avoids information overload and facilitates more effective comparison. This approach allows the user to focus on the most relevant and comparable offers, promoting informed and efficient decision-making. Recommendation presentation. - Closed Market recommendation.This recommendation is based on products from institutions with which the user already has a relationship. The recommendation is optimized to offer the best possible conditions within these institutions. - Open Market Recommendation. This recommendation is based on products available from institutions with which the user does not have a prior relationship. Optimized to offer a broad view of the most competitive options on the market. Update and Validation. - After applying these selection guidelines, the list of eligible products is updated. - This final list ensures that the offers presented are not only viable and suitable, but also highly competitive and aligned with the user's profile. - Evaluation and Offer Generation.Identifying the best offer in the open market: This involves analyzing all available offers in the open market—those from institutions with which the user has no prior relationship—and selecting the best option. Identifying the best offer in the closed market: Simultaneously, this involves analyzing offers from the closed market—those from institutions with which the user already has a relationship—and selecting the best option. The comparative display of the best open and closed market offers provides advantages over other issuers and marketplaces for credit or financial products. - Selecting the most attractive option: The method displays the best open and closed market offers and weights them according to the user's profile, selecting the option that best suits their needs and risk profile. - Improving the offer.The method of the invention comprises the step of taking the result of the “Selection of the Most Attractive Option” described in the preceding section and further improving it. This improvement, within the method, may include better rates, more favorable conditions, or additional benefits, such as better performance or lower cost, tailored to the user type and their risk profile. - Improved Final Offer. The method of the invention presents the user with an improved offer based on their selected preferences, ensuring that the user receives the best possible proposal. In certain embodiments, the method of the present invention allows for combining preference criteria variables to select the offer.Figure 2 shows a graph with two preference criteria variables for credit or financial products. For this example, the average annual percentage rate (APR) is indicated. The column labeled "Group" (cluster) displays groups or clusters of ten different products, from which the best alternative is offered to a user classified according to their risk profile. Figure 3 shows a diagram resulting from applying a decision tree technique to product classification in the invention. This diagram illustrates, as a non-limiting example, that the group of products designated as having the highest risk is identified with credit cards offering an interest rate greater than 70% (rate > 70 for node VI).Therefore, for users with the highest risk profile, the analysis of viable products will focus solely on those with this particular characteristic. Consequently, the best product for that user will be the one that offers the best conditions within this group, aligned with the interests the user defined at the beginning of the credit or financial product application process, including, but not limited to, consumer loans, mortgages, auto loans, personal loans, payroll loans, and credit cards. The parameters that define Figure 3 are presented in Table 1, where, using the average ordinary interest rate (TI) as an example. OPIn the case of credit cards, available credit or financial products are classified according to predetermined access values ​​for such products found in the market. As a non-limiting example, the average ordinary interest rate (TIOP) has been chosen for credit cards, but any preference criterion for classifying parameters to define a risk model can be used for applying decision tree techniques related to product classification, for example, credit limits (CL), as occurs for nodes IV and V in Table 1. Example of TC NODE Average ordinary interest rate (TI) OP Elite IT OP ≤ 28 28 < TI OP ≤ 59.9 Platinum and Gold II and III 28 < TI OP ≤ 39 39 < TI OP ≤ 59.9 59.9 < TI OP≤ 70 Classic and Entry level IV and V LC ≤ $5000 LC > $5000 High risk VI TIOP > 70 Table 1. Under the assumption of the emergence of new credit or financial products whose average interest rate is greater than 70%, through the analysis of information provided by the Central Bank and financial authorities such as the National Banking and Securities Commission (CNBV), the National Commission for the Protection and Defense of Financial Services Users (CONDUSEF) in Mexico; or the SEC (Securities and Exchange Commission) in the United States; as well as our own research, the method of the present invention is updated, and the method of the invention addresses or designates said products as part of node VI. Now, the method of the present invention performs a 0Risk profile segmentation through hierarchy assignment. This allows a well-positioned user to access all market offers, while a user with a less favorable profile has access to more limited credit offers. 