JP2003323542A - Business operation support device based on prediction of sale day - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a business operation support device for predicting the next sale day, especially for predicting a sale day on which the sales activity can be concentrated on such a customer or commodity whose order is abnormally delayed regardless of the seasonal variation. <P>SOLUTION: The device comprises a sales performance database, a customer/ commodity extracting part for extracting a specific combination of a customer and a commodity with little variation in the past pace of the sales, a sale day prediction determining part for predicting the next sale day of the extracted customer and commodity and determining the probability of the sale, and a warning output part for warning about the customer and commodity whose sale day is near or past. The device also has a previous year sales result time- series output part, so that the sales performance at the same time in the previous year can be referred to only for the warned customer and commodity. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a sales support device based on sales date prediction using a computer, and in particular,
It belongs to a sales support device that predicts sales dates for sales businesses that repeatedly sell the same product to customers.

[0002]

2. Description of the Related Art Conventionally, as a sales support device based on sales date prediction, there is an example of a device for performing a sales cycle management called "rice grain management" in a rice grain sales industry. "Rice management"
For customers who have sold a certain number of times in the past, take the average of the past sales cycle to predict the next sales date, and confirm the order by telephone on the forecast sales date. This is the method to use. However, there is a problem that it is difficult to apply the above-mentioned “rice box management” to sales industries other than the rice grain sales industry. The rice sales business has the feature that it repeatedly sells the same product to the same customer at almost the same sales pace.

However, in general, in the case of a sales business other than the rice sales business, there are many kinds of products to be handled, and there are products that are repeatedly sold and products that are sold only a few times at most. Further, among the products that are repeatedly sold, there are products whose sales pace is almost constant and products whose sales pace fluctuates greatly. Furthermore, even for the same product, there are customers whose sales paces are almost constant and those whose sales paces vary greatly. For customers and products whose sales pace fluctuates, the prediction accuracy will be low even if the next sales date is predicted from the past sales cycle. As described above, since many customers and products have low prediction accuracy, the sales support device based on sales date prediction using "rice box management" is not practical in the sales business other than the rice sales business.

[0004] Therefore, in order to apply the sales date forecast to the sales activities of the sales business other than the rice sales business, the customer who extracts only the combination of the customer and the product whose past sales pace has little fluctuation.
A sales support device having a product extraction unit was invented (demand forecasting system developed by K's Co., Ltd .: 2001).
The customer / commodity extraction unit statistically processes the past sales results several times, and extracts only the combination of the customer and the product in which the cumulative speed of the sales quantity does not change. For the extracted customers and products, the sales dates are highly accurate, so effective sales activities such as confirming orders according to the predicted sales dates can be performed.

[0005]

(1) In the conventional sales support device based on sales date prediction, when the order is delayed compared to the estimated sales date, it is determined whether the delay is normal or abnormal compared to the usual year. There was a problem that I could not. Some of the products are
Demand for certain products declines during certain seasons, resulting in a sharp decline in sales pace. In the conventional sales date forecast, the sales date is forecasted using the latest sales results, so it is possible to accurately forecast the sales date only when the latest sales pace continues, and when the sales pace suddenly decreases. Cannot be predicted correctly. Therefore, based on the forecasted sales date by the conventional forecasting method,
When an order is delayed, it cannot be determined whether it is a seasonal decrease in normal sales pace or an abnormal decrease in sales pace that is not related to seasonal fluctuations. Therefore, the problem to be solved by the present invention is that if the order is delayed with respect to the predicted sales date, it is a decrease in the normal sales pace of the season, or an abnormal sales that is not related to seasonal fluctuations. It is to be able to judge whether it is a decrease in pace.

(2) Further, the conventional sales support device based on sales date prediction has a problem that the result of the sales activity using the device cannot be evaluated later. In a conventional sales support device based on sales date prediction, the estimated sales date at that time and the final sales date are compared to determine which customers / products are close to the estimated sales date and which are past the estimated sales date. To output a warning. Based on the warning, the sales person confirms whether or not the customer who has been warned has ordered the corresponding product, and makes an effort to increase the number of orders. In order to manage the sales activity, a sales target is set for a warned customer / product, and sales performance is evaluated after a certain period of time,
It is necessary to repeat the cycle of goal setting and evaluation. However, the conventional sales support device only outputs the warning at the present time and does not manage the sales record of the warned customer / product after the warning date, so that the result of the sales activity cannot be evaluated. . Therefore, the problem to be solved by the present invention is to enable appropriate management of sales activities by making it possible to evaluate the sales performance after the warning date regarding the warned customer / product.

