Estimated commodity sales method

Abstract

The invention discloses a commodity sales pre-estimation method, which relates to the technical field of commodity sales pre-estimation and comprises the following steps: acquiring sales data of all commodities from an e-commerce platform, calculating sales volume correlation coefficients of a target commodity and other different types of commodities, and combining cross sales rates and covariances of the target commodity and related types of commodities to obtain a commodity sales pre-estimation result. Calculating sales influence factors of related types of commodities; obtaining sales data of all historical new commodities from an e-commerce platform, and calculating a historical new commodity price sensitivity coefficient and a historical new commodity sales average proportion; and calculating the related type commodity influence sales volume and the price influence sales volume, and calculating the new product estimated sales volume of the target commodity in combination with the sales quantity of the target commodity in unit time, the related type commodity influence sales volume, the price influence sales volume and the average proportion of the historical new product sales volume. The market condition can be reflected more comprehensively, and the market demand can be grasped more accurately.

Description

Technical Field

[0001] The present invention relates to the technical field of commodity sales estimation, and in particular to a method for estimating commodity sales. Background Art

[0002] With the rapid development of technologies such as big data, artificial intelligence, and machine learning, sales forecasting methods are also evolving. Traditional sales forecasting mainly relies on historical sales data and qualitative analysis methods, such as time series analysis and causal models. However, these methods often have limitations when facing new products, because new products lack sufficient historical data, and factors such as market environment and consumer behavior may also change. In today's highly competitive market environment, the launch of new products and sales forecasting have become key links in corporate strategic planning and operational management.

[0003] In a Chinese invention application with application publication number CN117853157A, a method, apparatus, device and storage medium for estimating the sales quantity of a target object are disclosed, including obtaining current characteristic parameters of the target object; wherein the current characteristic parameters include current independent user visits, and the current independent user visits are derived from a corrected historical independent user visits; each current characteristic parameter is multiplied by a corresponding target weight coefficient, and the sum of the calculated multiple products is used as the estimated sales quantity of the target object.

[0004] In the above invention application, the current characteristic parameters of the target object are obtained; wherein the current characteristic parameters include the current number of independent user visits, and the current number of independent user visits is derived from the corrected historical number of independent user visits; each current characteristic parameter is multiplied by the corresponding target weight coefficient, and the sum of the multiple products calculated is used as the estimated sales quantity of the target object. The sales quantity of the target object is estimated based on the user visits, but for new products, predictions can only be made after the new products are put on the shelves. When the new products have been put on the shelves and started to sell, sales forecasts mean that the company needs to obtain information about its potential market performance after the product has entered the market and may have begun to accumulate inventory. This may cause the company to be unable to adjust its production, inventory, marketing and sales strategies in a timely manner, thereby missing market opportunities or facing the risk of inventory backlogs.

[0005] To this end, the present invention provides a method for estimating commodity sales. Summary of the invention

[0006] 1. Technical issues to be solved

[0007] In view of the deficiencies in the prior art, the present invention provides a method for estimating commodity sales. The present invention calculates the sales volume Yux of the relevant type of commodity according to the sales volume of the relevant type of commodity corresponding to the target commodity per unit time and the sales volume influencing factor of the relevant type of commodity, and calculates the price-influenced sales volume Jys according to the sales volume, sales amount and historical new product price sensitivity coefficient A of the target commodity per unit time. The estimated new product sales volume Xy of the target commodity is calculated by combining the sales volume of the target commodity per unit time, the sales volume Yux of the relevant type of commodity, the price-influenced sales volume Jys and the average proportion of historical new product sales volume Zb. The method can more comprehensively reflect the market situation, more accurately grasp the market demand, optimize inventory management, pricing strategy, marketing strategy and other aspects, thereby reducing market risks, improving customer satisfaction and loyalty, and ultimately achieving business growth and sustainable development, thereby solving the technical problems recorded in the background technology.

