Quota-based sales prediction method and system

Abstract

The invention relates to the technical field of sales data analysis, in particular to a quota-based sales prediction method and system. The method comprises the following steps: cleaning and standardizing sales data, and distinguishing old companies from new companies; the sales basic value of the old company is the sales of the same month in the last year, and the growth factor is calculated according to a multi-weight product; the basic value of the new company is the last-month sales of the current year, and the growth factor is based on the last-month link ratio; the adjusted sales value is obtained through the product of the basic value and the growth factor, the apportionment coefficient (the proportion in the total value) of each company is calculated, the single-family predicted sales is obtained through the product of the total target and the apportionment coefficient, and the total predicted value is obtained through summarizing. The method can be applied to various sales scenes, parameter adjustment can be carried out according to specific business requirements, the influence of seasonal factors in growth factor calculation is increased, prediction can adapt to dynamic factors such as market fluctuation through a dynamic weight adjustment mechanism, and the problem that single-month target allocation is disjointed from an actual business scene is effectively solved.

Description

Technical Field

[0001] This application relates to the field of sales data analysis technology, and in particular to a method and system for predicting sales revenue based on quotas. Background Technology

[0002] In corporate sales management, the reasonable forecasting and allocation of sales targets are crucial for formulating sales strategies and allocating resources. With increasingly fierce market competition and the continuous development of data analysis technology, companies are placing higher demands on the accuracy of sales forecasts and the rationality of target allocation. Effective sales target management can not only motivate the sales team and optimize resource allocation, but also provide data support for the company's strategic planning.

[0003] However, existing sales forecasting methods mostly rely on single historical sales data. For newly added companies that lack historical sales data, the average allocation or subjective estimation is generally used to allocate targets, which makes it difficult to balance historical sales scale and growth potential, and cannot accurately reflect the real growth potential of sales, resulting in a disconnect between monthly target allocation and actual business scenarios. Summary of the Invention

[0004] This application provides a sales forecasting method and system based on quotas to solve the above-mentioned problems.

[0005] In a first aspect, this application provides a sales forecasting method based on a fixed quota, the method comprising: S1. Perform data cleaning, missing value imputation, outlier detection, and standardization on the raw sales data; S2. Classify companies as either established or new based on whether they have complete sales data from last year; S3. Calculate the baseline sales value. If the company is an established company, use the sales revenue from the same month last year as the baseline sales value; if the company is a new company, use the sales revenue from the previous month of this year as the baseline sales value; S4. Calculate the sales growth factor. If the company is an established company, calculate the sales growth factor using a multi-weighted product formula; if the company is a new company, calculate the sales growth factor based on the month-on-month comparison from the previous month of this year. S5. Calculate the adjusted sales value and allocation coefficient. The adjusted sales value of a single company is the product of the sales base value and the sales growth factor. The sum of the adjusted sales values ​​of all companies is the sum of the adjusted sales values ​​of the single company. The allocation coefficient of a single company is the adjusted sales value of that company divided by the sum of the adjusted sales values ​​of all companies. S6. Total target allocation: Based on the product of the total sales target and the company's allocation coefficient, the projected sales amount of a single company is obtained. Based on the sum of the projected sales amounts of all single companies, the total projected sales amount is determined.

[0006] Through the aforementioned technical solution, this approach optimizes the original sales data processing flow by introducing a data verification and preprocessing module and a dynamic weight adjustment mechanism to the existing sales forecasting algorithm. This enhances the algorithm's adaptability to complex market environments, achieves intelligent and robust data processing, improves the system's forecasting accuracy and allocation rationality in volatile business environments, and expands the algorithm's application scope in different industries and cyclical businesses. Using the core logic of "base value × growth factor," this solution respects historical sales volume while incentivizing growth potential, avoiding a "one-size-fits-all" allocation and thus improving the rationality of the target.

[0007] Optionally, the growth factor is used to adjust the sales base value, rewarding companies with good sales growth performance by increasing the sales revenue allocation ratio of the corresponding companies, while constraining companies with poor sales growth performance by reducing the sales revenue allocation ratio of the corresponding companies.

[0008] Through the above technical solutions, through growth factors The introduction of growth factors has enabled more accurate and rational sales forecasting. This approach effectively balances a company's historical sales volume and growth potential, avoiding the "one-size-fits-all" allocation found in traditional forecasting methods, thus improving the rationality of target allocation. It rewards companies with strong growth performance by increasing their sales allocation ratio, and vice versa. This not only improves forecast accuracy but also provides an effective incentive mechanism for companies. This mechanism allows sales target allocation to more accurately reflect a company's true growth potential and enhances the dynamic adaptability of sales target allocation to market changes. Therefore, this solution improves forecast accuracy while optimizing resource allocation efficiency, promoting overall sales performance improvement.

[0009] Optionally, if the company's sales revenue for the same month last year is 0, then the company's total sales revenue for the previous year divided by 12 is used as the sales base value.

[0010] By refining the calculation rules for the sales baseline value through the above technical solution, the prediction problem when the company's sales revenue for the same month last year was zero was solved, improving the rationality and accuracy of sales target allocation. Using last year's total sales revenue divided by 12 as a substitute effectively avoids prediction bias caused by abnormal data in a single month, ensuring the continuity and representativeness of the sales baseline value. This improvement allows the prediction model to better adapt to various complex sales data situations, enhances the robustness of the algorithm, and thus comprehensively improves the accuracy of sales revenue prediction.

[0011] Optionally, the multi-weighted product formula is specifically the following formula: ; in, For the aforementioned sales growth factor, Compared to the same period last year, This is the year-on-year weighting coefficient. Compared to the same period last year, This is the year-on-year weighting coefficient for the previous month. Compared to the previous month this year, This is the month-on-month weighting coefficient for the previous month.

[0012] The above technical solution, through a multi-weighted product formula, comprehensively integrates last year's total year-on-year comparison. Year-on-year comparison with last month Compared with the previous month this year These three growth trends across different time dimensions accurately depict the company's true growth potential. This approach overcomes the shortcomings of existing algorithms that only consider growth trends, significantly improving the rationality and accuracy of sales target allocation. This is achieved by dynamically adjusting the weighting coefficients. , , This solution can adapt to different user forecasting needs, making the forecasting results more flexible and intelligent, and effectively solving the problem of the disconnect between monthly target allocation and actual business scenarios.

[0013] Optionally, the year-on-year comparison of last year is used to characterize the long-term growth trend of the company's current sales; the year-on-year comparison of last month of this year is used to characterize the medium-term growth trend of the company's current sales, closely reflecting the sales growth trend within the current period; and the month-on-month comparison of last month of this year is used to characterize the short-term growth trend of the current formula sales, reflecting the recent sales growth momentum.

[0014] The above technical solution clearly defines the total year-on-year comparison for last year. Year-on-year comparison with last month Compared with the previous month this year This approach enables a comprehensive consideration of the company's long-term, medium-term, and short-term sales growth trends. This multi-dimensional analysis method accurately depicts the company's true growth potential, overcoming the shortcomings of existing algorithms that only consider growth trends. Therefore, this solution significantly improves the rationality and accuracy of sales target allocation and enhances its dynamic adaptability to market changes.

[0015] Optionally, if the month-on-month comparison of this year exists, then the growth factor is (1 + month-on-month comparison of this year). If the month-on-month comparison for the previous month of this year is missing, then the growth factor is set to 1.

[0016] The above technical solution refines the calculation rules for the S4 sales growth factor for new companies, resolving the issue of a lack of targeted processing for newly added companies. This is especially relevant when there is a month-on-month comparison with the previous month. At that time, it was directly used as a growth factor. The main basis for this accurately reflects the new company's recent business performance. (Compared to the previous month's figures) When missing, growth factor Setting it to 1 ensures robustness in forecasting and avoids unfounded predictions. This improvement significantly enhances the accuracy of sales forecasts for newly added companies, making the allocation of sales targets more reasonable and precise.

