TWI652638B - Smart marketing system and method thereof - Google Patents

Abstract

The present invention provides a smart marketing system and method therefor. The personnel enters the detection range of the merchandise shelf; records the behavior of the shopping process; if the personnel enters the detection range of other merchandise shelves, the shopping tendency analysis of the target merchandise is performed; if the personnel enters the detection range of other classified merchandise shelves, the classification is performed. Analysis of the shopping tendency of the product; if the comprehensive judgment personnel meet the promotion conditions, send a promotional message; record the consumption and promotion effectiveness after the personnel checkout. In this way, you can instantly analyze the shopping behavior of the customer's current purchase, and provide promotional information immediately when the customer is still in the store to enhance the purchase intention.

Description

Smart marketing system and method thereof

The invention relates to an intelligent analysis technology, and in particular to a smart marketing system based on intelligent analysis and a method thereof.

In order to influence the consumer's desire to buy, the merchants are all surprisingly recruiting to promote their own goods. With the advancement of the times, the computing power of computers is fast, and the processing of big data is not difficult. Of course, the analysis of purchasing desires can be realized. Therefore, most merchants have introduced intelligent marketing methods of intelligent analysis. The application of smart marketing is quite extensive, for example, improving the placement of products to increase sales, statistical analysis of customer spending records to recommend products. However, in the existing smart marketing technology, merchants can only understand the customer's needs through post-mortem analysis in order to carry out subsequent marketing. When the customer is still shopping in the mall, the merchant cannot provide the appropriate promotion immediately, so the customer's desire to purchase can not be effectively improved.

In view of this, the present invention provides a smart marketing system and a method thereof, which can instantly determine the shopping tendency of the in-store customers and provide suitable promotional information, thereby enhancing the customer's desire for consumption.

The smart marketing method of the present invention comprises the following steps. Judging personnel entering the merchandise shelf through at least one sensor. The person's shopping behavior for the goods on the shelf of the goods is recorded through the sensor. According to the shopping behavior of this person, the purchasing tendency of the person is determined. Send a promotional message based on the person’s shopping preferences.

The intelligent marketing system of the present invention includes at least one sensor and an arithmetic device. These sensors determine that the person enters the merchandise shelf. The computing device records the purchasing behavior of the goods on the shelf of the product through the sensor, determines the shopping tendency of the person according to the shopping behavior of the person, and sends a promotional message according to the shopping tendency of the person.

Based on the above, the embodiment of the present invention records the shopping behavior of the person by a sensor disposed around or on the shelf of the product, and analyzes the shopping behavior immediately, and then provides appropriate promotional content. In this way, the opportunity for customers to purchase goods will increase.

The above described features and advantages of the invention will be apparent from the following description.

1 is a schematic diagram of a smart marketing system 1 in accordance with an embodiment of the present invention. Referring to FIG. 1, the smart marketing system 1 includes one or more sensors 110 and an arithmetic device 120 in an environment in which commodity shelves PR1 to PRm (m is a positive integer) are provided (for example, a hypermarket, a retail of nursing products) Shop, toy store, etc.).

The sensor 110 can be a Wi-Fi Access Point (AP), a Beacon, a monitor, a Radio Frequency Identification (RFID) reader, an electronic shelf, or the like, which senses the behavior of the object. It may be a combination of the foregoing embodiments. Each sensor 110 is used for the mobile device around the goods shelf PR1~PRm or the mobile devices around the product shelves PR1~PRm (for example, the communication technology corresponding to the sensor 110 is loaded (for example, Wi-Fi, Bluetooth, etc.) (smart phones, tablets, wearable devices and other electronic devices) or personnel to sense. For example, recording a specific range of images, detecting that a particular item has been removed/replaced, detecting whether there is a mobile device or person within a specific range.

The computing device 120 can be a device such as a computer host, a server, a tablet computer, a workstation, or the like having an arithmetic processor (for example, a CPU, a microcontroller, a chip, etc.), and can obtain sensing data from the sensor 110, and The sensing data is analyzed, and the analysis process and results are described in the following examples.