[0008] (II) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for estimating commodity sales, characterized in that it includes the following steps:

[0010] Obtain sales data of all products from the e-commerce platform and calculate the sales correlation coefficient ρ between the target product and other different types of products i , combined with the cross-selling rate Jx of the target product and related products ab And covariance Cov(a, b), calculate the sales impact factor α of related types of products b ;

[0011] Obtain sales data of all historical new products from the e-commerce platform, and calculate the price sensitivity coefficient A of historical new products and the average sales volume of historical new products Zb;

[0012] According to the sales quantity Xs of the target product corresponding to the relevant type of product per unit time bj The sales impact factor α of related products b , calculate the sales volume Yux of related types of goods, and the sales volume Xs of the target goods per unit time aj 、Sales amountXe aj , the historical new product price sensitivity coefficient A, calculate the price impact on sales Jys, combined with the target product sales volume per unit time Xs aj , the impact of related types of goods on sales volume Yux, the impact of price on sales volume Jys and the average proportion of historical new product sales volume Zb, calculate the estimated new product sales volume Xy of the target product.

[0013] Furthermore, the sales data of all products are obtained from the e-commerce platform, including the sales quantity Xs per unit time. ij 、Sales amountXeij , calculate the sales correlation coefficient ρ between the target product and other different types of products i :

[0014]

[0015] Among them, i represents the sequential number of the product type, i=1, 2, 3, 4, ..., n, n represents the total number of product types, j represents the sequential number of different brands under the same product type, j=1, 2, 3, 4, ..., m, m represents the total number of different brands under the same product type, and a represents the type number of the target product.

[0016] Furthermore, the sales correlation coefficient ρ between the target product and other different types of products is obtained i , after sorting from largest to smallest, select the top 5 product types as related products of the target product.

[0017] Furthermore, the cross-selling rate Jx between the target product and related products is obtained. ab , and calculate the covariance Cov(a, b) between the target product and related products based on their sales quantity:

[0018]

[0019] Where b represents the serial number of related types of goods, b = 1, 2, 3, 4, 5. The cross-selling rate is the proportion of customers who purchased multiple products in a specific time period. Its calculation formula is as follows:

[0020] Cross-selling rate = (number of customers who purchased multiple products or services) / (total number of customers who purchased products or services) × 100%

[0021] For example, suppose that in a month, a total of 100 customers purchased product A, and 30 of them also purchased product B. Then the cross-selling rate is 30 divided by 100, which is 0.3, or 30%. This means that in this month, 30% of customers purchased both product A and product B.

[0022] Furthermore, the cross-selling rate Jx between the target product and related products is obtained. ab , covariance Cov(a, b) and sales volume correlation coefficient ρ b , calculate the sales impact factor α of related types of products b :

[0023] α b =ρ b *Cov(a,b)*Jx ab

[0024] The sales impact factor α of the corresponding related type of goods b The calculation formula is as above.

[0025] Furthermore, we obtain the sales data of all historical new products from the e-commerce platform, including the number of new product sales per unit time Sx i and sales amountJe i , combined with the sales quantity Xs of all products ij 、Sales amountXe ij , calculate the historical new product price sensitivity coefficient A:

[0026]

[0027] The corresponding calculation formula for the historical new product price sensitivity coefficient A is as above.

[0028] Furthermore, obtain the number of new product sales per unit time Sx i and the sales quantity Xs of all products ij , calculate the average historical new product sales share Zb:

[0029]

[0030] The calculation formula for the corresponding average historical new product sales proportion Zb is as above.

[0031] Furthermore, obtain the sales quantity Xs of the target product and the related type of products per unit time. bj The sales impact factor α of related products b , calculate the impact of related types of goods on sales volume Yux:

[0032]

[0033] The calculation formula for the impact of the corresponding related types of goods on sales volume Yux is as above.

[0034] Further, obtain the sales quantity Xs of the target product per unit time aj 、Sales amountXe aj , historical new product price sensitivity coefficient A, calculate the impact of price on sales volume Jys:

[0035]

[0036] The calculation formula for the corresponding price impact on sales volume Jys is as above.

[0037] Further, obtain the sales quantity Xs of the target product per unit time aj , the sales volume affected by related types of goods Yux, the sales volume affected by price Jys and the average proportion of historical new product sales Zb, calculate the estimated new product sales volume Xy of the target product:

[0038]

[0039] The calculation formula for the corresponding target product's new product estimated sales volume Xy is as above.