[0017] Optionally, the year-on-year weighting coefficient, the previous month's year-on-year weighting coefficient, and the previous month's month-on-month weighting coefficient are dynamically changed based on user-predicted demand; the user-predicted demand includes long-term demand, medium-term demand, and short-term demand.

[0018] Through the above technical solution, the year-on-year weighting coefficient is used. Year-on-year weighting coefficient of the previous month Month-on-month weighting coefficient By dynamically adjusting based on predicted user demand, this technology solves the problem of inaccurate predictions caused by fixed weights in existing technologies. This dynamic adjustment mechanism enhances the sales growth factor... The calculation can better match users' focus on growth trends across different time dimensions, thereby improving the rationality and accuracy of sales target allocation. It enhances the dynamic adaptability of sales target allocation to seasonal fluctuations and market changes, making the forecast results more intelligent and flexible.

[0019] Optionally, in step S1, the data cleaning includes removing duplicate records and correcting erroneous data formats; the missing value filling uses linear interpolation or industry average values; and the outlier detection is performed by identifying and processing outliers using the Z-score method.

[0020] The above technical solution, which incorporates data cleaning, missing value imputation, and outlier detection in step S1, addresses the issue of low quality in the original sales data. Removing duplicate records and correcting erroneous data formats improves data consistency and accuracy; using linear interpolation or industry averages for missing value imputation ensures data integrity; and identifying and handling outliers using the Z-score method significantly enhances data reliability. This comprehensive data preprocessing provides a solid data foundation for subsequent sales forecasting, thereby improving the rationality and accuracy of sales target allocation and enhancing the robustness of the algorithm.

[0021] Optionally, in step S5, if the total adjusted sales value of all companies is 0, the allocation coefficient for a single company is the proportion of the company's sales revenue in the previous month to the total sales revenue of all companies in the previous month; if the total sales revenue in the previous month is also 0, the allocation coefficient is the average value.

[0022] The above technical solution introduces a multi-level backup mechanism in step S5, resolving the allocation problem when the total adjusted sales value of all companies is 0. When this total is 0, the allocation is performed using the proportion of the previous month's sales revenue, ensuring the continuity of the forecast. If the total sales revenue of the previous month is also 0, the average value is further used for allocation, providing a robust fallback strategy. This improvement significantly enhances the robustness and accuracy of the sales forecasting algorithm, ensuring the rationality of allocation under extreme data conditions and effectively solving the problem of the disconnect between the monthly target allocation and the actual business scenario.

[0023] Secondly, this application provides a sales forecasting system based on fixed quotas, the system comprising: The data preprocessing module performs data cleaning, missing value imputation, outlier detection, and standardization on the raw sales data. The type differentiation module classifies companies into established and new companies based on whether they have complete sales data from last year. The basic analysis module calculates the basic sales value; if the company is an established company, the sales revenue from the same month last year is used as the basic sales value; if the company is a new company, the sales revenue from the previous month of this year is used as the basic sales value. The growth analysis module calculates the sales growth factor; if the company is an established company, the sales growth factor is calculated according to a multi-weighted product formula; if the company is a new company, the growth factor is calculated according to a multi-weighted product formula. Based on the month-on-month data from the previous month, the sales growth factor is determined; the adjustment and aggregation module is used to calculate the adjusted sales value and the allocation coefficient. The adjusted sales value of a single company is the product of the sales base value and the sales growth factor. The sum of the adjusted sales values ​​of all companies is the sum of the adjusted sales values ​​of a single company. The allocation coefficient of a single company is the adjusted sales value of that company divided by the sum of the adjusted sales values ​​of all companies; the sales forecast module is used for total target allocation. Based on the product of the total sales target and the company's allocation coefficient, the projected sales amount of a single company is obtained. Based on the sum of the projected sales amounts of all single companies, the total projected sales amount is determined. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram illustrating an application scenario provided in one embodiment of this application; Figure 2 A flowchart illustrating a sales forecasting method based on a fixed quota, provided as an embodiment of this application; Figure 3 This is a schematic diagram of a sales forecasting system based on a fixed quota, provided as an embodiment of this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0027] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0028] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0029] Existing sales forecasting methods mostly rely on single historical sales data. For newly added companies that lack historical sales data, the common approach is to allocate targets by average allocation or subjective estimation. This makes it difficult to balance historical sales volume with growth potential and fails to accurately reflect the true growth potential of sales, resulting in a disconnect between monthly target allocation and actual business scenarios.

[0030] Based on this, this application provides a sales forecasting system based on quotas. By introducing a data verification and preprocessing module and a dynamic weight adjustment mechanism to the original sales forecasting algorithm, it optimizes the original sales data processing flow, enhances the algorithm's adaptability to complex market environments, realizes intelligent and robust data processing, and improves the system's forecasting accuracy and allocation rationality in volatile business environments. It also expands the algorithm's application scope in different industries and cyclical businesses. Through the core logic of "base value × growth factor," this solution respects historical sales volume while incentivizing growth potential, avoiding a "one-size-fits-all" allocation, thereby improving the rationality of the target.

[0031] Figure 1 This application provides an illustration of an application scenario. In the process of sales analysis, the method provided in this application is applied to enable forecasts to adapt to dynamic factors such as market fluctuations, effectively solving the problem of the disconnect between monthly target allocation and actual business scenarios.

[0032] Specifically, the method provided in this application can be applied to any server. The server interacts with a sales data interface to obtain raw sales data provided by the interface. Based on real-time signal processing and dynamic control theory, combined with historical sales data and market trends, it achieves accurate sales forecasting and target allocation. The raw sales data first undergoes data cleaning, missing value imputation, outlier detection, and standardization to ensure data quality. The system categorizes companies as either established or new based on whether they have complete sales data from the previous year, and calculates the base sales value and sales growth factor accordingly. For established companies, the base sales value is the sales amount from the same month last year; for new companies, it is the sales amount from the previous month of this year. The sales growth factor is calculated for established companies using a multi-weighted product formula, while for new companies, it is determined based on the month-on-month data from the previous month of this year. Subsequently, the adjusted sales value and allocation coefficient are calculated. The adjusted sales value for a single company is the product of the base sales value and the sales growth factor, and the allocation coefficient is the adjusted sales value of that company divided by the sum of the adjusted sales values ​​of all companies. Finally, the overall target is allocated. The projected sales revenue for each company is obtained by multiplying the total sales revenue target by the allocation coefficient for each company. The total projected sales revenue is then aggregated and provided to the user, thus automating and intelligentizing the entire forecasting process.

[0033] For specific implementation details, please refer to the following examples.

[0034] Figure 2 This is a flowchart illustrating a sales forecasting method based on a fixed quota, as provided in one embodiment of this application. The method of this embodiment can be applied to the server in the above scenario. For example... Figure 2 As shown, the method includes: S1. Perform data cleaning, missing value imputation, outlier detection, and standardization on the original sales data.

[0035] S2. Companies are classified as old or new based on whether they have complete sales data from the previous year.

[0036] S3. Calculate the sales baseline. If the company is an established company, the sales revenue of the same month last year shall be used as the sales baseline. If the company is a new company, the sales revenue of the previous month this year shall be used as the sales baseline.

[0037] S4. Calculate the sales growth factor. If the company is an established company, calculate the sales growth factor according to the multi-weighted product formula. If the company is a new company, determine the sales growth factor based on the month-on-month data of the previous month.

[0038] S5. Calculate the adjusted sales value and the allocation factor. The adjusted sales value of a single company is the product of the sales base value and the sales growth factor. The sum of the adjusted sales values ​​of all companies is the sum of the adjusted sales values ​​of a single company. The allocation factor of a single company is the adjusted sales value of that company divided by the sum of the adjusted sales values ​​of all companies.

[0039] S6. Total Target Allocation: The projected sales revenue of each company is obtained by multiplying the total sales revenue target by the company's individual company allocation coefficient. The total projected sales revenue is determined by summing the projected sales revenue of all individual companies.