In order to facilitate the understanding of the operational flow of the present invention, various embodiments will be described in detail below. 2 is a flow chart showing a smart marketing method according to an embodiment of the invention. Referring to FIG. 2, in the following, the methods and embodiments of the present invention will be described in conjunction with the components and devices of the smart marketing system 1. The various processes of the method can be adjusted accordingly according to the implementation situation, and are not limited thereto.

The computing device 120 determines whether the person enters the detection range of any one of the product shelves PR1 to PRn through the sensor 110. For example, whether the signal intensity of the person or the person's mobile device appears in the image screen is greater than a predetermined intensity. In this embodiment, when the triggering condition that the detection result is that the person has entered any of the product shelves PR1, . . . or PRm occurs (step S201), the computing device 120 further records the shopping behavior of the person through the sensor 110 ( Step S202). Assume that this person enters the detection range of the product shelf PR1.

The shopping behavior recorded by the computing device 120 may be the take/replace behavior of the target item, the viewing time of the target item and the characteristics of the person (eg, gender, age, or combination), and the stay on the product shelf PR1. time. For example, the computing device 120 can recognize the image acquired by the sensor 110 to know that the person takes or puts back a certain item of merchandise, and records the interval between taking or returning the same item as the viewing time. The computing device 120 can obtain the gender and age information about the person sent by the mobile device carried by the person. The computing device 120 can determine whether it is within 1 meter of the commodity shelf PR1 through the size of the signal sent by the mobile device carried by the person, and record the time of entering and leaving the range to obtain the staying time. Depending on the type of sensor 110, the manner in which the shopping behavior is recorded and analyzed may be different, depending on the needs of those applying the embodiments of the present invention.

The computing device 120 determines through the sensor 110 whether the person or the mobile device carried by the user enters the detection range of other commodity shelves (ie, different from other commodity shelves monitored in step S202) PR2, . . . or PRm (step S203). . For example, the sensor 110 for the product shelf PR2 detects that the person enters the detection range of the product shelf PR2, or the signal strength of the mobile device carried by the person is less than a specific threshold value. If the other product shelves PR2, ... or PRm are not entered, the person's shopping behavior for the product shelf PR1 is continuously recorded. If the other product shelves have been entered (assuming entering the product shelf PR2), the computing device 120 performs shopping tendency analysis on the previous target products (step S204).

Specifically, the shopping tendency analysis of the embodiment of the present invention mainly combines the personal information of the person with the past shopping record according to the shopping behavior obtained in step S202 to obtain the shopping tendency of the person to the target product. Please refer to FIG. 3, which is a flow chart for calculating a shopping tendency of a target item according to an embodiment of the present invention. The computing device 120 first extracts the shopping behavior record related to the target item (step S301), and judges from the record of the shopping behavior of the person whether the target item is taken out/replaced from the product shelf PR1 (step S302). If so, the arithmetic unit 120 calculates the shopping tendency score A corresponding to the number of times of take-out/replacement (step S303). For example, the shopping tendency score A corresponding to the number of take/replace times is obtained by the formula (1): (1) If the number of take-out/replacement n is greater than 5, the shopping propensity score A is 1 and vice versa is n /5. It should be noted that this formula (1) can be adjusted according to requirements, but in principle, the lower the shopping tendency score A, the lower the number of take-out/return times.

The computing device 120 can also obtain the viewing time of each time the target item is taken out by the record of the shopping behavior when the person has taken out/replaced the target item, and summons the total viewing time of the person to take out the total item for the target item. And calculating the shopping tendency score B corresponding to the total viewing time of the target item (step S304). For example, from formula (2), the corresponding shopping propensity score B corresponding to the viewing time is obtained: (2) If the total viewing time i is greater than 5 minutes, the shopping propensity score B is 1 and vice versa is i /5. It should be noted that this formula (2) can be adjusted according to requirements, but in principle, the lower the total viewing time, the lower the shopping tendency score B.