[0040] (III) Beneficial effects

[0041] The present invention provides a method for estimating commodity sales, which has the following beneficial effects:

[0042] 1. Obtain sales data of all commodities from the e-commerce platform and calculate the sales correlation coefficient ρ between the target commodity and other commodities of different types i , combined with the cross-selling rate Jx of the target product and related products ab And covariance Cov(a, b), calculate the sales impact factor α of related types of products b , which can reflect the impact of related types of goods on the sales volume of target goods, help identify potential product combinations or matching sales opportunities, and optimize the product recommendation system.

[0043] 2. Obtain the sales data of all historical new products from the e-commerce platform, calculate the historical new product price sensitivity coefficient A and the average proportion of historical new product sales Zb, so as to more accurately understand the impact of new product price changes on consumer purchasing behavior, and fully understand the market performance and sales of new products. It can also conduct more scientific evaluation and management of new products, discover problems in time and take corresponding measures to adjust and optimize.

[0044] 3. According to the target product, the sales quantity Xs of the corresponding type of products per unit time bj The sales impact factor α of related products b , calculate the sales volume Yux of related types of goods, and the sales volume Xs of the target goods per unit time aj 、Sales amountXe aj , the historical new product price sensitivity coefficient A, calculate the price impact on sales Jys, combined with the target product sales volume per unit time Xs aj , the impact of related types of goods on sales volume Yux, the impact of price on sales volume Jys and the average proportion of historical new product sales volume Zb, and the calculation of the estimated new product sales volume Xy of the target product can more comprehensively reflect the market situation, more accurately grasp the market demand, optimize inventory management, pricing strategy, marketing strategy and other aspects, thereby reducing market risks, improving customer satisfaction and loyalty, and ultimately achieving business growth and sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 The figure is a flow chart of a method for estimating commodity sales according to the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] See also Figure 1 The present invention provides a method for estimating commodity sales, comprising the following steps:

[0048] Step 1: Obtain sales data of all products from the e-commerce platform and calculate the sales correlation coefficient ρ between the target product and other different types of products i , combined with the cross-selling rate Jx of the target product and related products ab And covariance Cov(a, b), calculate the sales impact factor α of related types of products b .

[0049] The step 1 includes the following contents:

[0050] Step 101: Obtain sales data of all products from the e-commerce platform, including the sales quantity Xs per unit time. ij 、Sales amountXe ij , calculate the sales correlation coefficient ρ between the target product and other different types of products i :

[0051]

[0052] Among them, i represents the sequential number of the product type, i=1, 2, 3, 4, ..., n, n represents the total number of product types, j represents the sequential number of different brands under the same product type, j=1, 2, 3, 4, ..., m, m represents the total number of different brands under the same product type, and a represents the type number of the target product.

[0053] Step 102: Obtain the sales volume correlation coefficient ρ between the target product and other products of different types i , after sorting from largest to smallest, select the top 5 product types as related products of the target product.

[0054] Step 103: Obtain the cross-selling rate Jx between the target product and related products ab , and calculate the covariance Cov(a, b) between the target product and related products based on their sales quantity:

[0055]

[0056] Where b represents the serial number of related types of goods, b = 1, 2, 3, 4, 5. The cross-selling rate is the proportion of customers who purchased multiple products in a specific time period. Its calculation formula is as follows:

[0057] Cross-selling rate = (number of customers who purchased multiple products or services) / (total number of customers who purchased products or services) × 100%

[0058] For example, suppose that in a month, 100 customers purchased product A, and 30 of them also purchased product B. Then the cross-selling rate is 30 divided by 100, which is 0.3, or 30%. This means that in this month, 30% of customers purchased both product A and product B.