[0040] This invention aims to address the problem that existing sales forecasting algorithms struggle to balance historical sales volume (base) and growth potential (trend) when allocating sales targets to companies. This will improve the rationality and accuracy of sales target allocation, enhance the accuracy of sales forecasting for newly added companies, comprehensively integrate long-term, medium-term, and short-term growth trends, accurately depict the company's true growth potential, and strengthen the dynamic adaptability of sales target allocation to seasonal fluctuations and market changes.

[0041] This solution, based on real-time signal processing and dynamic control theory, combined with historical sales data and market trends, achieves accurate sales forecasting and target allocation. The raw sales data first undergoes data cleaning, missing value imputation, outlier detection, and standardization (S1) to ensure data quality. The system categorizes companies as either established or new based on whether they have complete sales data from the previous year (S2), and calculates the base sales value (S3) and sales growth factor (S4) accordingly. The base sales value (S3) for established companies uses the sales amount from the same month last year, while for new companies it uses the sales amount from the previous month of this year. The sales growth factor (S4) for established companies is calculated using a multi-weighted product formula, while for new companies it is determined based on the month-on-month data from the previous month of this year. Subsequently, the adjusted sales value and allocation coefficient (S5) are calculated. The adjusted sales value for a single company is the product of the base sales value and the sales growth factor, and the allocation coefficient is the company's adjusted sales value divided by the sum of all companies' adjusted sales values. Finally, the total target allocation (S6) is performed. The predicted sales amount for each company is obtained by multiplying the total sales target by the allocation coefficient for each company, and these are then aggregated to obtain the total predicted sales amount. The modules work together through data flow and control logic to automate and intelligentize the overall forecasting process.

[0042] This solution includes a data preprocessing module, a data type differentiation module, a basic analysis module, a growth analysis module, an adjustment and summary module, and a sales forecasting module. The data preprocessing module performs data cleaning, missing value imputation, outlier detection, and standardization on the raw sales data (S1). The data type differentiation module classifies companies into established and new companies based on whether they have complete sales data from last year (S2). The basic analysis module calculates the basic sales value (S3); for established companies, it uses the sales revenue from the same month last year as the basic value; for new companies, it uses the sales revenue from the previous month of this year as the basic value. The growth analysis module calculates the sales growth factor (S4); for established companies, it calculates the sales growth factor using a multi-weighted product formula; for new companies, it determines the sales growth factor based on the month-on-month data from the previous month of this year. The adjustment and summary module calculates the adjusted sales value and allocation coefficient S5. The adjusted sales value for a single company is the product of the base sales value and the sales growth factor. The sum of the adjusted sales values ​​of all companies is the sum of the adjusted sales values ​​of individual companies. The allocation coefficient for a single company is its adjusted sales value divided by the sum of the adjusted sales values ​​of all companies. The sales forecast module is used for total target allocation S6. Based on the product of the total sales target and the company's individual allocation coefficient, the forecasted sales amount for each company is obtained. The total forecasted sales amount is determined by the sum of the forecasted sales amounts of all individual companies. These modules are interconnected through internal data interfaces and control signals to form a complete sales forecasting system. For example, the output of the data preprocessing module S1 serves as the input to the type differentiation module S2, and the output of the type differentiation module S2 guides the calculation logic of the basic analysis module S3 and the growth analysis module S4. The adjustment and summary module S5 receives the results from S3 and S4 and performs comprehensive calculations. Finally, the sales forecast module S6 completes the final sales forecast based on the output of S5.

[0043] This solution optimizes the original sales data processing flow by introducing a data verification and preprocessing module and a dynamic weight adjustment mechanism to the existing sales forecasting algorithm. This enhances the algorithm's adaptability to complex market environments, achieves intelligent and robust data processing, and improves the system's forecasting accuracy and allocation rationality in volatile business environments. It also expands the algorithm's application scope in different industries and cyclical businesses. Through the core logic of "base value × growth factor," this solution respects historical sales volume while incentivizing growth potential, avoiding a "one-size-fits-all" allocation and thus improving target rationality. For new companies lacking historical data, it uses recent data (sales revenue from the previous month this year) and short-term trends (month-on-month) for forecasting; for established companies, it integrates long-, medium-, and short-term trends, thereby improving the forecasting accuracy for different types of companies. This solution comprehensively integrates long-, medium-, and short-term growth trends to accurately depict the company's true growth potential. Simultaneously, it integrates seasonal data (sales revenue from the same month last year) and multi-dimensional growth indicators, and through a dynamic weight adjustment mechanism, enables forecasts to adapt to dynamic factors such as market fluctuations and industry cycles, effectively solving the problem of the disconnect between single-month target allocation and actual business scenarios. The prediction accuracy is significantly improved compared to traditional methods, and data processing efficiency is also effectively reduced by minimizing manual intervention. The algorithm's logic is transparent and explainable, its parameters are clearly defined, and the calculation process is traceable, making it easy for business personnel to understand and adjust, thus improving its feasibility for implementation. Furthermore, its dynamic weight adjustment mechanism allows it to adapt to the characteristics of different industries and cyclical businesses, demonstrating good applicability and scalability.

[0044] In the above scheme, data preprocessing S1 can employ various methods. For example, data cleaning may include removing duplicate records and correcting erroneous data formats; missing value imputation can use linear interpolation, industry averages, or predictive imputation based on machine learning models; outlier detection can be identified and processed using Z-score methods, IQR (interquartile range) methods, or cluster analysis methods. Company type differentiation S2, in addition to considering whether complete sales data from last year is available, can also consider auxiliary information such as the company's establishment year and registered capital for more refined classification. For calculating the sales baseline value S3, for established companies, in addition to sales in the same month last year, the average sales in the same period last year or the average sales over the past N months can also be considered as a reference; for new companies, in addition to sales in the previous month of this year, the average sales over the previous N months or the average sales of new companies in the industry can also be considered as the baseline value. The calculation of the sales growth factor S4, in addition to the multi-weighted product formula and month-on-month data from the previous month of this year, can also incorporate external factors such as industry growth rate and macroeconomic indicators for comprehensive consideration. In the calculation of adjusted sales value and allocation coefficient S5, if the sum of adjusted sales value of all companies is 0, the allocation coefficient for a single company can be the proportion of that company's sales revenue from the previous month to the total sales revenue of all companies from the previous month; if the total sales revenue from the previous month is also 0, the allocation coefficient can be the average value or a weighted average based on the company's registration time. The total target allocation S6 can also incorporate a manual calibration step after obtaining the projected sales revenue for each company, allowing business personnel to fine-tune it according to actual conditions. Furthermore, this method can be applied to various sales scenarios, such as retail, wholesale, and service industries, and parameters can be adjusted and the model optimized according to specific business needs. For example, in industries with significant seasonal fluctuations, the weight of seasonal factors in the growth factor calculation can be increased.

[0045] In some embodiments, the growth factor is used to adjust the sales base value, rewarding companies with good sales growth performance by increasing the corresponding company's sales revenue allocation ratio, while constraining companies with poor sales growth performance by reducing the corresponding company's sales revenue allocation ratio.

[0046] This solution aims to address the problem of existing sales forecasting algorithms struggling to balance historical sales volume (base) with growth potential (trend) when allocating sales targets to companies, thereby improving the rationality and accuracy of sales target allocation. Traditional forecasting methods often fail to consider both a company's historical size and current growth momentum, leading to over-allocation of targets for large companies and under-allocation for rapidly growing companies, resulting in poor target rationality. This solution solves this problem by introducing a growth factor. As a dynamic adjustment mechanism, the growth factor is based on quantifying a company's growth trend and applying it to a base sales value. The growth factor rewards companies with strong sales growth by increasing their sales allocation ratio, thus incentivizing them to maintain growth momentum; simultaneously, it constrains companies with poor sales growth by reducing their sales allocation ratio, prompting them to improve. This mechanism makes sales forecasting and allocation more aligned with the company's actual operating conditions and market dynamics, avoiding a "one-size-fits-all" approach to target allocation, thereby improving the rationality and accuracy of target allocation. The growth factor dynamically weights the company's sales revenue by adjusting the product of the sales baseline, taking into account both historical data and future growth potential, thus more accurately depicting the company's true sales capabilities and market competitiveness.