[0007]

In order to solve the above-mentioned problem (1), the present invention has a sales record database for the previous year and this year. The sales record database includes information on customer name, product name, sales volume (sales quantity or sales amount), and sales date. Furthermore, the present invention has a customer / product extraction unit that extracts a specific combination of a customer and a product. The customer / commodity extraction unit extracts from the combination of all customers and commodities only the combination of customers and commodities that have been sold a certain number of times in the past and whose sales pace has not fluctuated the last few times. To do.

Further, the present invention has a sales date prediction determination unit for predicting the next sales date from the sales record of the latest several times for the extracted customer / product. Since the number of past sales of the extracted customer / product is a certain number of times or more, it is possible to statistically predict the next sales date from the sales results. If the frequency with which the customer consumes the product changes, and if the customer purchases the product irregularly from multiple vendors, the sales pace of the customer / product changes. In the case of customers / products whose sales pace fluctuates, there is a problem that the prediction accuracy is low even if the next sales date is statistically predicted from the sales results of the latest several times. However, according to the present invention, since the customer / commodity extraction unit extracts only the combination of the customer and the product whose sales pace does not fluctuate, the sales date can be predicted with high accuracy.

Further, the present invention has a warning output unit for warning a customer / product near the predicted sales date and a customer / product whose sales are behind the predicted sales date. Confirming an order for a customer / product close to the estimated sales date will receive an order with a high probability and increase the sales volume, and at the same time, prevent the customer from missing an order, which is advantageous for both the company and the customer. .
Also, if you confirm an order for a customer or product whose order is behind the expected sale date, you can quickly discover that demand changes and that the customer is robbed of by another company, and at the same time, the customer can prevent omission of the order. And both the customer and the customer.

Further, the present invention is provided with a previous year performance time series output unit for outputting the sales performance of the same period of the previous year in time series only for warning customers and products. Since the sales date prediction determination unit predicts the next sales date based on the latest sales pace by using the sales results of the latest several times, it is not possible to predict a sudden decrease in the seasonal sales pace. Therefore, if only the warning output from the warning output unit is delayed, if the order is delayed, the delay is a seasonal decrease in the normal sales pace or an abnormal sales pace not related to the seasonal fluctuation. There is a problem that it is not possible to judge whether or not it is a decrease. However, in the present invention,
In order to output the sales results of the same customer for the same period of the previous year in a time series, refer to whether the sales volume has decreased in the same period of the previous year, and the delay of the current order is the same as the previous year. It is possible to judge whether the sales pace is decreasing or the sales pace is abnormally different from the previous year, which is not related to seasonal fluctuations.

Further, in order to solve the above-mentioned problem (1), the present invention is directed to a customer whose sales volume is decreasing in the same period of the previous year.
A customer / commodity secondary extraction unit is provided that automatically determines a product and identifies and outputs it with a warning output or excludes it from a warning output. The customer / commodity secondary extraction unit uses the previous year's sales record database to identify the customers whose sales volume has decreased during the same period of the previous year.
It has a function of automatically judging and extracting a product, and excluding it from the output of the warning output unit or identifying and outputting it.
By automatically identifying customers and products whose sales pace decreases seasonally by the customer / product secondary extraction unit, if the order is delayed compared to the predicted sales date, the delay will be normal for the season. It can be more easily determined whether it is a decrease in sales pace or an abnormal decrease in sales pace that is not related to seasonal fluctuations.

In order to solve the above problem (2), the present invention has an extraction / judgment history database. The extraction / judgment history database records the extraction history of the customer / commodity extraction unit and the customer / commodity secondary extraction unit, and further records the sales date prediction judgment history of the sales date prediction judgment unit. Furthermore, the present invention has an evaluation output unit that outputs the sales record after the warning date after a certain period of time for customers and products that have been warned in the past. The evaluation output unit refers to the extraction / judgment history database and the sales record database, and totalizes and outputs the sales records after the warning date for past warning customers and products. In business activities, first, a sales target is set for the product / customer warned by the warning output unit. next,
Warning For products and customers, the sales representative confirms the order with the customer and strives to achieve the sales target. After a certain period of time, the output of the evaluation output unit is referred to, and it is evaluated whether or not the sales record after the warning date has reached the sales target. According to this evaluation, if necessary, measures are taken, and at the same time, it is possible to effectively manage sales activities by reflecting on the performance evaluation of each employee.

