Automatic right and interest matching method based on member level in B2B2C platform

Abstract

The invention belongs to the technical field of member management, and particularly relates to an automatic right and interest matching method based on member levels in a B2B2C platform, which comprises the following steps of: acquiring consumption sum, consumption frequency and proximity three-dimensional behavior data of a user; calculating a user value score based on an industry feature weight factor preset by the platform; dynamically dividing member grades according to the value scores; and calling a right rule base, mapping the level to a right package, and pushing the right package to a user terminal in real time. According to the method, a formula capable of truly reflecting the value of a user is constructed by using the consumption sum, consumption frequency and proximity three-dimensional behavior data of the user, the value of the user is quantized by inputting the consumption sum, consumption frequency and proximity of the user into the formula, and then the value of the user is calculated according to the quantized value of the user. And a peer-to-peer or mapping relation between consumption amount and member points in the prior art is canceled, so that the underestimated value of the underestimated high-frequency low-consumption user can be accurately identified and prevented from being underestimated.

Description

Technical Field

[0001] This invention belongs to the field of membership management technology, specifically a method for automatically matching membership benefits based on membership level in a B2B2C platform. Background Technology

[0002] In traditional membership level / benefits matching systems, the accumulation of membership points is generally mapped to the accumulation of user spending. It can be understood that the higher the accumulated spending of any user, the higher the user's membership points. In different systems, spending and points may be equivalent, that is, a user spends one yuan and gets one point. However, in this case, the value of users who frequently offset spending is underestimated. B2B2C platforms, including the relationships between the platform and suppliers, and between the platform and users, can essentially be viewed as relationships between the platform and different types of users. Taking the relationship between the platform and users as an example, assuming a user makes a purchase on the platform, according to the traditional membership level / benefits matching system, the higher the user's purchase amount, the more points they accumulate. The number of points corresponds to different membership levels and the benefits associated with each membership level. Simply put, the higher a user's purchases, the higher their membership level and the more benefits they enjoy. However, in reality, user consumption behavior can be categorized into high-frequency low-spending, high-frequency high-spending, low-frequency high-spending, and low-frequency low-spending. High-frequency high-spending and low-frequency low-spending are easy to understand, and platforms can easily distinguish between these two types of users. However, high-frequency low-spending and low-frequency high-spending are difficult to distinguish, especially since high-frequency low-spending users have a higher user value to the platform than low-frequency high-spending users. However, based on the current membership level / benefits matching rules, this type of user is underestimated, resulting in an imbalance of benefits.

[0003] Therefore, this invention provides a method for automatically matching benefits based on membership level in a B2B2C platform. Summary of the Invention

[0004] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0005] The technical solution adopted by this invention to solve its technical problem is: a method for automatically matching benefits based on membership level in a B2B2C platform, comprising the following steps: S1: Collect user spending amount Consumption frequency Proximateity Three-dimensional behavioral data; S2: Industry-specific weighting factors based on platform presets Calculate user value score ; S3: Based on value score Dynamically classify membership levels ; S4: Call the rights and interests rule base and assign levels Mapped to a benefits package and pushed to the user's terminal in real time.

[0006] Preferably, in step S2, a user value score is calculated. The method is as follows: retrieve user's spending amount Consumption frequency Proximateity Three-dimensional behavioral data; the amount of consumption Consumption frequency Based on rolling time window calculate; According to the formula: ; in, This is the highest spending amount in the platform's history. This represents the highest consumption frequency in the platform's history. .

[0007] Preferably, in step S1, the consumption frequency is obtained. Previously, this also included merging and detecting user orders, including: If the same user If multiple orders are placed consecutively within an hour, they will be combined into a single transaction, and , ; At the same time, the user is marked as a risky account, triggering... Weight downgrade: According to the formula: ; ; ; ; in, for The penalty factor for the weight.

[0008] Preferably, in step S2, the industry characteristic weighting factor The method for obtaining it is as follows: Analysis of high-frequency user retention rates in historical data of the platform and low-frequency user churn rate ; when Start Dynamic adjustment; among which, For threshold; According to the formula: ; in, To adjust the coefficient, and .

