JP2007502482A - Rating system and method for identifying desirable customers - Google Patents

Abstract

Progressive rating methods and systems for identifying desirable customers. A predictive index is calculated for each of the customers to predict trends in the profits that the customers can generate. The predictive index is calculated based on various customer data including at least two types of customer data selected from the customer asset level, demographic information about the customer, and customer transaction history. A score is determined for each of the selected types of customer data. Furthermore, appropriate weighting corresponding to each type of customer data is obtained. The prediction index is then calculated using an advanced algorithm based on each of the weights and scores corresponding to the selected type of customer data. The prediction index is compared with a preset threshold value to determine if the customer is desirable.

Description

Related applications

This application is a priority from the following US provisional application: US Provisional Patent Application No. 60 / 472,422 entitled "Customer Scoring Model" filed May 22, 2003. US Patent Provisional Application No. 60 / 472,412 entitled “LIFETIME REVENUE MODEL” filed May 22, 2003, filed May 22, 2003, on May 23, 2003 US Patent Provisional Application No. 60 / 472,748, filed May 23, 2003, filed “FINANCE DATA MART ACCOUNT PROFITABILITY MODEL” Activity analysis model (FINANCAL D US Patent Provisional Application No. 60 / 472,747 entitled “ATA MART ATTRITION ANALYSIS MODEL”, US patent entitled “CUSTOMER REVENUE PREDICTION METHOD AND SYSTEM” filed concurrently with this application Application number No. (Attorney Case Number 67389-038), US Patent Application No. “ACTIVITY-DRIVEN, CUSTOMER PROFITABILITY CALCULATION SYSTEM” filed concurrently with this application (Attorney Case Number 67389-039) and US Patent Application No. “METHOD AND SYSTEM FOR PREDICTING ATTRITION CUSTOMERS” filed concurrently with this application. (Agent case number 67389-040). The disclosure of the above patent application is incorporated herein by reference in its entirety.

  The disclosure of this specification relates broadly to rating methods and systems for identifying desirable customers, and more specifically based on customer-related attributes such as asset levels, demographic information, and / or transaction history. The present invention relates to a rating method and system for identifying desirable customers by calculating for each customer a predictive index that predicts the profits that each customer is likely to generate.

  It is important for companies to be able to identify desirable customers from an existing customer pool. The degree of desirability of a customer can be determined based on, for example, the profit that the customer is considered to have generated or the profit that is expected to be generated. Companies should try their best to lock in the desired customers and should leave out customers that produce limited or minimal profits for the company. Providing better treatment and services to the desired customer so that the desired customer stays with the company is reasonable from an economic point of view.

  Today, some companies use hierarchical systems to determine the type of treatment they give customers based on the degree of desirability for the company. For example, a stockbroker may wish to provide such desirable customers with additional considerations such as providing select services, additional discounts, promotions, service referrals, and the like. Customer service centers also use automated systems to connect incoming calls from customers depending on the amount of profits they generate or can generate. For example, the computer system of the customer service center determines the identity of the incoming call based on the caller ID or the account number entered by the caller. The calling customer's profile is then read to determine the priority for answering the call. If the customer's profile indicates that the originating customer is the desired customer (one who has already generated or can generate significant profits), the computer system will make the incoming call the highest priority. And quickly connect this phone to one of the staff who specializes in serving good customers. On the other hand, if the customer's profile indicates that the profit generated by the customer is not sufficient to be classified as a good customer, the system will use this incoming call to the next customer service representative who will answer the call. Assign to a general queue waiting to become available.

  Determining the degree of customer desirability based on the future benefits that the customer can generate is straightforward and understandable, but to predict which types of customers can bring more benefits to the company There is no effective methodology. In the past, stockbrokers thought that the profits that a customer could generate correlated with the customer's asset level. Thus, some stockbrokers assigned a customer score to each customer based on their respective asset level such that the higher the customer asset level, the higher the customer score assigned. If the customer score exceeds a predetermined threshold, the customer is identified as a desirable customer and will receive better treatment.