[0059] Step 104: Obtain the cross-selling rate Jx between the target product and related products ab , covariance Cov(a, b) and sales volume correlation coefficient ρ b , calculate the sales impact factor α of related types of products b :

[0060] α b =ρ b *Cov(a,b)*Jx ab

[0061] When using, combine the contents in steps 101 to 104:

[0062] Obtain sales data of all products from the e-commerce platform and calculate the sales correlation coefficient ρ between the target product and other different types of products i , combined with the cross-selling rate Jx of the target product and related products ab And covariance Cov(a, b), calculate the sales impact factor α of related types of products b , which can reflect the impact of related types of goods on the sales volume of target goods, help identify potential product combinations or matching sales opportunities, and optimize the product recommendation system.

[0063] Step 2: Obtain the sales data of all historical new products from the e-commerce platform, and calculate the price sensitivity coefficient A of the historical new products and the average sales share Zb of the historical new products.

[0064] The step 2 includes the following contents:

[0065] Step 201: Obtain all historical new product sales data from the e-commerce platform, including the number of new product sales per unit time Sx i and sales amountJe i , combined with the sales quantity Xs of all products ij 、Sales amountXe ij, calculate the historical new product price sensitivity coefficient A:

[0066]

[0067] The corresponding calculation formula for the historical new product price sensitivity coefficient A is as above.

[0068] Step 202: Obtain the number of new product sales per unit time Sx i and the sales quantity Xs of all products ij , calculate the average historical new product sales share Zb:

[0069]

[0070] The calculation formula for the corresponding average historical new product sales proportion Zb is as above.

[0071] When using, combine the contents in steps 201 and 202:

[0072] By obtaining the sales data of all historical new products from the e-commerce platform and calculating the historical new product price sensitivity coefficient A and the average proportion of historical new product sales Zb, we can more accurately understand the impact of new product price changes on consumer purchasing behavior, and fully understand the market performance and sales of new products. We can evaluate and manage new products more scientifically, discover problems in a timely manner, and take corresponding measures to adjust and optimize them.

[0073] Step 3: Calculate the sales volume Xs of the corresponding type of products in a unit time according to the target product bj The sales impact factor α of related products b , calculate the sales volume Yux of related types of goods, and the sales volume Xs of the target goods per unit time aj 、Sales amountXe aj , the historical new product price sensitivity coefficient A, calculate the price impact on sales Jys, combined with the target product sales volume per unit time Xs aj , the impact of related types of goods on sales volume Yux, the impact of price on sales volume Jys and the average proportion of historical new product sales volume Zb, calculate the estimated new product sales volume Xy of the target product.

[0074] The step three includes the following contents:

[0075] Step 301: Obtain the sales quantity Xs of the target product and the related type of product per unit time bj The sales impact factor α of related products b , calculate the impact of related types of goods on sales volume Yux:

[0076]

[0077] The calculation formula for the impact of the corresponding related types of goods on sales volume Yux is as above.

[0078] Step 302: Obtain the sales quantity Xs of the target product per unit time aj 、Sales amountXe aj , historical new product price sensitivity coefficient A, calculate the impact of price on sales volume Jys:

[0079]

[0080] The calculation formula for the corresponding price impact on sales volume Jys is as above.

[0081] Step 303: Obtain the sales quantity Xs of the target product per unit time aj , the sales volume affected by related types of goods Yux, the sales volume affected by price Jys and the average proportion of historical new product sales Zb, calculate the estimated new product sales volume Xy of the target product:

[0082]

[0083] The calculation formula for the corresponding target product's new product estimated sales volume Xy is as above.

[0084] When using, combine the contents in steps 301 to 303:

[0085] According to the sales quantity Xs of the target product corresponding to the relevant type of product per unit time bj The sales impact factor α of related products b , calculate the sales volume Yux of related types of goods, and the sales volume Xs of the target goods per unit time aj 、Sales amountXe aj , the historical new product price sensitivity coefficient A, calculate the price impact on sales Jys, combined with the target product sales volume per unit time Xs aj , the impact of related types of goods on sales volume Yux, the impact of price on sales volume Jys and the average proportion of historical new product sales volume Zb, and the calculation of the estimated new product sales volume Xy of the target product can more comprehensively reflect the market situation, more accurately grasp the market demand, optimize inventory management, pricing strategy, marketing strategy and other aspects, thereby reducing market risks, improving customer satisfaction and loyalty, and ultimately achieving business growth and sustainable development.