The arithmetic unit 120 also calculates the shopping tendency score C corresponding to the person's stay time before the target item or the product shelf PR1 (step S305). For example, from formula (3), the corresponding shopping propensity score C corresponding to the dwell time is obtained: (3) If the stay time j is greater than 10 minutes, the shopping tendency score C is 1 and vice versa is j /10. It should be noted that this formula (3) can be adjusted according to requirements, but in principle, the lower the staying time, the lower the shopping tendency score C.

The computing device 120 further extracts the correlation between each product and the target product from the product records previously consumed by the customer, and calculates the shopping tendency score D corresponding to the previous purchase of the product (ie, the previous shopping tendency) (step S306. If the result of the determination in step S302 is that the behavior of the target item is not taken out/replaced, the step is directly performed). For example, it is defined as the highest value of the current target product (eg, product category, brand, etc.) among all the products purchased in the previous purchase, and the correlation between the products may be a previously defined value. D is a defined value between 0 and 1, ie .

Further, the arithmetic unit 120 calculates the correlation ( E , F ) of the gender, age (i.e., characteristic association) of the customer with the target item (step S307). For example, the correlation between gender and target goods, age and target goods E and F are respectively between 0 and 1, ie , .

Next, the computing device 120 respectively assigns the weighting value of the number of taking/returning behaviors, the viewing time, the staying time, the previous shopping tendency, and the characteristic association, and performs a weighting calculation to calculate the purchasing propensity score of the person for the target item. T1 (step S308), and this shopping tendency score T1 is taken as the shopping tendency of the target item. For example, the shopping tendency score T1 of the target item is obtained by the formula (4): ...(4) α, β, γ, δ, θ, μ are the weight values corresponding to each shopping behavior.

Referring back to FIG. 2, the computing device 120 determines whether the client enters the detection range of the other classified merchandise shelf (the merchandise category is different from the merchandise shelf monitored in step S202) PR2, . . . or PRm (step S203). If the detection range of the other classified product shelves PR2, . . . or PRm is entered, the computing device 120 performs shopping tendency analysis on the previously classified products (step S204).

Specifically, the analysis of the classified products is mainly based on the shopping behavior of the monitoring, and the combination of the specific shopping trends of the classified goods (ie, belonging to the same product category), the personal information of the customer, and the past shopping records. Calculate the customer's shopping propensity for this classified item. Referring to FIG. 4, a flow chart for calculating a shopping tendency of classified merchandise according to an embodiment of the present invention is shown. The computing device 120 calculates the shopping tendency of the classified products corresponding to the shopping tendency of each shelf product (that is, various products in the classified product), and averages the classified product shopping tendency score G to avoid the shopping tendency of the single target product. The shopping tendency of the overall classified products. For example, from the formula (5), the classification product shopping tendency score G of each shelf item is obtained: (5) That is, the average of the shopping tendency scores A _k of the specific products of the classified products, assuming that the classified products have a total of k specific commodities.

Next, the arithmetic unit 120 calculates the shopping tendency score H corresponding to the total stay time of the classified item (refer to step S305). When a person's shopping propensity scores for various specific products are not high (for example, less than 0.2, 0.3, etc.), but there is a higher total retention time of the classified goods (for example, more than 10, 15 minutes, etc.), it can represent people. The classified goods for this product classification are still relatively high in purchasing intention. For example, the shopping tendency score H of the total residence time of the classified goods is obtained by the formula (6): (6) If the total stay time l is greater than 20 minutes, the shopping propensity score H is 1 and vice versa is l / 20.

Next, the arithmetic unit 120 calculates the classified item shopping tendency score T2 corresponding to the item type based on the average value of the classified items and the total staying time (step S403). Since the information such as the gender, age, and other characteristics of the person and the relevance of the previous purchase have been used to calculate the shopping tendency G of each shelf item, the calculation of the classification product shopping tendency score T2 can no longer be repeated. Value. For example, from the formula (7), the classified product shopping tendency score T2 is obtained : ...(7) ζ, η are the weighting values of the shopping propensity scores G and H.