[0009] Preferably, an acceleration benefit channel is opened for the aforementioned high-frequency, low-consumption users, and the method is as follows: Daily consumption frequency At that time, temporary rights are triggered; among them, This is the threshold for daily consumption frequency. The temporary rights include: Temporarily upgrade membership level to , And the validity period of the rights is ,in, The interval between the next purchase, in units of .

[0010] Preferably, in S1, the proximity factor The calculation method is as follows: Prominence calculated based on the time decay model ; According to the formula: ; in, For consumption time intervals, This is the industry attenuation factor.

[0011] Preferably, in step S3, based on the value score Dynamically classify membership levels The method is as follows: S31: Rolling cumulative historical value; According to the formula: ,in, For the first A rolling time window Internally calculated value score, This represents the cumulative number of windows. S32: Level Dynamic Threshold Matching Preset membership level threshold range Corresponding level to ; S33: Acquire Cumulative Historical Value and identify The level corresponding to the threshold range of belonging This refers to the user's membership level.

[0012] Preferably, in step S4, the rights and interests rule base stores multi-level membership levels. The corresponding benefits package matrix is ​​based on the user's current level. It calls the corresponding set of rights items in the rights package matrix and pushes the set of rights items to the user terminal.

[0013] The beneficial effects of this invention are as follows: This invention discloses an automatic benefit matching method based on membership level in a B2B2C platform, utilizing user spending amounts. Consumption frequency Proximateity Using three-dimensional behavioral data, a formula can be constructed to truly reflect user value, based on the user's spending amount. Consumption frequency Proximateity By inputting this formula, user value is quantified. Based on the quantified user value, the existing technology's equivalence or mapping relationship between consumption amount and membership points is eliminated, thereby accurately identifying and avoiding underestimation of the value of high-frequency, low-spending users. Attached Figure Description

[0014] The invention will now be further described with reference to the accompanying drawings.

[0015] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0016] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0017] like Figure 1 As shown in the embodiment of the present invention, an automatic matching method for benefits based on membership level in a B2B2C platform includes the following steps: S1: Collect user spending amount Consumption frequency Proximateity Three-dimensional behavioral data; S2: Industry-specific weighting factors based on platform presets Calculate user value score ; S3: Based on value score Dynamically classify membership levels ; S4: Call the rights and interests rule base and assign levels Mapped to a benefits package and pushed to the user's terminal in real time.

[0018] B2B2C platforms, including the relationships between the platform and suppliers, and between the platform and users, can essentially be viewed as relationships between the platform and different types of users. Taking the relationship between the platform and users as an example, assuming a user makes a purchase on the platform, according to the traditional membership level / benefits matching system, the higher the user's purchase amount, the more points they accumulate. The number of points corresponds to different membership levels and the benefits associated with each membership level. Simply put, the higher a user's purchases, the higher their membership level and the more benefits they enjoy. However, in reality, user consumption behavior can be categorized into high-frequency low-spending, high-frequency high-spending, low-frequency high-spending, and low-frequency low-spending. High-frequency high-spending and low-frequency low-spending are easy to understand, and platforms can easily distinguish between these two types of users. However, high-frequency low-spending and low-frequency high-spending are difficult to distinguish, especially since high-frequency low-spending users have a higher user value to the platform than low-frequency high-spending users. However, based on the current membership level / benefits matching rules, this type of user is underestimated, resulting in an imbalance of benefits.