  However, it has become clear that relying solely on asset levels to identify desirable customers does not work very well. For example, in a stock broker, some customers may have high asset levels, but do not frequently participate in investment activities such as stock or mutual fund transactions, and only a limited fee is charged to the stock broker. Will not bring. Thus, such customers actually have very little revenue for stock brokers despite having high asset levels. Some customers, on the other hand, may not have a very high level of assets, but they actually do quite often like a day trader. Despite the asset level not being so high, this type of customer makes more profits for stock brokers and is therefore more desirable than those with higher asset levels that generate limited income for stock brokers. No difference. Thus, a need exists for a more accurate system or technique for identifying desirable customers.

  This specification presents an advanced rating method and system for identifying desirable customers. The advantage of such a rating method and system is that the degree of customer desirability is determined based on a number of factors rather than relying solely on the asset level. A prediction index is provided to indicate the degree of desirability of each customer. In addition, this progressive rating method and system employs a unique weighting system to properly address the different importance of various factors that can affect the accuracy of the rating.

  By way of example, the customer rating method includes a predictive index for each customer, at least two types of data selected from the customer's asset level, demographic information about the customer, and customer transaction history. Calculate based on customer data. That is, a score is determined for each selected type of customer data. For example, a score for a customer's asset level can be determined by using a lookup table that includes a relationship between the asset level and the corresponding score to find a score corresponding to the customer's asset level. After a score is determined for each of the selected types of data, a prediction index for the customer is calculated based on the score. The resulting forecast index predicts the trend of profits that customers can generate, such as the amount of profits.

  In one embodiment, a prediction index for a customer is calculated by adding a score for each of the selected types of customer data. In other embodiments, a specific weighting system is used to reflect the different importance of the various customer data when calculating the prediction index. For example, a predetermined weight for each type of customer data is applied to each of the scores for each type of data, for example by multiplying the score by the weight, and a weighted score is generated. Next, a weighted score for the selected type of customer data is subjected to a mathematical operation such as addition to generate a prediction index. The weighting for each selected type of customer data can be determined empirically, such as by regression.

  In order to determine customer desirability, this progressive rating method can compare the prediction index to one or more preset thresholds. Based on the results of this comparison, each customer can be assigned a degree of desirability such as highly desirable, fairly desirable, average, undesirable, etc., which can be used for further processing or evaluation. it can.

  A data processing system, such as a computer, can be used to implement the rating methods and systems disclosed herein. The data processing system can include a processor for processing data, a data storage device connected to the processor, and data transmission means. The data storage device retains instructions for causing the data processing system to perform the functions described herein when executed by the processor. The customer database, reference database, and weighting database can be implemented on this data storage device or any other data storage device accessible to the data processing system. The instructions can be incorporated into a machine readable medium to control the data processing system to perform customer ratings. Machine readable media include optical storage media such as CD-ROM and DVD, magnetic storage media such as floppy disks and tapes, and / or solid state storage devices such as memory cards and flash ROMs. sell. Further, such instructions may be transmitted and transmitted using a carrier wave.

  Other additional advantages of the methods and systems disclosed herein will become readily apparent from the following detailed description, which is exemplary and not limiting of the invention. As can be appreciated, other different embodiments of this customer rating method and system are possible, and it is clear that some of their details can be varied in various ways, all of which are This disclosure does not depart from the disclosure herein. Accordingly, the drawings and descriptions are to be understood as being exemplary in nature and not as limiting the invention.

  In the following description, for the purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the disclosure herein. However, those skilled in the art will appreciate that the method and system of the present invention may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to avoid unnecessarily obscuring the disclosure herein.

  For illustrative purposes, the following discussion considers an exemplary rating method and system for use in identifying desirable customers in a stockbroker. It can be appreciated that the grading methods and systems disclosed herein are applicable to many other industries, can have various variations, and are encompassed within the scope of this application. I will. A schematic block diagram of a typical customer rating system 100 is shown in FIG. A data processing system 102 such as a computer is provided for generating a prediction index 110 for each of a plurality of customers based on various types of customer data. The prediction index 110 indicates how much profit a customer can generate or provides an indicator to predict. The data processing system 102 has access to three databases: a customer database 104, a reference database 106, and a weight information database 108. The customer database 104 stores various types of customer data for each of a plurality of customers. Various customer data can include, but is not limited to, asset levels, demographic information, transaction history, and the like. Data processing system 102 can select some or all of the customer data stored in customer database 104 to calculate a predictive index for multiple customers. For example, the data processing system can select asset level and demographic information, or asset level and transaction history, to calculate a predictive index.