[0047] In this plan, growth factors It plays a central role in sales forecasting methods, its function being to dynamically adjust baseline sales values. Specifically, growth factors... The ability to adjust rewards and penalties based on company sales performance: For companies with strong sales growth, the growth factor... A value greater than 1 amplifies the base sales value, thereby increasing the corresponding company's sales revenue allocation ratio; conversely, for companies with poor sales growth, the growth factor... A value less than 1 reduces the base sales value, thereby lowering the corresponding company's sales allocation ratio. This adjustment mechanism ensures fairness and incentive in sales allocation, allowing resources to be directed towards companies with greater growth potential. Growth Factor The calculation method (such as a multi-weighted product formula or month-on-month data from the previous month this year) determines the adjustment range for the sales baseline. For example, when the growth factor... The multi-weighted product formula comprehensively considers various growth trends, including last year's total year-on-year growth, this year's first-month year-on-year growth, and this year's first-month month-on-month growth, making the adjustment more comprehensive and accurate. In this way, the growth factor... This enabled more refined management of the company's sales revenue, making its forecasts more accurate.

[0048] This plan utilizes growth factors. The introduction of growth factors has enabled more accurate and rational sales forecasting. This approach effectively balances a company's historical sales volume and growth potential, avoiding the "one-size-fits-all" allocation found in traditional forecasting methods, thus improving the rationality of target allocation. It rewards companies with strong growth performance by increasing their sales allocation ratio, and vice versa. This not only improves forecast accuracy but also provides an effective incentive mechanism for companies. This mechanism allows sales target allocation to more accurately reflect a company's true growth potential and enhances the dynamic adaptability of sales target allocation to market changes. Therefore, this solution improves forecast accuracy while optimizing resource allocation efficiency, promoting overall sales performance improvement.

[0049] In the above scheme, growth factor The specific calculation method can be adjusted according to different industries and market environments. For example, in rapidly growing industries, more weight can be given to short-term growth indicators (such as month-on-month changes this year) to more sensitively capture market changes. In industries with stable growth, more reliance can be placed on long-term growth indicators (such as year-on-year total growth). Furthermore, growth factors... Further consideration can be given to non-sales data, such as customer satisfaction, market share changes, and product innovation cycles, to more comprehensively assess a company's growth potential. For example, for companies with high customer satisfaction, even if short-term sales growth is not outstanding, their growth factor can be appropriately increased. The weighting of growth factors is used to encourage their long-term development. The adjustment frequency can also be dynamically set according to actual business needs, such as monthly, quarterly, or annually. In cases of extreme market volatility, an emergency adjustment mechanism can be activated to adjust growth factors. Make temporary adjustments to deal with unforeseen circumstances.

[0050] In some embodiments, if the company’s sales revenue for the same month last year is 0, then the company’s total sales revenue for the previous year divided by 12 is used as the sales base value.

[0051] This solution aims to address the shortcomings of existing sales forecasting algorithms in allocating sales targets to companies. For example, if a company's sales in the same month last year were zero, traditional methods may fail to effectively calculate the baseline sales value, leading to reduced forecast accuracy. This solution resolves this technical issue by refining the rules for calculating the baseline sales value. The underlying technology is that when sales in a specific month are zero, the monthly average of the company's total sales from the previous year is introduced as a substitute, ensuring the continuity and reasonableness of the baseline sales value. This approach more accurately reflects the company's historical sales volume, avoiding forecast bias caused by occasional zero sales, thereby improving the reasonableness and accuracy of sales target allocation. Through this improvement, this solution can better adapt to various complex sales data situations, ensuring the robustness of the forecasting model under conditions of incomplete or abnormal data.

[0052] In this scheme, calculating the sales baseline S3 is a crucial step in sales forecasting. Typically, for established companies, the sales baseline S3 is their sales revenue for the same month last year. However, considering the possibility that sales revenue for the same month last year might be zero in actual business operations—for example, due to no business activity, missing data records, or product line adjustments—directly using zero as the baseline would distort subsequent forecasts. Therefore, this scheme introduces a supplementary rule: if a company's corresponding sales revenue for the same month last year is zero, then the company's total sales revenue for last year divided by 12 is used as the sales baseline. The logic behind this approach is that by calculating the monthly average of last year's total sales revenue, a relatively stable and representative historical sales benchmark can be provided even when sales revenue for the same month last year is missing or zero. This ensures that the calculation of the sales baseline S3 is always based on reliable data, avoiding the impact of abnormal data in a single month on the accuracy of the overall forecast. In this way, this scheme can more comprehensively and accurately reflect the company's historical sales scale, providing a reliable benchmark for subsequent adjustments to the sales growth factor S4 and the allocation of the total target S6.

[0053] This solution addresses the forecasting challenge of zero sales revenue in the same month of the previous year by refining the calculation rules for the sales baseline, thus improving the rationality and accuracy of sales target allocation. By using last year's total sales revenue divided by 12 as a substitute, it effectively avoids forecasting bias caused by data anomalies in a single month, ensuring the continuity and representativeness of the sales baseline. This improvement allows the forecasting model to better adapt to various complex sales data situations, enhancing the algorithm's robustness and comprehensively improving the accuracy of sales revenue forecasting.

[0054] In the above scheme, when sales in the same month last year were zero, in addition to using the total sales of last year divided by 12 as the sales baseline, other alternatives can be considered. For example, the average sales of the past N months (excluding months with zero sales) can be used as a reference, or the average monthly sales of similar companies in the industry can be introduced as a benchmark. In certain specific cases, if the company has special events in that month (such as production stoppage or transformation), adjustments can also be made through manual intervention or by combining other business data. In addition, for new companies, if sales in the previous month of this year were also zero, the rules can be further refined, such as using the average monthly sales since its establishment, or referring to the average sales of newly added companies in the industry. Regarding the description of the parameter range, the total sales of last year divided by 12 is a general average, but if the company's sales have obvious seasonality, the average of the months with non-zero sales last year can be used, or a weighted average based on seasonality indices can be used. This flexibility allows this scheme to better adapt to the specific circumstances of different industries and companies, ensuring the accuracy and applicability of the forecast.

[0055] In some embodiments, the multi-weighted product formula is specifically the following formula: ; in, As a sales growth factor, Compared to the same period last year, This is the year-on-year weighting coefficient. Compared to the same period last year, This is the year-on-year weighting coefficient for the previous month. Compared to the previous month this year, This is the month-on-month weighting coefficient for the previous month.

[0056] This solution aims to address the problem that existing sales forecasting algorithms, when allocating targets to established companies, fail to comprehensively consider long-term, medium-term, and short-term growth trends, thus failing to accurately depict the company's true growth potential. Traditional methods often focus only on growth within a single time dimension, leading to biased forecast results. This solution solves this technical problem by introducing a multi-weighted product formula. Its technical basis lies in forming a comprehensive sales growth factor by weighting and multiplying growth indicators from three different time dimensions: year-on-year growth, year-on-year growth of the previous month, and month-on-month growth of the previous month. This approach comprehensively reflects the company's long-term, medium-term, and short-term growth trends, making sales growth factors more accurate. This allows for a more accurate depiction of the company's true growth potential, thereby improving the rationality and accuracy of sales target allocation. This is achieved through dynamically adjusting weighting coefficients. , , This solution can adapt to different forecasting needs and market changes, enhancing the dynamic adaptability of sales target allocation.