5 Risk categories identified by product clusters MOB Range Category Limit Brief description BCS Range (Months on (Credit Score With example of Credit Card and books) minimum months credit bureau NODE TDC. credit) experience required with TDC TDC These are cards intended for people with no credit history or with a negative history, also known as secured cards and their main characteristic is that they require an initial deposit as collateral to reduce risk.Entry-level cards are designed for people who already have a basic credit history. They don't require a deposit as collateral, but generally have low credit limits and few additional benefits. Classic cards are conventional cards for people with a stable credit history. Basic cards offer a moderate credit limit and some additional benefits such as insurance or simple business services. Gold cards offer additional benefits such as more generous rewards and travel insurance; they are intended for people with a solid credit history and credit card experience. Platinum cards offer significant benefits such as VIP access, travel rewards, and high-level insurance; they are intended for experienced users with a robust credit history.Elite cards are the most exclusive on the market; they offer luxury benefits such as a personal concierge, access to exclusive events, and high reward accumulation rates. They are for users with excellent credit histories and a history of credit card use. Table 2. Figure 3 and Table 1 show a grouping of products, where product classification parameters can be observed. Depending on the user's risk profile, as shown in the examples in Table 2, a specific risk segmentation is established. Specifically, Node I, which contains the best cards, corresponds to products classified as elite. Users with this classification can access any product in the spectrum, even those considered high-risk, as they would be approved.A clear example is that this segment includes preferred credit cards. Users with this profile could access any product, so it's simply a matter of ordering the options according to their preferences. For example, if they're looking for the lowest possible interest rate but with the highest credit limit, they could select the card with no credit limit. Conversely, in the case of Node VI, which groups high-risk products, the cards with the highest interest rates are found. As in the previous case, the method establishes additional preferences that influence the selection of the best product. For example, if a user in this segment is looking for the card with the highest interest rate but also with the highest credit limit, the choice would be adjusted to those priorities.The steps or stages of the method of the present invention are carried out by means of a computer system or parts thereof, which can be implemented as computer-readable code to program the processing conditions of said computer system. Various embodiments of the invention are described in terms of this computer system. For example, the computer-implemented method of Figure 1 can be implemented on such a system. The results or derivations illustrated by the diagrams in Figures 2 and 3 can also be implemented through such a system. After reading this description, it will be evident to a person skilled in the art how to implement the invention using other computer systems.At least one input to the computer system must be an application that may have as a parameter a credit score or default risk score generated by evaluating the user's credit data; in other embodiments, such credit data may include, among others, public records data, the user's credit score and classification, credit / debit ratio, available credit, delinquent accounts, negative accounts, instances of negative information on the credit report, average account age, debit, or a combination of such parameters, as well as their financial information. The computer system of the present invention includes one or more processors. The processor(s) may be a special-purpose or general-purpose processor. The processor is connected to a communication infrastructure (e.g., a network).The computer system also includes a user input interface connected to one or more input devices and a display interface connected to one or more screens, which may be integrated input and display components. A computer screen, along with the display interface, can be used as a screen and can display the results shown in Figures 2 and 3, using a web application as an illustrative, but not limiting, example. EXAMPLES Example 1. Using the criteria established in the user strata, data source, and risk model, parameters are classified to define a personalization model and a risk model.Member or user registration occurs when a guest decides to register and submits a loan or financial product application. Once the parameters have been classified, data is entered to define the user type and risk profile for subsequent evaluation and offer generation. It is important to note that, under this classification, access to the offered products depends on the user's profile. Example 2. As explained above, the method allows for combining user preferences. For this example, two preference variables are selected, preferably and illustratively, but not exhaustively: i) the average ordinary rate based on, ii) the total annual cost (CAT); see Figure 2.At this point, the groups or clusters of different products are presented, from which the best alternative is offered in terms of the choice of a user who has been classified according to the segmentation within the risk model, product elimination, and candidate filtering. The method of the present invention prioritizes the personalization of the offer since, within a group (cluster) of products, the user has the option to choose the offer they prefer from the provided recommendation; continuing with the example, within the group of products numbered 6 in Figure 2, the user can choose the one that offers the lowest rate. See the box in Figure 2. Example 3.Based on the estimates in Examples 1 and 2, this method offers an automated stage for improving the offer. If the user prefers a product with a 15% interest rate, this method is authorized to propose an improvement, namely, reducing the rate to 13%, and / or adding additional benefits such as cashback or reducing certain fees, to ensure, through personalization, a better offer or recommendation for the user. Part of the description of this application contains material subject to industrial property rights protection. The owner of these rights has no objection to the reproduction by facsimile of the patent document or the description of the application by any person, as it appears in the patent file or records at the Patent and Trademark Office, but otherwise reserves all industrial property rights.