  The data processing system 102 assigns a score to each of the selected types of customer data based on their respective contents. The criteria database 106 includes criteria data that allows the data processing system 102 to determine which score to assign based on each value or range of each type of customer data. For example, the criteria database 106 can include one or more lookup tables and can present assigned scores for each of the entries for customer data. The weight information database 108 stores weights stored in advance for each type of customer data. Details on how the weighting is determined will be described shortly. The database as shown in FIG. 1 can be implemented in one or more data storage devices connected to the data processing system 102, such as a hard disk or non-volatile memory. The data storage device may be local to the data processing system 102 or may be located on another computer and connected to the data processing system 102 via a data transmission link such as a LAN (Local Area Network) or the Internet. .

  In calculating the predictive index for a particular customer, the data processing system 102 accesses the customer database 104 to retrieve the selected type of customer data corresponding to that particular customer. Further, the data processing system 102 accesses the reference database 106 and reads the reference data regarding the selected type of customer data. The data processing system 102 then assigns a score based on the reference data for each of the selected types of customer data. For example, for all data entries of the selected type of customer data, the data processing system 102 accesses a lookup table stored in the criteria database 106 to obtain a corresponding score to be assigned to each data entry. decide. Next, the processing system 102 calculates a prediction index for the particular customer using a unique algorithm based on the score assigned to each of the selected types of customer data corresponding to the customer. . In one embodiment, the data processing system 102 accesses the weight information database 108 when generating the prediction index so that the different importance of each type of customer data is taken into account in the generation of the prediction index, and the selection The weights stored in advance for the selected type of customer data are read, and the respective weights are applied to the respective scores assigned to the selected type of customer data.

In one embodiment, data processing system 102 uses the following algorithm to determine a predictive index for a customer:
C = aA + bB + cC + dD + eE + fF + gG (a)
here,
C is the prediction index to be calculated,
A, B, C, D, E, F, and G are the scores assigned to the various customer data of the customer,
a, b, c, d, e, f, g are predetermined weights corresponding to each type of customer data (the process for determining each weight will be described shortly).

  Equation (a) uses 6 types of customer data to calculate the forecast index, but the exact number and / or type of customer data used to generate the forecast index is fixed at 6. It has not been done. Rather, it depends on design choices. The type of customer data used to calculate the prediction index may be more or less. For example, the customer database 102 can store customer data regarding asset levels, demographic information, and transaction history. However, the algorithm used by the data processing system 102 may use only two types of customer data to generate the prediction index. For example, the algorithm may use only asset level and demographic information to generate a predictive index.

Details of the customer database 102, the reference database 106, and the weight information database 108 will be described below.
(1) Customer database The customer database 104 stores data entries for each customer. Data entries in the customer database 104 include various types of customer data such as asset levels, transaction history, and demographic data. A customer's asset level is defined as the sum of all assets owned by that customer (as long as data is available). In the example of a stock broker, assets considered to be owned by customers include, but are not limited to, common stock accounts, preferred stocks, subscription rights / warrants, units, options, bonds, CMO / MBS / Includes ABS, money market, municipal bonds, US bonds, investment trusts, investment trusts with sales commissions, UIT, and / or any other type of means and assets that may be owned by a customer.

  Demographic data is defined as information about attributes and / or features about the customer or information that can be used to identify the customer. For example, demographic data includes, but is not limited to, the duration of transactions with the stock broker, customers in the same family, city size, age, gender, educational background, marital status, income, address , Home ownership status, number and / or type of vehicles owned, household income, number of families, number of children, age of children, frequency of eating out, hobbies, etc. This list is not exhaustive. Any attributes relating to the customer can be used to generate a prediction index after empirical studies have been performed on the impact each of them has on the prediction index.

  Data relating to a transaction history is defined as all kinds of information about every transaction that a user has executed in the past. It is possible to use other transaction data (if known), but this data is usually a brokerage firm in our example, but with a trader who wants to calculate and use a profit forecasting index. Related to the transaction history. In such an example, transaction history data includes transaction date, transaction type, transaction amount, transaction frequency, average transaction amount, number of transactions per month, average transaction per month, within a specific time period. Total trades, number of shares per trade, 12-month moving average of total trades per month, etc. can be included. Further, the transaction history data can include actual revenue or revenue data, metrics derived from revenue or revenue, such as the amount of brokerage fees, or actual or average commissions.