[0019] To address the above problems, in one embodiment of the present invention, the user's consumption amount is utilized. Consumption frequency Proximateity Using three-dimensional behavioral data, a formula can be constructed to truly reflect user value, based on the user's spending amount. Consumption frequency Proximateity By inputting this formula, user value is quantified. Based on this quantified user value, the existing technology's equivalence or mapping relationship between consumption amount and membership points is eliminated, thereby accurately identifying and avoiding the underestimation of the value of high-frequency, low-spending users. The following specific examples will be used for explanation: Assuming a user A makes 5 purchases in a month on a certain beauty platform, then The average amount spent per transaction is 200 yuan, and the cumulative amount spent is... And if the most recent purchase was made 3 days ago, then If user B makes a purchase once a month, then... If a single purchase is 5000 yuan, then If the most recent purchase was made 30 days ago, then... ; According to the system's preset weighting industry characteristic weighting factors ,in , , ; Based on the calculation, the user value score corresponding to user A is... User B's corresponding user value score ; Based on the aforementioned calculations of user value scores for users A and B, data analysis shows that user A's user value score is significantly higher than user B's. Therefore, it can be understood that user A's user value to the platform is much higher than user B's. In existing technologies, directly mapping user spending amount to membership points would mistakenly assign user B a higher membership level, allowing them to enjoy more benefits, while assigning user A a lower membership level, resulting in significantly reduced benefits. This embodiment, however, can identify the impact of user spending frequency on user value, accurately identify user value, and scientifically link user value with membership level, preventing high-frequency, low-spending users from being underestimated.

[0020] In one embodiment, in step S2, a user value score is calculated. The method is as follows: retrieve user's spending amount Consumption frequency Proximateity Three-dimensional behavioral data; the amount of consumption Consumption frequency Based on rolling time window calculate; According to the formula: ; in, This is the highest spending amount in the platform's history. This represents the highest consumption frequency in the platform's history. .

[0021] The formula will now be explained using specific data: The time window here In other words, in this embodiment, a rolling time window is defined as 30 days, and the total value score for each rolling time window is calculated based on the accumulated value score generated within each rolling time window. This total value score is then summed with the previously accumulated value score to obtain the value score corresponding to the user. Assume user A is a high-frequency, low-consumption user: and within the time window The following are the consumer behaviors within the company: Only one purchase per day, and the total number of purchases is 15. ; If you spend 200 yuan each time, then ; If the most recent purchase was today, then ; User B is a low-frequency, high-consumption user, and within the time window... The following are the consumer behaviors within the company: Only one purchase, i.e. ; If the consumption amount is 5000 yuan, then ; If the most recent purchase was made 29 days ago, then ; Calculate the value scores for User A and User B using the formulas above: ; ; Based on the above calculations, user A, who has a high frequency but low consumption type, is within this time window. The value score accumulated by domestic consumption is Similarly, user B, who has a low-frequency, high-consumption type, during this time window... The value score accumulated by domestic consumption is ; Based on the above, assume user B is within the time window Additional actions, such as user B making a purchase of 2000 yuan on day 25, then... , , ; Recalculate: ; From the above, we can see that if user B is within the time window... Even if user B only makes one purchase, the amount of user B's purchase... This is the amount spent by user A. It is 1.6 times that, but based on the calculation of the value score above, we can conclude that... , From a data perspective, user A's value score for high-frequency consumption is significantly higher than user B's. This means that in one embodiment of the invention, greater emphasis is placed on the value of user consumption behavior. Compared to low-frequency consumers, the system favors high-frequency consumers, and it does not underestimate the value of high-frequency, low-consumption users; on the contrary, it inflates their value. Based on the above, when user B is within the time window... If memory is appended, user B's value score increases; It is worth noting that, among them, the rolling time window within the consumption amount For rolling time windows The sum of all order amounts, corresponding to the purchase frequency. Also a rolling time window The total number of all orders.

[0022] In one embodiment, in step S1, the consumption frequency is obtained. Previously, this also included merging and detecting user orders, including: If the same user If multiple orders are placed consecutively within an hour, they will be combined into a single transaction, and , ; At the same time, the user is marked as a risky account, triggering... Weight downgrade: According to the formula: ; ; ; ; in, for The penalty factor for the weight.