[0013]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to FIG.
~ It demonstrates with reference to FIG. The following embodiment is a form of a computer system that combines a sales management system and a sales support system. FIG. 1 shows a device configuration in the embodiment. The device configuration includes a sales management system server, a sales support system server, a sales support system printer, and a network device connecting these. The sales management system manages sales processing such as order processing, sales processing, and invoicing processing, and has a sales record database for the past two years or more. The sales support system copies a part of the sales performance database from the sales management system to the database of its own system through the network, extracts customers and products, makes a judgment on the sales date, and outputs a warning through the display screen and the printer. Output and evaluation output. In addition to the device configuration of FIG. 1, a device configuration in which a sales management system server and a sales support system server are integrated is also conceivable. However, if you continue to use the existing sales management system and add a new sales support function by forecasting sales dates, the device configuration that has a sales support system server separately from the existing sales management system server Addition work becomes easy.

FIG. 2 shows an information processing structure processed by the sales support system server. Details of the information processing will be described with reference to FIGS.
An outline of information processing will be described with reference to. First, from the sales result database managed by the sales system server,
Copy a part of the previous year and this year's data to the previous year's sales record database (A) and this year's sales record database (B) managed by the sales support system server. The sales record database (A and B) manages customer names, product names, sales quantities, sales amounts, and sales dates. Next, in the customer / commodity primary extraction unit (C), the number of past sales is a certain number or more, and
Only the combinations of customers and merchandise whose sales pace does not change are extracted.

Next, in the sales date prediction judging section (D),
For the extracted customers and products, the next sale date is predicted by the statistical processing described later. Next, in the warning output unit (E), the sales probability ranks are divided according to the difference between the predicted sales date and the final sales date, and the customer / product is output for each rank. Further, in the previous year actual result time series output unit (F), the sales results of the same period of the previous year are output in time series only for the customers and products for which the warning output unit (E) warned.
Further, in the customer / product secondary extraction section (G), the customer / product which is warned by the warning output section (E) is judged as a customer / product whose sales pace is decreasing in the same period of the previous year, and a warning is issued. You can exclude it from the output or display the warning output with an identifying symbol. The extraction / judgment history database (H) records a history of customer / commodity information extracted by the customer / commodity extraction unit (C and G) and predicted sales date information judged by the sales date prediction judgment unit (D). Next, the evaluation output unit (I), based on the information in the extraction / judgment history database (H) and the sales record database (B), records the sales record after the warning date for the customer / product that was previously warned. Is totaled and output.

Next, with reference to the graph of FIG. 3, a method of extracting a combination of a customer and a product in the customer / product primary extraction section (C) whose sales pace does not fluctuate will be described. The customer / commodity primary extraction unit (C) first extracts combinations of customers and commodities that have been sold 5 times or more in the past. Next, based on the past five sales history, the fluctuation of the sales pace is determined. FIG. 3 is a line graph showing a history of accumulated sales volume for an example of a combination of a specific customer and a product, in order to explain a method of determining a change in the sales pace from the past five sales history. . The vertical axis of the graph represents the cumulative value of the sales amount, and the horizontal axis of the graph represents the sales date. Graph A. Is an example of a combination of a customer and a product whose sales pace does not fluctuate. The polygonal line connecting the past five sales history is close to a straight line, which means that the sales pace in the past five sales is almost constant. In this case, there is a high possibility that the product will be sold at the same sales rate from the next time. Graph B. Is an example of a combination of customers and products whose sales pace changes. The polygonal line connecting the past five sales history is largely deviated from the straight line, which means that the sales pace in the past five sales varies greatly. In this case, the sales pace after the next time cannot be predicted. In the customer / product primary extraction section (C), the graph A.1 is selected from all combinations of customers and products that have been sold 5 times or more in the past. Only the combinations of customers and merchandise whose sales pace does not fluctuate are extracted by statistical processing using the least squares method. If the least squares method is used, whether the polygonal line is close to the straight line or deviates from the straight line is evaluated with a statistic called residual,
It is possible to extract only the combination of the customer and the product whose sales pace does not fluctuate.