  Other types of customer data may also be included in the customer database 104 for use in calculating the prediction index. For example, for stock brokers, the following types of customer information: average market value for the last 3 months, average market value for the last 3 months, recent market value Average total assets for the last three months, average total assets for the last three months, average total assets for the last twelve months, fees for the last three months, interest or other expenses for the last three months, The number of transactions in the last three months, deposit of funds in the last three months, withdrawal of funds in the last three months, number of account types, and / or days of deposit delay, etc. can also be used. The number and / or type of customer data contained in the customer database 104 depends on design choices. Empirically determine whether a certain type of customer data affects the trend in profit generated by the customer, so whether a variable, ie some type of data, can correlate with a trend in profit generation Thus, regression can be used.

FIG. 2 shows the data structure of a typical data entry 204 in the customer database 104. A unique customer ID 211 is assigned to each customer for identification. Data entry 204 includes various types of customer data, such as asset level 213, regional information 215, transaction history 217, and other types of customer data 218 that can be used to generate forecast index 110. As already mentioned, information corresponding to each type of customer data is stored in the data fields 223, 225, 227, 229.
(2) Criteria database Criteria database 106 stores criteria data used by data processing system 102 to determine a score to be assigned for each of the selected types of customer data corresponding to a customer. Yes. In one example, the baseline data is implemented in one or more lookup tables that contain relationships between various customer data and the corresponding scores to be assigned. FIG. 3 shows the data structure of a typical lookup table 306 in the criteria database 106. The data field 311 specifies the type of customer data, and the data field 312 lists contents or ranges corresponding to each type of customer data. Data field 313 shows the score assigned to the range or content of customer data specified in data field 312. For example, in the data field 322, the type of customer data specified is “asset level”. In addition, asset levels are $ 0, $ 0 to $ 1,000, $ 1,000 to $ 10,000, $ 10,000 to $ 100,000, $ 100,000 to $ 1,000,000, and It is divided into six ranges of over 1,000,000 dollars. A score is assigned to each range at the asset level. As shown in FIG. 3, a customer with an asset level of $ 0 is assigned a score of 1.67, and a customer with an asset level between $ 0 and $ 1,000 is assigned a score of 3. A customer with 33 and an asset level between $ 1,000 and $ 10,000 is assigned a score of 5.

  In order to determine a score based on the customer's asset level, the data processing system 102 first accesses the customer database 102 to retrieve data about the customer's assets and calculates the total amount of the customer's assets. The data processing system 102 then determines the score to be assigned to the customer by finding the corresponding range in the “asset level” 322 of the lookup table 306. For example, when the total amount of customer assets is determined to be $ 375,000, the customer assets are between $ 100,000 and $ 1,000,000. As shown in FIG. 3, the score corresponding to this range is 8.33. Therefore, a score of 8.33 is assigned to the customer based on his / her asset level. Lookup table 306 may also include other types of transactions, such as trading activity, duration of transaction with the merchant, customer age, number of customers in the home, customer net worth, and population of the city where the customer resides. Contains information about customer data and corresponding scores.

The score allocation and score assignment for a particular type of data need not be consistent across all types of customer data. The score assigned within a particular type of data can be determined by how important a variable, ie a type of customer data, can be to predict the profits that the customer will generate. More important customer data can be assigned higher scores, and less important customer data can be assigned lower scores. Further, the score distribution for a particular type of customer data may be in a variety of forms, such as a linear distribution or a normal distribution.
(3) Weight information database As described above, the data processing system 102 determines a score for each type of customer data corresponding to a specific customer, and then calculates a prediction index for the customer. a) can be used. Equation (a) is reprinted below.

C = aA + bB + cC + dD + eE + fF + gG (a)
here,
C is the prediction index to be calculated,
A, B, C, D, E, F, and G are the scores assigned to the various customer data of the customer,
a, b, c, d, e, f, and g are weights corresponding to the respective types of customer data.

  The weight information database 108 stores predetermined weight information corresponding to each type of customer data used for generating the prediction index.