[0057] In this scheme, the calculation of the sales growth factor S4 for established companies employs a multi-weighted product formula. Its design aims to comprehensively capture the company's long-term, medium-term, and short-term sales growth trends, assigning different weights based on their importance. This is the final sales growth factor used to adjust the sales baseline value S3. This represents the year-on-year total, used to characterize the company's long-term sales growth trend and reflect the company's overall performance over the past year. Year-on-year weighting coefficient. The growth factor determines the long-term trend. Its influence in China. This represents the year-on-year comparison with the previous month, used to characterize the company's medium-term sales growth trend, and more closely reflects the sales growth trend within the current cycle. (Year-on-year comparison weighting coefficient for the previous month) It determines the influence of medium-term trends. This represents the month-on-month change compared to the previous month, used to characterize the company's short-term sales growth trend and reflect the company's recent sales growth momentum. Month-on-month weighting coefficient. This determines the influence of short-term trends. Through this product-like form, the various growth trends interact and jointly determine the sales growth factor. The size of the factor. For example, the sales growth factor of a company with stable long-term growth, good medium-term performance, and strong recent momentum. This will be significantly greater than 1, resulting in a higher sales allocation ratio. This refined calculation method ensures the sales growth factor... It can more accurately reflect the company's true growth potential and effectively solves the problem of existing algorithms' one-sided consideration of growth trends. (Appendix) Figure 2 The formula's structure and the function of each parameter are explained in detail.

[0058] This scheme comprehensively integrates last year's total year-on-year comparison using a multi-weighted product formula. Year-on-year comparison with last month Compared with the previous month this year These three growth trends across different time dimensions accurately depict the company's true growth potential. This approach overcomes the shortcomings of existing algorithms that only consider growth trends, significantly improving the rationality and accuracy of sales target allocation. This is achieved by dynamically adjusting the weighting coefficients. , , This solution can adapt to different user forecasting needs, making the forecasting results more flexible and intelligent, and effectively solving the problem of the disconnect between monthly target allocation and actual business scenarios.

[0059] In the above scheme, the weight coefficients of the multi-weight product formula , , The weighting can be dynamically adjusted based on anticipated user demand. For example, when users are more focused on long-term growth, the year-on-year weighting coefficient 'a' can be increased; when they are more focused on recent performance, the month-on-month weighting coefficient can be increased. The initial values ​​of these weighting coefficients can be set based on industry experience or historical data, for example... =0.2, =0.5, =0.3. Simultaneously, machine learning algorithms can be introduced to automatically optimize weighting coefficients based on historical prediction errors, enabling adaptive adjustments to improve prediction accuracy. Furthermore, in addition to year-on-year, month-on-month, and year-on-year comparisons, other relevant growth indicators can be considered, such as customer growth rate, product line growth rate, and market share changes, to further enrich the growth factors. The dimensions to consider. In certain special circumstances, such as when a company makes a major strategic adjustment or when the market environment changes drastically, certain growth indicators can be temporarily disabled or their weights reset to avoid inaccurate predictions.

[0060] In some embodiments, the year-on-year comparison of last year is used to characterize the long-term growth trend of the company's current sales; the year-on-year comparison of the previous month is used to characterize the medium-term growth trend of the company's current sales, closely reflecting the sales growth trend within the current period; and the month-on-month comparison of the previous month is used to characterize the short-term growth trend of the current formula sales, reflecting the recent sales growth momentum.

[0061] This solution aims to address the problem that existing sales forecasting algorithms only consider growth trends in a one-sided manner, failing to comprehensively integrate long-term, medium-term, and short-term trends, and thus unable to accurately depict the company's true growth potential. Traditional methods often focus only on single or partial growth indicators, resulting in incomplete and inaccurate forecasts. This solution solves this technical problem by clearly defining the roles of last year's total year-on-year, this year's previous month's year-on-year, and this year's previous month's month-on-month changes in the sales growth factor S4. Its technical basis lies in the hierarchical definition and application of growth indicators across different time dimensions, ensuring that the sales growth factor S4 comprehensively reflects the company's long-term, medium-term, and short-term growth trends. (Last year's total year-on-year...) Focusing on macroeconomic and stable long-term trends, the year-on-year comparison for last month of this year Focus on medium-term trends that are more closely tied to the current cycle, while the month-on-month comparison for last month of this year... This allows for the capture of immediate and sensitive short-term growth momentum. This multi-dimensional and hierarchical consideration of growth trends enables the sales growth factor S4 to more accurately depict the company's true growth potential, thereby improving the rationality and accuracy of sales target allocation.

[0062] In this plan, the sales growth factor The construction of this relies on the precise quantification of growth trends across different time dimensions. Among these, the total year-on-year growth rate last year... It is used to characterize the long-term growth trend of a company's current sales. It compares the company's total sales over the past year with those of the previous year, reflecting the company's stable development and market competitiveness over a longer period. For example, if a company's total sales increased by 20% last year compared to the previous year, then... The figure of 0.2 indicates strong long-term growth potential. (Year-on-year comparison for last month of this year) It is used to characterize the medium-term growth trend of a company's current sales, and it is closer to the sales growth trend within the current cycle. This refers to comparing last month's sales figures with those of the same period last year. It reflects a company's growth over the past year relative to the current month, helping to capture mid-term performance amidst seasonal or cyclical fluctuations. For example, if last month's sales this year increased by 15% compared to the same month last year, then... The figure of 0.15 indicates that the company has maintained a good growth momentum in the medium term. (Month-on-month comparison from last month of this year) It is used to characterize the short-term growth trend of a company's current sales, reflecting the company's recent sales growth momentum. This refers to comparing the sales figures for the previous month of the current year with those for the month before that, allowing for a quick assessment of the latest market changes and company performance. For example, if the sales figures for the previous month of this year increased by 10% compared to the month before that, then... A value of 0.1 indicates that the company has strong growth momentum in the short term. By comprehensively utilizing these three indicators, this plan can fully and multidimensionally assess the company's sales growth potential and ensure the sales growth factor... The accuracy and comprehensiveness of the information.

[0063] This plan defines the total year-on-year comparison for last year. Year-on-year comparison with last month Compared with the previous month this year This approach enables a comprehensive consideration of the company's long-term, medium-term, and short-term sales growth trends. This multi-dimensional analysis method accurately depicts the company's true growth potential, overcoming the shortcomings of existing algorithms that only consider growth trends. Therefore, this solution significantly improves the rationality and accuracy of sales target allocation and enhances its dynamic adaptability to market changes.

[0064] In the above scheme, in addition to the three core indicators, more detailed growth trend indicators can be introduced. For example, an "average month-on-month change over the past three months" can be added to smooth short-term fluctuations, or an "industry average growth rate" can be introduced as an external reference. For industries with obvious seasonal characteristics, a "seasonal index adjustment item" can be added to more accurately reflect the true growth trend. Furthermore, the calculation of these growth trends can be adjusted according to data availability and business needs. For example, if some data is missing, interpolation or industry averages can be used to fill in the gaps. Regarding the description of the parameter range, these year-on-year and month-on-month data are usually expressed as percentages, but need to be converted to decimal form in calculations. Theoretically, the range can go from negative infinity to positive infinity, but in practical applications, there are usually reasonable range limitations, such as a growth rate between -100% (sales of 0) and positive infinity.

[0065] In some embodiments, if the month-on-month comparison of the previous month exists, the growth factor is (1 + month-on-month comparison of the previous month) to the power of c; if the month-on-month comparison of the previous month is missing, the growth factor is set to 1.

[0066] This solution aims to address the problem of low prediction accuracy in existing sales forecasting algorithms when allocating targets to new companies. These algorithms often rely on average allocation or subjective estimation, ignoring recent business performance. This is particularly problematic for new companies, where limited historical data makes it impossible to calculate comprehensive growth trend indicators. This solution resolves this technical issue by refining the calculation rules for the S4 sales growth factor for new companies. Its technical basis lies in the fact that when only recent data (month-on-month comparison) is available for new companies... When available, it should be used as the primary indicator of growth trend; while when the month-on-month comparison of the previous month is available... If the growth factor is also missing, the growth factor G is set to a neutral value of 1 to avoid unfounded predictions. This approach can more accurately reflect the recent business performance and growth momentum of new companies, thereby improving the rationality and accuracy of sales target allocation and enhancing the accuracy of sales forecasts for newly added companies.