In addition to the function of this embodiment, a filter function may be provided for removing a combination of a customer and a product designated in advance from extraction. In the filter function, combinations of customers and products that do not require warning are registered in advance. Examples of combinations of customers and products that do not require warnings include customers / products for which the cause of lost orders is known, and extreme seasonal products.

Next, using the graph of FIG. 4, a method of predicting the next sales date in the sales date prediction determination section (D),
A method of determining the sales probability rank in the warning output unit (E) will be described. FIG. 4 is a line graph showing a history of cumulative sales volume for an example of a combination of a specific customer and a product. The vertical axis of the graph represents the cumulative value of the sales volume, and the horizontal axis of the graph represents the sales date. The diamonds (◆) in the graph represent past sales results, and the circles (○) represent forecasts of the next sales date. Also, the current date is represented by a vertical line. The estimated sales date is calculated by the sales pace calculated from the past five sales results by the least squares method and the average sales amount. Graph A. Is an example in which the current date has not yet reached the predicted sales date. The customers and products in this stage have a high future sales probability, but the sales probability at this point is still low. Graph B. Is an example where the current date has just reached the projected sale date. Customers and products at this stage have a high sales probability. Therefore, if sales activities are carried out for customers and products at this stage, there is a high probability that an order will be received, and at the same time, customers will be able to prevent omission of orders, which is advantageous for both themselves and customers. Graph C. Is an example in which the current date is far past the predicted sales date. The sales probability of customers and merchandise at this stage has decreased, and there is likely to be no demand, or the customer may have purchased from another company. Therefore, by conducting sales activities for customers and products at this stage, it is possible to quickly discover changes in demand and the loss of customers to other companies,
At the same time, it is possible for the customer to prevent omission of orders, which is advantageous for both the customer and the customer. In the warning output section (E), FIG.
Depending on the relationship between the current date and the forecasted sales date
Determine the sales probability rank. Although the definition range of the sales probability rank varies depending on the type of industry, for example, rank A is a customer / product that has not reached the predicted sales date, rank B is a customer / product that is close to the predicted sales date, and rank C is the predicted sales date. The customer / product will be slightly delayed, and the rank D will be defined to be the customer / product significantly delayed from the predicted sales date.

FIG. 5 shows a warning list output by the warning output unit (E). The warning output unit (E) outputs a warning list for customers and products extracted by combining extraction conditions such as sales probability rank, predicted sales date, sales person in charge, customers, and products according to the purpose. For example, the salesperson uses the rank B warning list close to the predicted sales date for daily visit activities and telephone sales activities. Further, the sales manager checks the decrease in demand with reference to the warning list of rank D, which is largely behind the predicted sales date. Further, by managing the information of the delivery route and outputting and utilizing the warning list of rank B close to the estimated sales date for each delivery route, the delivery person can easily carry out business activities simultaneously with delivery.

FIG. 6 shows a previous year result list output by the previous year result time series output section (F). In the previous year's result list, the sales results for the same period of the previous year are output in chronological order only for the customers and products warned by the warning output unit (E). By referring to this previous year's performance list, it is possible to confirm the increase or decrease in sales volume during the same period of the previous year. For customers / products whose orders are delayed compared to the forecasted sales date, if the sales volume decreases in the same period of the previous year, it is the customers / products whose seasonal demand decreases and the order delay is normal. I can judge. On the other hand, for customers / products whose orders are delayed compared to the predicted sales date, if the sales volume does not decrease in the same period of the previous year, it means that the customers / products have no seasonal demand decrease, and the order delay is Since it is abnormal, it can be determined that the cause needs to be confirmed. The customer / product secondary extraction unit (G) extracts only the customers / products whose sales pace has decreased in the same period of the previous year from the customers / products warned by the warning output unit (E) and issues a warning. You can exclude it from the output or show an identifying symbol in the warning output. In the warning list of FIG. 6, a double circle mark (⊚) is attached to a customer / product whose sales pace has decreased in the same period of the previous year so that the customer / product can be distinguished from other customers / products.