According to one embodiment, each weighting value corresponding to each type of customer data can be determined using regression. For example, to obtain the values of weights ag in equation (a), the following regression equation is used:
R = aA + bB + cC + dD + eE + fF + gG (b)
here,
R is a known profit generated by each customer, or a prediction assigned in advance to each customer based on profit already generated by each customer or expected to be generated according to actual data and empirical considerations An index,
AG are each of the scores corresponding to the various types of actual customer data entered into equation (a),
“a” to “g” are corresponding weightings for each type of customer data selected.

  In the regression process, the customer data read from the known customer pool is provided to the regression equation (b) in order to find the respective coefficients (weights) ag corresponding to the various customer data, Each coefficient corresponding to customer data corresponds to the trend of the impact that the various customer data has on the profit or forecast index. After the regression process, the values of the weights a to g corresponding to the respective types of customer data are determined, and accessed when the data processing system 102 such as a hard disk calculates the prediction index using the formula (a). Can be stored in a data storage device.

  According to one embodiment, each weighting for each type of customer data can be included in the reference data. For example, in the reference table stored in the standard database, the score assigned to each type of customer data reflects in advance the corresponding weighting for each type of data. Certain types of customer data that play a more important role in predicting profits generated by customers are given or assigned higher scores than other types of customer data that have less impact, thereby enabling the customer rating system of the present invention However, when calculating the prediction index, the step of applying a weight to each of the determined customer scores can be omitted.

  After determining the predictive index for the customer, the data processing system 102 applies one or more preset thresholds to the determined predictive index to determine whether the customer is desirable for the stock broker. can do. For example, the preset threshold value may be as follows.

Customer score desirability 80 <highly desirable 60-80 highly desirable 40-60 desirable 20-40 average 0-20 undesirable Data processing system 102 has confirmed the degree of desirability for each customer of a stockbroker Later, a report showing the degree of desirability of each customer can be created. This report can be implemented as a computer file for further access by data processing system 102 or other data processing systems to provide customers with different levels of service based on their respective predictive indexes. For example, a customer phone computer accesses this report, which phone should be preferentially answered based on which customer has called and how desirable that customer is to the broker. In order to judge, it is possible to distinguish the incoming call. Calls from a first customer with a large predictive index should be given higher priority than calls from a second customer with a smaller predictive index, even if the second customer calls first. is there.

  FIG. 4 is a flow diagram illustrating a process for determining the degree of customer desirability. In step 401, the data processing system 102 accesses the customer database 104 and retrieves various types of customer data for the customer. In step 403, the data processing system 102 accesses the reference database 106 for reference data. Next, the data processing system 102 assigns a score to each type of customer data corresponding to the customer based on the reference data and the customer data (step 405). In step 407, the data processing system 102 accesses the weight information database 108 to obtain weighting information for each type of customer data. In step 409, the data processing system 102 calculates a prediction index for the customer by applying the respective weights and assigned scores for the customer data to equation (a) as previously described. Next, the data processing system 102 applies a preset threshold to the prediction index obtained by calculation, and determines the degree of desirability of the customer (step 411). In FIG. 4, steps 401, 403, and 405 are shown to be performed in order, but these steps may be performed simultaneously. Alternatively, the weighting information and reference data can be used for subsequent access to the data processing system 102 so that the data processing system 102 may first perform steps 403 and 405 and not repeat steps 403 and 405 for each customer. It may be stored in a memory.

  FIG. 5 shows a block diagram of an exemplary data processing system 500 in which customer rating system 100 and / or data processing system 102 may be implemented. Data processing system device 500 includes a bus 502 or other communication mechanism for communicating information, and a data processor 504 connected to bus 502 for processing data. In addition, data processing system 500 connects main memory 506, such as random access memory (RAM) or other dynamic storage device, to bus 502 for storing information and instructions executed by processor 504. Prepared. Furthermore, main memory 506 can also be used to store temporary numeric variables or other intermediate information during execution of instructions executed by data processor 504. In addition, the data processing system 500 includes a read only memory (ROM) 508 or other static storage device connected to the bus 502 to store static information and instructions for the processor 504. . A storage device 510 such as a magnetic disk or optical disk is provided for storing information and instructions and is connected to the bus 502. Further, the data processing system 500 may have suitable software and / or hardware for converting data from one format to another. An example of such a conversion operation is to convert the data format available on the system 500 to another format, such as a format for facilitating data transmission.