[0067] In this plan, the calculation of the sales growth factor S4 has been specifically optimized for new companies to address the challenge of insufficient historical data. For new companies, due to the lack of last year's total year-on-year comparison... Compared with the same period last year Based on long-term and medium-term data, this plan will compare the month-on-month data from the previous month of this year. As its main indicator of growth trend. Specifically, if there is a month-on-month comparison with the previous month of this year... Then the sales growth factor Directly set as (1 + month-on-month comparison of this year) )of The power of. Here This is the month-on-month weighting coefficient used to adjust the month-on-month data in the growth factor. The influence of this approach ensures that the new company's recent growth momentum can be effectively quantified and incorporated into the forecasting model. This method can more accurately capture the new company's short-term sales performance, avoiding the biases caused by simple averaging or subjective estimation in traditional methods. However, in some cases, the new company may not even show a month-on-month increase compared to the previous month. There may also be missing data, such as data from newly established companies, data from the previous month that is unavailable, or data records that are incomplete. To avoid forecast interruptions or negative impacts due to missing data, this plan stipulates that if the month-on-month comparison of the previous month is... If missing, then sales growth factor Set the growth factor to 1. Setting it to 1 means not adjusting for growth or contraction in the sales baseline S3, i.e., maintaining a neutral forecast, which is a robust approach when effective growth data is lacking. Through this differentiated and robust approach, this solution is better suited to the characteristics of new companies and improves the accuracy of its sales forecasts.

[0068] This solution addresses the lack of targeted processing for newly added companies by refining the calculation rules for the S4 sales growth factor. It also addresses the issue of month-on-month growth compared to the previous month. At that time, it was directly used as a growth factor. The main basis for this accurately reflects the new company's recent business performance. (Compared to the previous month's figures) When missing, growth factor Setting it to 1 ensures robustness in forecasting and avoids unfounded predictions. This improvement significantly enhances the accuracy of sales forecasts for newly added companies, making the allocation of sales targets more reasonable and precise.

[0069] In the aforementioned plan, for the new company, in addition to the month-on-month comparison of last month this year... Other auxiliary indicators can also be considered to determine sales growth factors. For example, one can refer to the average growth rate of new companies of similar size in the same industry, or calculate the average growth rate based on their cumulative sales and number of months since inception. If the new company has a clear business plan or marketing strategy, this information can also be taken into consideration and appropriately adjusted based on expert experience. (When comparing the month-on-month growth rate of the previous month...) Missing and growth factors Setting it to 1 is a conservative strategy. In practice, alternative values ​​can be considered based on business risk appetite and data availability. For example, a default, lower growth factor (such as 0.95) can be set, or a preliminary assessment can be made based on non-sales data such as the new company's registered capital and number of employees. Furthermore, the month-on-month weighting coefficient should be adjusted accordingly. The selection can be dynamically adjusted according to the growth stage of the new company. For example, for a startup, it can be given... Larger values ​​are used to more sensitively capture early growth.

[0070] In some embodiments, the year-on-year weighting coefficient, the previous month's year-on-year weighting coefficient, and the previous month's month-on-month weighting coefficient are dynamically changed based on user-predicted demand; user-predicted demand includes long-term demand, medium-term demand, and short-term demand.

[0071] This solution aims to address the problem of inaccurate forecasts caused by fixed weights in existing sales forecasting algorithms, which are ill-suited to adapt to varying forecasting needs and market changes. Traditional methods typically use fixed weighting coefficients, which cannot be flexibly adjusted to reflect users' different emphases on long-term, medium-term, or short-term trends. This solution solves this technical problem by introducing a dynamic weighting coefficient adjustment mechanism. Its technical basis lies in the year-on-year weighting coefficient... Year-on-year weighting coefficient of the previous month Month-on-month weighting coefficient The calculation of the sales growth factor G can be dynamically adjusted based on users' projected demand (long-term, medium-term, and short-term). This mechanism allows the calculation of the sales growth factor G to better match users' focus on growth trends across different time dimensions, thereby improving the rationality and accuracy of sales target allocation and enhancing the dynamic adaptability of sales target allocation to seasonal fluctuations and market changes.

[0072] In this scheme, the year-on-year weighting coefficient is used in the multi-weighted product formula of sales growth factor S4. Year-on-year weighting coefficient of the previous month Month-on-month weighting coefficient Dynamic adjustment is one of its core innovations. These weighting coefficients are no longer fixed values, but change dynamically based on users' predicted needs. Specifically, users' predicted needs can include: long-term needs, where users are more concerned about the company's long-term development trend and stability; medium-term needs, where users are more concerned about the company's performance and medium-term growth potential in the current cycle; and short-term needs, where users are more concerned about the company's recent sales momentum and market response. When users select long-term needs, the system will increase the year-on-year weighting coefficient. Reduce the year-on-year weighting coefficient of the previous month. Month-on-month weighting coefficient This makes the sales growth factor It places more emphasis on reflecting the company's long-term growth potential. For example, It can be set to 0.6. It is 0.2. The value is 0.2. When a user selects a medium-term demand, the system will increase the year-on-year weighting coefficient from the previous month. This makes the medium-term trend in growth factors It occupies a dominant position. For example, It can be set to 0.2. It is 0.6. The value is 0.2. When a user selects short-term needs, the system will increase the month-on-month weighting coefficient c, making the recent sales momentum have a greater impact on the growth factor. The impact is greatest. For example, It can be set to 0.2. It is 0.2. The value is 0.6. This dynamic adjustment mechanism gives users greater flexibility, allowing them to customize the sales growth factor based on actual business scenarios and forecast targets. The calculation logic is different. For example, during periods of high market volatility or the early stages of a new product launch, users might prioritize short-term needs to capture immediate market feedback. In contrast, during the company's strategic planning phase, the focus might be more on long-term needs.

[0073] This plan uses a year-on-year weighting coefficient. Year-on-year weighting coefficient of the previous month Month-on-month weighting coefficient By dynamically adjusting based on predicted user demand, this technology solves the problem of inaccurate predictions caused by fixed weights in existing technologies. This dynamic adjustment mechanism enhances the sales growth factor... The calculation can better match users' focus on growth trends across different time dimensions, thereby improving the rationality and accuracy of sales target allocation. It enhances the dynamic adaptability of sales target allocation to seasonal fluctuations and market changes, making the forecast results more intelligent and flexible.

[0074] In the above scheme, the weighting coefficient , , The dynamic adjustment of the model can be based not only on manually set forecasting needs but also on automated adjustments through machine learning models. For example, a model can be trained using external data such as historical forecast errors, market sentiment indices, and industry cycles to automatically optimize weight coefficients, enabling adaptive adjustments to improve forecast accuracy. Furthermore, user forecasting needs can be more finely categorized, such as "steady growth needs" and "aggressive growth needs," each corresponding to a set of preset weight coefficients. The adjustment range of the weight coefficients can be set from 0 to 1, and the sum of the three can be limited to 1 to maintain their mathematical significance as weights. In practical applications, a user interface can be provided, allowing users to intuitively drag sliders to adjust weights and preview forecast results in real time, thereby improving user experience and model usability.

[0075] In some embodiments, in step S1, data cleaning includes removing duplicate records and correcting erroneous data formats; missing value filling is performed using linear interpolation or industry averages; outlier detection is performed by identifying and processing outliers using the Z-score method.

[0076] This solution aims to address the issue of low-quality raw sales data in existing sales forecasting algorithms, which includes duplicate records, incorrect formatting, missing values, and outliers, leading to reduced forecast accuracy. Traditional methods often neglect data preprocessing, directly using raw data for calculations, thus affecting the reliability of the final forecast results. This solution resolves this technical problem by refining and standardizing the S1 step of data preprocessing. Its technical foundation lies in employing systematic data cleaning, missing value imputation, and outlier detection methods to ensure that the raw sales data input to the forecasting model is high-quality, accurate, and complete. Data cleaning improves data consistency by removing duplicate records and correcting incorrect data formats; missing value imputation ensures data integrity through linear interpolation or industry averages; and outlier detection uses the Z-score method to identify and handle extreme data points, improving data reliability. This comprehensive data preprocessing workflow provides a solid data foundation for subsequent sales forecasting, thereby improving the rationality and accuracy of sales target allocation.