[0023] In this technical respect, to correct the behavior of users splitting orders to increase their value score, it is also necessary to downgrade the weights in the value score calculation formula. Specifically, in this embodiment, the downgraded weights are: ; Taking the above embodiment as an example, the weight Calculated based on the above formula: ; Based on the calculation, the real-time values ​​of the other weights after dynamic adjustment are obtained, where: ; ; According to the formula: User B's real-time value score is: ; Based on the above, user B's value score before being penalized is: The value score after punishment Compared to the value score before the penalty, the value score after the penalty is significantly lower. Based on the above, this can be used to prevent users from splitting large orders into multiple smaller orders, increasing the frequency of consumption based on multiple smaller orders, and then using the frequency of consumption to increase the value score, thereby matching higher membership levels and obtaining unreasonable benefits.

[0024] In one embodiment, in step S2, the industry characteristic weighting factor The method for obtaining it is as follows: Analysis of high-frequency user retention rates in historical data of the platform and low-frequency user churn rate ; when Start Dynamic adjustment; among which, For threshold; According to the formula: ; in, To adjust the coefficient, and .

[0025] Based on the above, when users break down large orders into multiple smaller orders and rely on increased purchase frequency to inflate user value in order to obtain higher membership levels and benefits, a penalty factor is needed to weaken the value score calculation process. Furthermore, in actual operation, different products, such as fast-moving consumer goods and luxury goods, exhibit different user consumption behaviors during the actual sales process. Analysis shows that when high-frequency user retention rates are monitored... low-frequency user churn rate Assuming a threshold Adjustment coefficient Then, according to the calculation: ; Based on the comparison, Then the weighting coefficients need to be adjusted automatically: ; Correspondingly, as mentioned above, when any weight After adjustment, the weighting will be automatically activated. The adjustment differs from the above embodiments, where the weighting of user-split large orders and deliberate generation of numerous consumption frequency reductions is adjusted to target high-frequency user retention rates. and low-frequency user churn rate To achieve fixed weight factors Dynamic adjustments help match user consumption behavior in the current environment, making the match between user consumption behavior and user value score more accurate. Furthermore, in cases where users split orders, the weighting factors are redefined accordingly. and complete the weighting Adjustments.

[0026] In one embodiment, an acceleration benefit channel is opened for the high-frequency, low-consumption users, the method being as follows: Daily consumption frequency At that time, temporary rights are triggered; among them, This is the threshold for daily consumption frequency. The temporary rights include: Temporarily upgrade membership level to , And the validity period of the rights is ,in, The interval between the next purchase, in units of .

[0027] The value score calculated from consumption behavior is used to accumulate and map user membership levels. It is understood that the value score generated by user consumption is accumulated on the original total value score after a rolling time window, thereby constructing a total value score corresponding to the user's account. Different membership levels are mapped based on the user's total value score. However, the total value score differs significantly between membership levels, and the accumulation of value score in a single transaction or rolling time window is limited, making it impossible to quickly transition between membership levels. It is understood that if membership level transitions and adjustments were easily made, membership benefits would be ineffective, rendering them meaningless. In one embodiment, considering the difficulty in adjusting user membership levels in a short time (i.e., the delay in traditional benefit matching, which fails to immediately incentivize high-frequency behavior), in one embodiment of this invention, it is assumed that a user's daily purchase frequency... If so, it can be considered that the user has high-frequency consumption behavior on that day. It is worth noting that, as mentioned in the above embodiment, in order to prevent users from splitting orders, it is necessary to monitor the user's behavior. Consumption behavior generated within an hour is monitored to prevent the splitting of orders to increase purchase frequency. This differs from the temporary incentive measures activated based on the number of user purchases. It's understandable that the monitoring period for the aforementioned order splitting behavior is short, allowing for... Hours, and the number of daily purchases here This occurs within a single day, that is, within 24 hours, and is related to the user's daily purchase frequency. It allows for temporary adjustments to a user's membership level and assigns a validity period, which can be... Based on the above, the maximum is 24. In other words, in order to incentivize users to make real, high-frequency purchases on a single day, the user's membership level can be temporarily adjusted, allowing the user to enjoy higher benefits.

[0028] In one embodiment, in S1, the proximity degree The calculation method is as follows: Prominence calculated based on the time decay model ; According to the formula: ; in, For consumption time intervals, This is the industry attenuation factor.