Claims

CLAIMS 1. A computer-implemented method for evaluating and allocating credit or financial product offers, the method comprising the steps of: creating a user profile; registering data; classifying parameters to define a personalization model; comparing parameters to segment a risk model; eliminating products from offers based on separation criteria; filtering applications based on rejection criteria; and presenting, via an interface, a recommendation based on comparative estimation of closed market, open market, and enhanced offer.

2. The computer-implemented method for evaluating and allocating credit or financial product offers according to claim 1, characterized in that the step of creating a user profile comprises the sub-steps of registering as a member or user and submitting a financial product application. 3.- The computer-implemented method for evaluating and allocating credit or financial product offers according to claim 1, characterized in that the data registration stage considers information sources for natural persons and legal entities.

4. The computer-implemented method for evaluating and allocating credit or financial product offers according to claim 3, characterized in that the data registration stage with information sources for natural persons comprises the sub-stages of collecting selected information from the data group of. Contact information; personal data; credit report / rating; credit bureau information; credit information; alternative data; information from the Central Bank and administrative bodies; information from additional investigations into entities not reported to the Central Bank; and information on the user's tenure and the products they have used over time.

5. The computer-implemented method for evaluating and assigning credit or financial product offers in accordance with claim 3, characterized in that the stage of registering data with information sources for legal entities comprises the sub-stages of collecting selected information from the group comprising annual returns and invoicing; business information sources; criminal records and profiling; shareholder reports; digitization of financial statements; and classification of economic activity. 6.- The computer-implemented method for evaluating and allocating credit or financial product offers according to claim 1, characterized in that the stage of classifying parameters to define a personalization model comprises the sub-stages of collecting user preferences, where such preferences are criteria selected from the group comprising lower interest rate; longer repayment term; shorter repayment term; higher credit line; cashback; no annual fee; rewards and points for purchases; interest-free installment promotions; low fees and commissions; additional benefits such as insurance and assistance; efficient customer service; access to cash advances; fraud protection; higher yield; mobile application; application handling and fast approval.

7. The computer-implemented method for evaluating and allocating credit or financial product offers according to claim 1. with claim 1, characterized in that the stage of comparing parameters to segment the risk model is carried out using information and data from the Central Bank, administrative bodies, proprietary research, and other sources available in the market.

8. The computer-implemented method for evaluating and allocating credit or financial product offers according to claim 1, characterized in that the stage of eliminating products from offers using separation criteria comprises the sub-stages of identifying products from offers; verifying maturity; eliminating products with a Zero Interest Rate; updating the list of offers; and validating that the updated list meets the user's preference criteria and risk model. 9.- The computer-implemented method for evaluating and allocating credit or financial product offers according to claim 1, characterized in that the application filtering stage using rejection criteria comprises the sub-stage of considering selected factors from the group comprising minimum income; user's monthly income; score ranges by cluster; accounts with fraud; serious debts; number of inquiries in the last six months; risky utilization levels; over-indebtedness levels; current delinquency level; fraud observation codes in the bureau; amounts requested and products offered. 10.- The computer-implemented method for evaluating and allocating credit or financial product offers in accordance with claim 1, characterized in that the stage of presenting a recommendation under comparative estimation comprises the sub-stage of refining the list of credit or financial product offerings by means of selection guidelines.

11. The computer-implemented method for evaluating and allocating credit or financial product offers according to claim 10, characterized in that the sub-step of refining the list of credit or financial products comprises, in turn, the sequential sub-steps of: classifying each product based on the current maximum credit line, the time on credit bureau (MOBS) of the oldest account, and a segmentation based on a decision tree; offering products aligned with the user's credit, economic, risk, sociodemographic profile and preference; eliminating duplicate products; dividing into open and closed market categories; and limiting products by institution; wherein the sequential sub-step of dividing into categories comprises identifying that the user already has products from certain financial institutions and defining those products as closed market;Identify if the products are from institutions with which the user has no prior relationship, define those products as open market; limit the view to 1 to 3 products from different institutions for the open market and 1 to 3 products for the closed market; update the list of eligible products; and improve the offer.

12. A computer system configured to perform the evaluation and allocation of credit or financial product offers according to the method of claim 1, the system comprising: one or more processors connected to a communication infrastructure; a user input interface connected to one or more input devices; and a display interface connected to; one or more screens that can be integrated input and display components; where the system is configured to perform the stages of: creating a user profile; registering data; classifying parameters to define a personalization model; comparing parameters to segment a risk model; removing products from offers using separation criteria; filtering applications using rejection criteria; and presenting, through an interface, a recommendation under comparative estimation of closed market, open market, and improved offer.

Images