  A data processing system 500 is connected via a bus 502 to a display device 512 such as a cathode ray tube (CRT), a plasma display panel, or a liquid crystal display (LCD) for displaying information to an operator. Can do. An input device 514 containing alphanumeric characters and other keys is connected to the bus 502 for communicating information and command selections to the processor 504. Other types of user input devices communicate information about directions and command selections to the processor 504, and cursors such as a mouse, touchpad, trackball, or cursor direction keys for controlling cursor movement on the display device 512. Control (not shown).

  Data processing system 500 is controlled in response to a processor 504 that executes one or more sequences of one or more instructions contained in main memory 506. Such instructions can be read into main memory 506 from other machine-readable media such as storage device 510. When executing the instructions / sequences contained in main memory 506, processor 504 executes the process steps described herein. For example, under the control of pre-stored instructions, the data processor 504 may store customer data, reference data, and data stored in the data storage 510 and / or other data storage connected to the data processing system. Access the weighted data and generate a customer score and / or a predictive index for the customer. In other embodiments, a hardwired circuit can be used to perform the customer ratings disclosed herein instead of or in combination with software instructions. . That is, customer rating embodiments are in no way limited to a specific combination of hardware circuitry and software.

  The term “machine-readable medium” as used herein refers to any medium that participates in providing instructions to processor 504 for execution or providing data to processor 504 for processing. Point to. Such a medium may take many forms, including but not limited to, non-volatile media, volatile media, and transmission media. Non-volatile media includes, for example, optical or magnetic disks such as storage device 510. Volatile media includes dynamic memory, such as main memory 506. Transmission media includes coaxial cables, copper wires, and optical fibers, including the wires that make up the bus 502 or external network. Further, the transmission medium may take the form of acoustic or light waves that can be carried on a bus or external network link, such as those generated in radio wave and infrared data communications.

  Common forms of machine readable media include, for example, floppy disk, flexible disk, hard disk, magnetic tape, or any other magnetic medium, CD-ROM, any other optical medium, punch card, paper tape , Any other physical media with hole patterns, RAM, PROM, and EPROM, Flash EPROM, any other memory chip or cartridge, carrier wave as described below, or data processing system performs read Any medium capable is included.

  Various forms of machine-readable media may be involved in communicating one or more sequences of one or more instructions to processor 504 for execution. For example, the instructions can be initially stored on a magnetic disk of a remote data processing system such as a server. The remote data processing system can load the instructions into its dynamic memory and send the instructions over a telephone line using a modem. A modem at hand in the data processing system 500 can receive data on the telephone line and use an infrared transmitter to convert the data into an infrared signal. An infrared detector can receive the data carried in the infrared signal, and appropriate circuitry can place this data on the bus 502. Of course, various broadband communication technologies / equipment can be used for any of these links. Bus 502 carries data to main memory 506 and processor 504 reads and executes instructions from main memory 506 and / or processes the data. The instructions and / or data received by main memory 506 may be stored on storage device 510 either before or after execution by processor 504 or other processing, depending on the circumstances.

  The data processing system 500 further includes a communication interface 518 connected to the bus 502. Communication interface 518 provides a two-way data communication connection to a network link 520 that is connected to a local network. For example, the communication interface 518 may be an integrated digital communication service network (ISDN) card or modem that provides a data communication connection to a corresponding type of telephone line. As another example, communication interface 518 may be a wired or wireless local area network (LAN) card that provides a data communication connection to a compatible LAN. In any such implementation, communication interface 518 sends and receives electrical, electromagnetic or optical signals that carry digital data streams representing various types of information.

  Network link 520 typically provides data communication to other data devices over one or more networks. For example, the network link 520 can provide a connection to a data facility operated by an Internet service provider (ISP) 526 via a local network. ISP 526 then provides a data communication service over a world wide packet data communication network now widely referred to as the Internet 527. Both the local ISP network 526 and the Internet 527 use electrical, electromagnetic or optical signals that carry digital data streams. The signals that pass digital data to and from the various networks that carry digital data to and from the data processing system 500 and the signals that pass over the network link 520 and through the communication interface 518 are typical of carrier waves carrying information. It is a typical form.