[0077] In this solution, data preprocessing S1 is the foundation of the entire sales forecasting method, aiming to ensure the quality and reliability of the input data. This step includes three key stages: data cleaning, missing value imputation, and outlier detection. Data cleaning improves the accuracy and consistency of the data, specifically including: removing duplicate records, i.e., identifying and deleting identical redundant entries in the dataset to avoid duplicate calculations of statistical results; correcting erroneous data formats, such as standardizing date formats and removing non-numeric characters from numerical data, ensuring that the data conforms to standard formats for easy subsequent processing. Missing value imputation handles blank or invalid data in the dataset due to various reasons to ensure data integrity. This solution uses linear interpolation or industry averages for imputation: linear interpolation is suitable for data with time-series characteristics, extrapolating missing values ​​based on the trend of preceding and following data points. For example, if sales figures for a certain month are missing, linear prediction can be made based on sales figures for preceding and following months; imputation using industry averages is suitable for situations lacking trends or where individual data vary significantly, supplementing missing data by referencing the overall industry level. Outlier detection is used to identify and handle extreme values ​​in a dataset that deviate from the normal range. These outliers may be caused by data entry errors, system malfunctions, or special events, and if left untreated, they can severely impact the accuracy of the predictive model. This solution uses the Z-score method to identify and handle outliers: the Z-score method determines whether a data point is an outlier by calculating the multiple of the standard deviation of the data point from the mean. A threshold is typically set (e.g., 2 or 3 standard deviations). Data points exceeding this threshold are marked as outliers and, depending on the situation, are deleted, replaced with the mean or median, etc. Through these meticulous data preprocessing steps (S1), this solution ensures the high quality of the original sales data, providing reliable input for subsequent calculations of sales baseline values ​​(S3), sales growth factors (S4), etc.

[0078] This solution addresses the issue of low quality in raw sales data by introducing data cleaning, missing value imputation, and outlier detection in step S1. Removing duplicate records and correcting erroneous data formats improves data consistency and accuracy; using linear interpolation or industry averages for missing value imputation ensures data integrity; and identifying and handling outliers using the Z-score method significantly improves data reliability. This comprehensive data preprocessing provides a solid data foundation for subsequent sales forecasting, thereby improving the rationality and accuracy of sales target allocation and enhancing the robustness of the algorithm.

[0079] In the above solutions, data cleaning, besides removing duplicate records and correcting erroneous data formats, can also include data type conversion, unit standardization, and string standardization. For example, sales data can be converted from string to numeric type, and different currency units can be standardized to a single unit. Missing value imputation, in addition to linear interpolation and industry averages, can also employ regression prediction, K-nearest neighbor (KNN) algorithms, or imputation based on historical data. For example, for seasonal data, values ​​from the same period last year can be imputed. Outlier detection, besides the Z-score method, can also use box plots, DBSCAN clustering algorithms, or Isolation Forest. For identified outliers, the handling method can be flexibly chosen according to the business scenario, such as direct deletion, replacement with nearest neighbor values, median, or mean, or marking them as special events for separate analysis. Regarding parameter range description, the Z-score threshold can be adjusted according to the characteristics of the data distribution. For example, for normally distributed data, 2 or 3 standard deviations are commonly used thresholds, but for non-normally distributed data, a stricter or more lenient threshold may be needed.

[0080] In some embodiments, in step S5, if the total adjusted sales value of all companies is 0, the allocation coefficient for a single company is the proportion of the company's sales revenue in the previous month to the total sales revenue of all companies in the previous month; if the total sales revenue in the previous month is also 0, the allocation coefficient is the average value.

[0081] This solution aims to address the problem that existing sales forecasting algorithms fail to fully consider extreme cases when calculating the allocation factor S5. For example, when the sum of adjusted sales values ​​of all companies is 0, traditional methods may fail to calculate the allocation factor due to the divisor being zero, leading to forecast interruptions or errors. This extreme case may occur when the base sales value and growth factor of all companies are 0 or at extremely low values. This solution resolves this technical problem by refining the calculation rules for the allocation factor S5. Its technical basis lies in introducing a backup calculation logic when the sum of adjusted sales values ​​of all companies is 0: first, it attempts to allocate based on the proportion of the previous month's sales; if the sum of the previous month's sales is also 0, then it uses the average value for allocation. This multi-level backup mechanism ensures that the allocation factor S5 can be effectively calculated under any extreme circumstances, thereby guaranteeing the continuity and robustness of the sales forecasting process and improving the rationality and accuracy of sales target allocation.

[0082] In this scheme, extreme cases are considered when calculating the adjusted sales value and allocation coefficient in step S5. Normally, the allocation coefficient for a single company is its adjusted sales value divided by the sum of the adjusted sales values ​​of all companies. However, in certain special business scenarios, the sum of the adjusted sales values ​​of all companies may be 0. This could be because the sales baseline value S3 and sales growth factor S4 for all companies are calculated to be 0, or all of them are 0 after adjustment. To avoid calculation errors or program interruptions due to division by zero, this scheme introduces a backup logic: if the sum of the adjusted sales values ​​of all companies is 0, the allocation coefficient for a single company is no longer calculated based on the adjusted sales value, but rather as the proportion of that company's sales revenue from the previous month to the total sales revenue of all companies from the previous month. The logic behind this approach is that when the adjusted sales value calculated by the model cannot provide a valid basis for allocation, it reverts to the most recent actual sales performance (sales revenue from the previous month) as the allocation benchmark to maintain the continuity of the forecast. However, if the total sales revenue of all companies last month is also zero, this indicates that none of the companies had sales data or zero sales revenue in the most recent period. In this case, any proportional allocation based on historical sales data will be invalid. In this more extreme scenario, this solution further stipulates that the allocation coefficient is the average. That is, the total target sales revenue is evenly distributed among all companies, which is a fair and robust fallback strategy when effective data support is lacking. Through this multi-level backup mechanism, this solution ensures the calculability and rationality of the allocation coefficient S5 under any extreme data conditions, thereby improving the robustness and accuracy of sales target allocation.

[0083] This solution addresses the allocation challenge when the total adjusted sales value of all companies equals zero by introducing a multi-level backup mechanism in step S5. When this total is zero, allocation is performed using the previous month's sales revenue as a percentage, ensuring forecast continuity. If the previous month's sales revenue also equals zero, average allocation is further employed, providing a robust fallback strategy. This improvement significantly enhances the robustness and accuracy of the sales revenue forecasting algorithm, ensuring reasonable allocation even under extreme data conditions and effectively resolving the disconnect between monthly target allocation and actual business scenarios.

[0084] In the above schemes, besides the previous month's sales ratio and average, other alternative allocation strategies can be considered. For example, a weighted average allocation can be performed based on non-sales data such as the company's registration date, number of employees, and market potential, or expert experience can be introduced for manual intervention. In certain specific industries, if the company has clear annual or quarterly targets, these targets can also be used as the basis for allocation. For cases where "the total sales of the previous month are also 0," if the number of companies is small, the total target can be evenly distributed among all companies, or a weighted average can be applied based on their highest historical sales. Regarding the parameter range, the allocation coefficient typically ranges from 0 to 1, and the sum of the allocation coefficients for all companies should be 1. In extreme cases, if the calculated allocation coefficient is negative or abnormally large, additional verification and correction should be performed.

[0085] Figure 3 A schematic diagram of a sales forecasting system based on a fixed quota, provided in an embodiment of this application, is shown below. Figure 3 As shown, a sales forecasting system 300 based on quotas in this embodiment includes: a data preprocessing module 301, a type differentiation module 302, a basic analysis module 303, a growth analysis module 304, an adjustment and summary module 305, and a sales forecasting module 306.