[0086] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0087] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0088] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for estimating commodity sales, characterized by: The steps include: Obtain sales data of all products from the e-commerce platform and calculate the sales correlation coefficient ρ between the target product and other different types of products i , combined with the cross-selling rate Jx of the target product and related products ab And covariance Cov(a, b), calculate the sales impact factor α of related types of products b ; Obtain sales data of all historical new products from the e-commerce platform, and calculate the price sensitivity coefficient A of historical new products and the average sales volume of historical new products Zb; According to the sales quantity Xs of the target product corresponding to the relevant type of product per unit time bj The sales impact factor α of related products b , calculate the sales volume Yux of related types of goods, and the sales volume Xs of the target goods per unit time aj 、Sales amountXe aj , the historical new product price sensitivity coefficient A, calculate the price impact on sales Jys, combined with the target product sales volume per unit time Xs aj , the impact of related types of goods on sales volume Yux, the impact of price on sales volume Jys and the average proportion of historical new product sales volume Zb, calculate the estimated new product sales volume Xy of the target product.

2. A method for estimating commodity sales according to claim 1, characterized in that: Get sales data of all products from the e-commerce platform, including the sales quantity Xs per unit time ij 、Sales amountXe ij , calculate the sales correlation coefficient ρ between the target product and other different types of products i : Among them, i represents the sequential number of the product type, i=1, 2, 3, 4, ..., n, n represents the total number of product types, j represents the sequential number of different brands under the same product type, j=1, 2, 3, 4, ..., m, m represents the total number of different brands under the same product type, and a represents the type number of the target product.

3. A method for estimating commodity sales according to claim 2, characterized in that: Get the sales correlation coefficient ρ between the target product and other different types of products i , after sorting from largest to smallest, select the top 5 product types as related products of the target product.

4. A method for estimating commodity sales according to claim 3, characterized in that: Get the cross-selling rate Jx of the target product and related products ab , and calculate the covariance Cov(a, b) between the target product and related products based on their sales quantity: Among them, b represents the sequential number of the related type of goods, b=1, 2, 3, 4, 5.

5. A method for estimating commodity sales according to claim 4, characterized in that: Get the cross-selling rate Jx of the target product and related products ab , covariance Cov(a, b) and sales volume correlation coefficient ρ b , calculate the sales impact factor α of related types of products b : α b =ρ b *Cov(a,b)*Jx ab The sales impact factor α of the corresponding related type of goods b The calculation formula is as above.

6. A method for estimating commodity sales according to claim 1, characterized in that: Get all historical new product sales data from the e-commerce platform, including the number of new product sales per unit time Sx i and sales amountJe i , combined with the sales quantity Xs of all products ij 、Sales amountXe ij , calculate the historical new product price sensitivity coefficient A: The corresponding calculation formula for the historical new product price sensitivity coefficient A is as above.

7. A method for estimating commodity sales according to claim 1, characterized in that: Get the number of new product sales per unit time Sx i and the sales quantity Xs of all products ij , calculate the average historical new product sales share Zb: The calculation formula for the corresponding average historical new product sales proportion Zb is as above.

8. A method for estimating commodity sales according to claim 5, characterized in that: Get the sales quantity Xs of the target product and the corresponding type of product per unit time bj The sales impact factor α of related products b , calculate the impact of related types of goods on sales volume Yux: The calculation formula for the impact of the corresponding related types of goods on sales volume Yux is as above.

9. A method for estimating commodity sales according to claim 6, characterized in that: Get the sales quantity Xs of the target product per unit time aj 、Sales amountXe aj , historical new product price sensitivity coefficient A, calculate the impact of price on sales volume Jys: The calculation formula for the corresponding price impact on sales volume Jys is as above.

10. A method for estimating commodity sales according to claim 9, characterized in that: Get the sales quantity Xs of the target product per unit time aj , the sales volume affected by related types of goods Yux, the sales volume affected by price Jys and the average proportion of historical new product sales Zb, calculate the estimated new product sales volume Xy of the target product: The calculation formula for the corresponding target product's new product estimated sales volume Xy is as above.

Images