FIG. 7 shows a performance evaluation list output by the evaluation output unit (I). In the performance evaluation list, the sales performance after the warning date is output for the customer / product that has been warned in the past. In business activities, first, a future sales target is set for the customer / product for which the warning output unit (E) warned. Next, for the warning product / customer, the sales person confirms the order with the customer and makes a sales effort to achieve the sales target. After a certain period, the evaluation output unit (I) confirms the sales record after the warning date. In this way, it is possible to effectively manage sales activities by evaluating the achievement level of sales targets, taking measures as necessary, and reflecting them in the performance evaluation of individual employees. The output of the warning output unit (E) and the evaluation output unit (I) can be confirmed not only on the print list but also on the display screen. If you confirm on the display screen, and if you select a combination of a specific customer and product, you can display the details of past sales history or display a graph to display the sales of the customer / product that you are interested in. You can investigate the trends in more detail.

[0022]

The present invention predicts the next sales date from the latest sales record, compares the predicted sales date with the current date, determines the sales probability, and outputs a warning. At the same time, the sales in the same period of the previous year can be performed. Achievements can be referenced in chronological order. Therefore, if the order is delayed with respect to the forecasted sales date, refer to the sales results for the same period of the previous year to see if that is a seasonally normal decrease in the sales pace, or if there is an abnormality that is not related to seasonal fluctuations. It will be possible to concentrate on sales activities only to customers and products whose sales pace is abnormally decreasing by judging whether the sales pace is decreasing.

Further, according to the present invention, it is possible to automatically judge a customer / commodity whose sales volume has decreased in the same period of the previous year, exclude it from the warning output, or identify and output the warning output. Therefore, if an order is delayed with respect to the forecasted sales date, it can be more easily determined whether it is a seasonal decrease in normal sales pace or an abnormal decrease in sales pace that is not related to seasonal fluctuations. It becomes possible to judge. Furthermore, the present invention relates to customers and products warned in the past,
Sales results after the warning date can be aggregated and output. For this reason, it is evaluated whether the sales target set at the time of warning output is achieved after a certain period of time, and if necessary, countermeasures are taken and it is reflected in the performance evaluation of individual employees to appropriately conduct sales activities. It becomes possible to manage.

[Brief description of drawings]

FIG. 1 is a diagram showing a device configuration according to an embodiment of the present invention.

FIG. 2 is a diagram showing an information processing structure processed by a sales support system server according to the embodiment of the present invention.

FIG. 3 is a line graph showing a history of cumulative sales volume for an example of a combination of a specific customer and a product in order to explain a change in sales pace.

FIG. 4 is a line graph showing a history of cumulative sales volume for an example of a combination of a specific customer and a product in order to explain a method of predicting a next sales date and a method of determining a sales probability.

FIG. 5 is a diagram of a warning list created and output by a warning output unit according to the embodiment of the present invention.

FIG. 6 is a diagram of a previous year sales record list created and output by a previous year record time series output unit according to the embodiment of the present invention.

FIG. 7 is a diagram of a performance evaluation list created and output by an evaluation output unit according to the embodiment of the present invention.

Claims (3)

[Claims] 1. A customer / product extraction unit that has a sales record database and extracts a specific combination of a customer and a product whose past sales pace has not fluctuated, and predicts the next sale date for the extracted customer / product. In the sales support device, a sales date prediction determination unit that determines a sales probability, and a warning output unit that warns a customer / product close to the predicted sales date and a customer / product that has passed the predicted sales date are provided. A sales support device based on sales date forecast, which is equipped with a previous year time series output unit that outputs the sales results of the same period of the previous year in time series only for products. 2. The sales support device according to claim 1, wherein the sales support device automatically predicts a warning customer / product whose sales volume has decreased during the same period of the previous year, and identifies and outputs the warning output. A sales support device based on sales date prediction, which is characterized by performing or excluding from warning output. 3. The sales support apparatus according to claim 1, further comprising an extraction / judgment history database for recording a history of extracted customers / commodities and a history of predicted sales dates,
Sales support device based on sales date prediction, which is provided with an evaluation output unit that outputs the sales results after the warning date for past warning customers and products.