Referring back to FIG. 2, the computing device 120 can then determine whether the person meets the promotion condition (step S207). The promotion condition can be the location of the person, whether the relevant item has been purchased, the shopping tendency analyzed by the foregoing part, and the past promotion effect. . Please refer to FIG. 5, which is a flowchart for determining whether a promotion condition is met according to an embodiment of the present invention. The computing device 120 can instantly obtain the current location of the person through the sensor 110 (step S501). For example, it is determined that the sensor 110 for a certain product shelf PR1, . . . or PRm captures the person within the detection range, or determines that the mobile device carried by the person detected in the sensor 110 is issued. The signal strength is the strongest. The computing device 120 can determine whether the location of the person exceeds a certain distance (for example, 10, 15 meters, etc.) of a certain sensor 110, whether to go to another floor, whether to go to checkout, etc., to determine whether the position of the person meets the sending condition. (Step S502). If the position of the person is still within the judgment condition (for example, within 10 meters, on the same floor, etc.), the arithmetic unit 120 determines that the transmission condition is met.

If the position of the person meets the transmission condition, the computing device 120 determines whether the person has a higher shopping tendency for the specific item in the former classified item (for example, the shopping tendency score T1 exceeds a specific value) to confirm the specific item of the former classified item. Whether the shopping tendency conforms to the transmission condition (step S503) (for example, if the specific value is exceeded). If it does, you don't need to judge whether there is a shopping tendency for this classified product, so as to avoid sending the same or similar promotional messages repeatedly. If not, the computing device 120 determines whether the customer has a higher shopping tendency for the former classified product (for example, the shopping tendency score T2 exceeds a preset value) to determine whether the shopping tendency of the former classified product meets the sending condition (step S504) (for example, if the preset value is exceeded).

Next, the computing device 120 determines whether the received promotional message (ie, the number of received) exceeds the transmission threshold (eg, 3, 5 times, etc.) (step S505) to avoid sending too many promotional messages to cause a counter effect. The computing device 120 also determines whether the promotional effect of the person (ie, the number of successful sales successfully sold through the promotional message) is greater than the performance threshold (step S506) to avoid sending an ineffective promotional message. Finally, the computing device 120 integrates the various determinations of the foregoing steps to determine whether the person meets the promotion condition (for example, if the location of the person is on the same floor, the shopping tendency for a particular product is lower than a specific value, and the shopping tendency for the classified product is higher than If the preset value, the received quantity is greater than the sending threshold, and the successful number is greater than the performance threshold, etc., the promotion condition is met; otherwise, the promotion condition is not met) (step S507).

Please return to FIG. 2, if the promotion condition is met, the computing device 120 may use the short message, the push message or the mobile device application notification to send a promotional message related to the target product or the classified product (for example, the preferential price of the target product, the classified product) The discount code, etc.) is sent to the mobile device of the person for reference by the person or the discount code in the promotional message at the time of checkout (step S208). Then, after the person completes the checkout, the computing device 120 connects with the checkout system and records the content of the current purchase, and the effect of the promotion (whether or not the product recommended by the promotion message is purchased) (step S209), as a judgment of the person The promotion effect, the promotion effect of various commodities, etc., and improve the content of the promotion or the adjustment of the above calculation formulas (1) to (7).

In summary, the embodiments of the present invention utilize various shopping behaviors of a person in an in-store shopping process to analyze the shopping tendency of a specific product or a classified product, and then determine whether to meet the promotion condition according to the location thereof, and according to the promotion. The judgment result of the condition sends a promotional message to the person, and the person records the promotion effect after the checkout, thereby providing accurate marketing content. In this way, the merchant can instantly analyze the classification of the products that he wants to purchase at present through various behavioral information of the user in the in-store shopping process. In addition, the merchant can promptly send the promotional message of the product to be purchased when the customer is still in the store, thereby increasing the willingness of the customer to consume.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention, and any one of ordinary skill in the art can make some changes and refinements without departing from the spirit and scope of the present invention. The scope of the invention is defined by the scope of the appended claims.