  The data processing system 500 can send messages and receive data including program code via the network, network link 520, and communication interface 518. In the Internet example, the server 530 may send the requested code for the application program over the Internet 527, ISP 526, local network, and communication interface 518. The program can, for example, carry out the customer ratings already mentioned. In addition, the ability to communicate enables the loading of relevant data into the system for processing according to customer rating applications.

  Further, the data processing system 500 has various signal input / output ports for connection to and communication with peripheral devices such as printers and display devices. These input / output ports may include USB ports, PS / 2 ports, serial ports, parallel ports, IEEE-1394 ports, infrared communication ports, and / or other manufacturer specific ports. The data processing system 500 can communicate with other data processing systems via such signal input / output ports.

  Although the most common format at this time, those skilled in the art will understand that a personal computer (PC) is the only type of data processing system that can be used to implement a rating system. . Other end user devices include personal digital assistants (PDAs) with appropriate communication interfaces, mobile phones or other wireless telephone devices with web or internet access capabilities, web TV devices, and the like.

  The rating systems and methods described herein can be implemented using a single data processing system, such as a single PC, or can be implemented using a combination of various types of data processing systems. For example, a client-server structure, or a distributed data processing architecture can be used to implement the rating system herein, where multiple data processing systems connect to a network for mutual communication. Is done. Some of the data processing systems serve as servers that process data streams, provide computing services or access to customer data, and / or update software on other data processing systems connected to the network it can.

  All the contents included in the above description and all the contents shown in the accompanying drawings are to be interpreted as examples and should not be construed as limiting the present invention. Also, the following claims are all of the generic and specific features disclosed herein, and the scope of the various ideas of the invention that can be said to be encompassed between them as a matter of expression. It should be understood that it is intended to include all descriptions of

  The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments.

It is the general | schematic block diagram which showed the structure of the typical customer rating system. 2 shows a typical customer database data structure. Fig. 3 shows a typical lookup table contained in a reference database. FIG. 2 depicts a flowchart illustrating an exemplary process for determining customer desirability. FIG. 2 shows a schematic block diagram of a data processing system in which the exemplary customer rating system disclosed herein can be implemented.

Claims (38)