[0029] As mentioned above, recency is mainly used to characterize the difference between the time a user made a purchase and the current time. The closer the user's most recent purchase time is to the current time, the higher the recency. Assume user A's purchase interval is... Heaven, then , where settings .

[0030] In one embodiment, in S3, based on value scores Dynamically classify membership levels The method is as follows: S31: Rolling cumulative historical value; According to the formula: ,in, For the first A rolling time window Internally calculated value score, This represents the cumulative number of windows. S32: Level Dynamic Threshold Matching Preset membership level threshold range Corresponding level to ; S33: Acquire Cumulative Historical Value and identify The level corresponding to the threshold range of belonging This refers to the user's membership level.

[0031] In one embodiment, in step S4, the rights and interests rule base stores multi-level membership levels. The corresponding benefits package matrix is ​​based on the user's current level. It calls the corresponding set of rights items in the rights package matrix and pushes the set of rights items to the user terminal.

[0032] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for automatically matching membership benefits based on membership level in a B2B2C platform, characterized in that: Includes the following steps: S1: Collect user spending amount Consumption frequency Proximateity Three-dimensional behavioral data; S2: Industry-specific weighting factors based on platform presets Calculate user value score ; S3: Based on value score Dynamically classify membership levels ; S4: Call the rights and interests rule base and assign levels Mapped to a benefits package and pushed to the user's terminal in real time.

2. The method for automatically matching benefits based on membership level in a B2B2C platform according to claim 1, characterized in that: In step S2, the user value score is calculated. The method is as follows: retrieve user's spending amount Consumption frequency Proximateity Three-dimensional behavioral data; the amount of consumption Consumption frequency Based on rolling time window calculate; According to the formula: ； in, This is the highest spending amount in the platform's history. This represents the highest consumption frequency in the platform's history. .

3. The method for automatically matching membership levels and benefits in a B2B2C platform according to claim 2, characterized in that: In step S1, the consumption frequency is obtained. Previously, this also included merging and detecting user orders, including: If the same user If multiple orders are placed consecutively within an hour, they will be combined into a single transaction, and , ; At the same time, the user is marked as a risky account, triggering... Weight downgrade: According to the formula: ； ； ； ； in, for The penalty factor for the weight.

4. The method for automatically matching membership levels and benefits in a B2B2C platform according to claim 3, characterized in that: In S2, the industry characteristic weight factor The method for obtaining it is as follows: Analysis of high-frequency user retention rates in historical data of the platform and low-frequency user churn rate ; when Start Dynamic adjustment; among which, For threshold; According to the formula: ； in, To adjust the coefficient, and .

5. The method for automatically matching benefits based on membership level in a B2B2C platform according to claim 4, characterized in that: To enable the acceleration benefit channel for the aforementioned high-frequency, low-consumption users, the method is as follows: Daily consumption frequency At that time, temporary rights are triggered; among them, This is the threshold for daily consumption frequency. The temporary rights include: Temporarily upgrade membership level to , And the validity period of the rights is ,in, The interval between the next purchase, in units of .

6. The method for automatically matching membership levels and benefits in a B2B2C platform according to claim 5, characterized in that: In S1, the proximal factor The calculation method is as follows: Prominence calculated based on the time decay model ; According to the formula: ； in, For consumption time intervals, This is the industry attenuation factor.

7. The method for automatically matching membership levels and benefits in a B2B2C platform according to claim 6, characterized in that: In S3, based on value score Dynamically classify membership levels The method is as follows: S31: Rolling cumulative historical value; According to the formula: ,in, For the first A rolling time window Internally calculated value score, This represents the cumulative number of windows. S32: Level Dynamic Threshold Matching Preset membership level threshold range Corresponding level to ; S33: Acquire Cumulative Historical Value and identify The level corresponding to the threshold range of belonging This refers to the user's membership level.

8. The method for automatically matching benefits based on membership level in a B2B2C platform according to claim 7, characterized in that: In S4, the rights and interests rule base stores multi-level membership levels. The corresponding benefits package matrix is ​​based on the user's current level. It calls the corresponding set of rights items in the rights package matrix and pushes the set of rights items to the user terminal.

Images