100‧‧‧Smart Marketing System

110‧‧‧ Sensor

120‧‧‧ arithmetic device

PR1~PRm‧‧‧Commodity shelves

S201~S209, S301~S308, S401~S403, S501~S507‧‧‧ steps

1 is a schematic diagram of a smart marketing system in accordance with an embodiment of the present invention. 2 is a flow chart of a smart marketing method in accordance with an embodiment of the present invention. 3 is a flow chart for calculating a shopping tendency of a target item according to an embodiment of the present invention. 4 is a flow chart for calculating a shopping tendency of a classified item according to an embodiment of the present invention. FIG. 5 is a flow chart for determining whether a promotion condition is met in accordance with an embodiment of the present invention.

Claims (8)

A smart marketing method includes: determining, by at least one sensor, a person entering a merchandise shelf; recording, by the at least one sensor, the purchasing behavior of the person on the merchandise shelf of the merchandise; determining the person according to the purchasing behavior of the merchanding a shopping tendency; and sending a promotional message based on the purchasing tendency of the person, wherein the shopping behavior includes a take-to-action of a target item, a viewing time and a characteristic association, and a stay time of the shelf of the item, the characteristic association including gender The age, or a combination thereof, and the step of determining the shopping tendency of the person according to the shopping behavior of the person includes: determining whether the shopping behavior has the taking behavior of the target product; calculating the number of times the taking behavior is performed; The viewing time of the target product; calculating the stay time of the shelf of the product; obtaining the previous shopping tendency of the person who has the highest relevance to the target product in the previous purchase; obtaining the characteristic association of the person; and the number of times the action is given respectively , the viewing time, the dwell time, the previous shopping tendency and characteristics The weight value and weighted to determine the person's shopping for the target product propensity score. The smart marketing method according to claim 1, wherein the step of determining the shopping tendency of the person according to the shopping behavior of the person comprises: calculating an average shopping tendency of the plurality of classified products on the shelf of the product and other products. a value, wherein the classified commodities belong to the same commodity category as the target commodity; calculate a total residence time of the classified commodities; and calculate a shopping propensity score of the commodity category according to the average value of the classified commodities and the total staying time. The smart marketing method of claim 1, wherein the step of transmitting the promotional message according to the shopping tendency of the person comprises: determining, by the at least one sensor, the location of the person; and determining whether the location is based on the location of the person Send the promotion message. The smart marketing method of claim 1, wherein the step of sending the promotional message according to the shopping tendency of the person comprises: counting the number of received promotional messages received by the person; and counting the success of the person passing the promotional message The number of successful sales; and whether to send the promotional message based on the received quantity and the number of successes. A smart marketing system comprising: at least one sensor for determining a person entering a merchandise shelf; and an computing device for recording, by the at least one sensor, the purchasing behavior of the person on the merchandise shelf of the merchandise, according to the personnel The shopping behavior determines the purchasing tendency of the person, and sends a promotional message according to the purchasing tendency of the person. The shopping behavior includes an act of taking a target item, a viewing time and a characteristic association, and a stay time of the shelf of the item, the characteristic association including a gender, an age, or a combination thereof, and the computing device determines whether the shopping behavior is The act of taking the target product, calculating the number of times of taking the action, calculating the viewing time of the target product, calculating the stay time of the shelf of the product, and obtaining the highest correlation with the target product in the previous purchased product. The previous shopping tendency, obtaining the characteristic association of the person, and assigning the weighting value of the number of times of taking the action, the viewing time, the staying time, the previous shopping tendency and the characteristic, and performing weighting calculation to determine the person The shopping propensity score for the target item. The smart marketing system according to claim 5, wherein the computing device calculates an average of shopping trends of the plurality of classified products on the shelf of the commodity and other commodities, wherein the classified commodities are the same as the target commodity A commodity category, calculating a total residence time of the classified commodities, and calculating a shopping propensity score of the commodity category according to the average value of the classified commodities and the total staying time. The smart marketing system of claim 5, wherein the computing device determines the location of the person through the at least one sensor, and determines whether to send the promotional message according to the location of the person. The smart marketing system of claim 5, wherein the computing device counts the number of received promotional messages received by the person, and counts the number of successful sales of the person through the promotional message, and according to the received quantity and The number of successes determines whether or not to send the promotional message.