Accessing data relating to the customer, the data comprising at least two types of data selected from the group consisting of customer asset level, demographic information about the customer and customer transaction history;
Determining a score for each of the selected types of data relating to the customer;
Calculating a prediction index for the customer based on a score for each of the selected types of data regarding the customer;
A customer rating method, wherein the prediction index predicts a trend in profits that the customer can generate.   The method of claim 1, wherein calculating a predictive index for the customer comprises adding a score for each of the selected types of data for the customer. Said calculating step comprises:
Accessing a weight for each of the selected types of data relating to the customer;
Calculating a prediction index for the customer based on a score for each of the selected types of data for the customer and a weight for each of the selected types of data for the customer; Item 2. The method according to Item 1.   The method of claim 3, wherein a weight for each of the selected types of data relating to the customer is determined by regression. Accessing data on profit thresholds;
Comparing the prediction index with data relating to the profit threshold;
The method of claim 1, further comprising: informing whether the customer is desirable based on the result of the comparing step.   The method of claim 1, wherein the profit trend is indicative of a customer trend in profit generation.   The method of claim 6, wherein the profit is related to operating profit or profit from brokerage.   The method of claim 1, further comprising determining a level of service to the customer based on the calculated prediction index.   The method of claim 8, wherein the service level is related to a priority of a response to a call made by the customer. Determining a score for each of the selected types of data relating to the customer;
Accessing reference data including a score to be assigned to each of the selected types of data;
Comparing each of the selected types of data with corresponding reference data;
And determining a score for each of the selected types of data based on the result of the comparing step.   The method according to claim 10, wherein the reference data includes a look-up table including a relationship between each of the selected types of data and a corresponding score. A data processing system for rating customers,
A processor for processing the data;
A data storage device connected to the processor;
The data storage device is
Accessing data relating to the customer, the data comprising at least two types of data selected from the group consisting of customer asset level, demographic information about the customer and customer transaction history;
Determining a score for each of the selected types of data relating to the customer;
Calculating a prediction index for the customer based on a score for each of the selected types of data relating to the customer;
The system, wherein the prediction index predicts a trend in profits that the customer can generate.   The system of claim 12, wherein the data processing system is controlled to calculate a prediction index for the customer by adding a score for each of the selected types of data for the customer. The data storage device is
Accessing a weight for each of the selected types of data relating to the customer;
Calculating a prediction index for the customer based on a score for each of the selected types of data relating to the customer and a weight for each of the selected types of data relating to the customer; The system of claim 12, further comprising instructions for causing a processing system to execute.   15. The system of claim 14, wherein the data processing system is controlled to calculate a weight for each of the selected type of data relating to the customer by regression. The data processing system provides a score for each of the selected types of data relating to the customer,
Accessing reference data including a score to be assigned to each of the selected types of data;
Comparing each of the selected types of data with corresponding reference data;
13. The system of claim 12, wherein the system is controlled to determine by performing a step of determining a score for each of the selected types of data based on the result of the comparing step. The data storage device is
Accessing data on profit thresholds;
Comparing the prediction index with data relating to the profit threshold;
13. The system of claim 12, further comprising instructions for causing the data processing system to perform a step of informing whether the customer is desirable based on a result of the comparing step.   The system of claim 12, wherein the profit trend represents a customer trend in generating profits.   The system of claim 18, wherein the profit is related to operating profit or profit from brokerage.   The system of claim 12, wherein the data storage device further comprises instructions for causing the data processing system to determine a level of service to the customer based on the calculated prediction index.   21. The system of claim 20, wherein the service level is related to a priority of a response to a call made by the customer.   Instructions for controlling a data processing system to rate a customer, which may be embodied on a machine-readable medium and claim the data processing system when executed by the data processing system A program comprising instructions for executing each step of the method according to 1.   23. The program product of claim 22, wherein calculating the prediction index includes adding a score for each of the selected types of data related to the customer. Said calculating step comprises:
Accessing a weight for each of the selected types of data relating to the customer;
Calculating a prediction index for the customer based on a score for each of the selected type of data relating to the customer and a weight for each of the selected type of data relating to the customer. The program according to claim 22.   25. The program product of claim 24, wherein the data processing system is controlled to calculate a weight for each of the selected type of data relating to the customer by regression. Determining a score for each of the selected types of data relating to the customer;
Accessing reference data including a score to be assigned to each of the selected types of data;
Comparing each of the selected types of data with corresponding reference data;
23. The program according to claim 22, further comprising: determining a score for each of the selected types of data based on a result of the comparison step. The data processing system;
Accessing data on profit thresholds;
Comparing the prediction index with data relating to the profit threshold;
The program according to claim 22, further controlling to execute whether or not the customer is desirable based on a result of the comparison step. Accessing data relating to the customer, the data comprising at least two types of data selected from the group consisting of customer asset level, demographic information about the customer and customer transaction history;
Determining a prediction index for the customer based on the selected type of data relating to the customer;
A customer rating method, wherein the prediction index predicts a trend in profits that the customer can generate. The prediction index is
Determining a score for each of the selected types of data relating to the customer;
29. The method of claim 28, comprising: calculating a prediction index for the customer based on a score for each of the selected types of data relating to the customer.   30. The method of claim 29, wherein calculating a predictive index for the customer includes adding a score for each of the selected types of data for the customer. Said calculating step comprises:
Accessing a weight for each of the selected types of data relating to the customer;
Calculating a prediction index for the customer based on a score for each of the selected type of data relating to the customer and a weight for each of the selected type of data relating to the customer. 30. The method of claim 29.   32. The method of claim 31, wherein a weight for each of the selected types of data related to the customer is determined by regression. Determining a score for each of the selected types of data relating to the customer;
Accessing reference data including a score to be assigned to each of the selected types of data;
Comparing each of the selected types of data with corresponding reference data;
30. The method of claim 29, further comprising: determining a score for each of the selected types of data based on the result of the comparing step. Accessing data on profit thresholds;
Comparing the prediction index with data relating to the profit threshold;
29. The method of claim 28, further comprising informing whether the customer is desirable based on the result of the comparing step.   29. The method of claim 28, wherein the profit trend indicates a customer trend in generating profit.   36. The method of claim 35, wherein the profit is related to operating profit or profit from brokerage.   30. The method of claim 28, further comprising determining a level of service to the customer based on the calculated prediction index.   38. The method of claim 37, wherein the service level is related to a priority of telephone responses made by the customer.

Images