[0086] The data preprocessing module 301 is used to perform data cleaning, missing value imputation, outlier detection, and standardization on the raw sales data; the type differentiation module 302 is used to classify companies as old or new companies based on whether they have complete sales data from last year; the basic analysis module 303 is used to calculate the basic sales value. If the company is an old company, the sales amount of the same month last year is used as the basic sales value; if the company is a new company, the sales amount of the previous month this year is used as the basic sales value; the growth analysis module 304 is used to calculate the sales growth factor. If the company is an old company, the sales growth factor is calculated according to a multi-weighted product formula; if the company is a new company... Based on the month-on-month data from the previous month, the sales growth factor is determined; the adjustment and summary module 305 is used to calculate the adjusted sales value and the allocation coefficient. The adjusted sales value of a single company is the product of the sales base value and the sales growth factor. The sum of the adjusted sales values ​​of all companies is the sum of the adjusted sales values ​​of a single company. The allocation coefficient of a single company is the adjusted sales value of that company divided by the sum of the adjusted sales values ​​of all companies; the sales forecast module 306 is used for total target allocation. Based on the product of the total sales target and the company's allocation coefficient, the predicted sales amount of a single company is obtained. Based on the sum of the predicted sales amounts of all the single companies, the total predicted sales amount is determined.

[0087] Optionally, in the growth analysis module 304, the growth factor is used to adjust the sales base value, reward companies with good sales growth performance by increasing the sales revenue allocation ratio of the corresponding companies, and at the same time constrain companies with poor sales growth performance by reducing the sales revenue allocation ratio of the corresponding companies.

[0088] Optionally, in the basic analysis module 303, if the company's sales revenue for the same month last year is 0, then the company's total sales revenue for last year divided by 12 is used as the basic sales value.

[0089] Optionally, in the growth analysis module 304, the multi-weighted product formula is specifically the following formula: ; in, For the aforementioned sales growth factor, Compared to the same period last year, This is the year-on-year weighting coefficient. Compared to the same period last year, This is the year-on-year weighting coefficient for the previous month. Compared to the previous month this year, This is the month-on-month weighting coefficient for the previous month.

[0090] Optionally, in the growth analysis module 304, the year-on-year total is used to characterize the long-term growth trend of the company's current sales. The year-on-year comparison from the previous month this year is used to characterize the current company's mid-term sales growth trend, closely reflecting the sales growth trend within the current cycle; The month-on-month comparison mentioned earlier this year is used to characterize the short-term growth trend of sales in the current formula, reflecting the recent sales growth momentum.

[0091] Optionally, in the growth analysis module 304, if the month-on-month comparison of this year exists, then the growth factor is (1 + month-on-month comparison of this year). If the month-on-month comparison for the previous month of this year is missing, then the growth factor is set to 1.

[0092] Optionally, in the growth analysis module 304, the year-on-year weighting coefficient, the previous month's year-on-year weighting coefficient, and the previous month's month-on-month weighting coefficient are dynamically changed based on user-predicted demand; the user-predicted demand includes long-term demand, medium-term demand, and short-term demand.

[0093] Optionally, in the data preprocessing module 301, the data cleaning includes removing duplicate records and correcting erroneous data formats; the missing value filling uses linear interpolation or industry average values; and the outlier detection is performed by identifying and processing outliers using the Z-score method.

[0094] Optionally, in the adjustment summary module 305, if the total sales value of all companies after adjustment is 0, the allocation coefficient for a single company is the proportion of the company's sales revenue in the previous month to the total sales revenue of all companies in the previous month; if the total sales revenue in the previous month is also 0, the allocation coefficient is the average value.

[0095] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

Claims

1. A sales forecasting method based on quotas, characterized in that, include: S1. Perform data cleaning, missing value imputation, outlier detection, and standardization on the raw sales data; S2. Companies are categorized as either "old companies" or "new companies" based on whether they have complete sales data from the previous year. S3. Calculate the sales baseline. If the company is an established company, the sales revenue of the same month last year shall be used as the sales baseline. If the company is a new company, the sales revenue of the previous month this year shall be used as the sales baseline. S4. Calculate the sales growth factor. If the company is an established company, calculate the sales growth factor according to the multi-weighted product formula. If the company is a new company, determine the sales growth factor based on the month-on-month data of the previous month of this year. S5. Calculate the adjusted sales value and the allocation factor. The adjusted sales value of a single company is the product of the sales base value and the sales growth factor. The sum of the adjusted sales values ​​of all companies is the sum of the adjusted sales values ​​of a single company. The allocation factor of a single company is the adjusted sales value of that company divided by the sum of the adjusted sales values ​​of all companies. S6. Total Target Allocation: The projected sales revenue for each company is obtained by multiplying the total sales revenue target by the company's individual company allocation coefficient. The total projected sales revenue is determined by summing the projected sales revenue of all individual companies.

2. The method according to claim 1, characterized in that, The growth factor is used to adjust the sales base value, rewarding companies with good sales growth performance by increasing their sales revenue allocation ratio, while constraining companies with poor sales growth performance by reducing their sales revenue allocation ratio.

3. The method according to claim 1, characterized in that, If the company's sales revenue for the same month last year is 0, then the company's total sales revenue for the previous year divided by 12 is used as the basic sales value.

4. The method according to claim 2, characterized in that, The multi-weighted product formula is specifically as follows: ； in, For the aforementioned sales growth factor, Compared to the same period last year, This is the year-on-year weighting coefficient. Compared to the same period last year, This is the year-on-year weighting coefficient for the previous month. Compared to the previous month this year, This is the month-on-month weighting coefficient for the previous month.

5. The method according to claim 2, characterized in that, The year-on-year total is used to characterize the long-term growth trend of the company's current sales. The year-on-year comparison from the previous month this year is used to characterize the current company's mid-term sales growth trend, closely reflecting the sales growth trend within the current cycle; The month-on-month comparison mentioned earlier this year is used to characterize the short-term growth trend of sales in the current formula, reflecting the recent sales growth momentum.

6. The method according to claim 4, characterized in that, If the month-on-month comparison of this year exists, then the growth factor is (1 + month-on-month comparison of this year). If the month-on-month comparison for the previous month of this year is missing, then the growth factor is set to 1.

7. The method according to claim 4, characterized in that, The year-on-year weighting coefficient, the previous month's year-on-year weighting coefficient, and the previous month's month-on-month weighting coefficient are dynamically changed based on user-predicted demand. The predicted user demand includes long-term demand, medium-term demand, and short-term demand.

8. The method according to claim 5, characterized in that, In step S1, the data cleaning includes removing duplicate records and correcting erroneous data formats; The missing values ​​are filled using linear interpolation or by using industry averages; The outlier detection is performed by identifying and processing outliers using the Z-score method.

9. The method according to claim 5, characterized in that, In step S5, if the total adjusted sales value of all companies is 0, then the allocation coefficient for a single company is the proportion of that company's sales revenue in the previous month to the total sales revenue of all companies in the previous month. If the total sales revenue of the previous month is also 0, then the average value of the allocation coefficient is taken.

10. A sales forecasting system based on quotas, characterized in that, The method applied to any one of claims 1-9 includes: The data preprocessing module is used to perform data cleaning, missing value imputation, outlier detection, and standardization on the raw sales data. The type differentiation module is used to classify companies as old companies or new companies based on whether they have complete sales data from last year; The basic analysis module is used to calculate the basic sales value. If the company is an established company, the sales revenue of the same month last year is used as the basic sales value; if the company is a new company, the sales revenue of the previous month this year is used as the basic sales value. The growth analysis module is used to calculate the sales growth factor. If the company is an established company, the sales growth factor is calculated according to the multi-weighted product formula; if the company is a new company, the sales growth factor is determined based on the month-on-month data of the previous month of this year. The adjustment summary module is used to calculate the adjusted sales value and the allocation coefficient. The adjusted sales value of a single company is the product of the sales base value and the sales growth factor. The sum of the adjusted sales values ​​of all companies is the sum of the adjusted sales values ​​of a single company. The allocation coefficient of a single company is the adjusted sales value of that company divided by the sum of the adjusted sales values ​​of all companies. The sales forecasting module is used for total target allocation. It obtains the projected sales revenue of each company by multiplying the total sales revenue target by the allocation coefficient of each company. The total projected sales revenue is determined by summing the projected sales revenue of all companies.

Images