CN110458587A - A kind of method and system of analysis insurance electric business platform customer service work - Google Patents
A kind of method and system of analysis insurance electric business platform customer service work Download PDFInfo
- Publication number
- CN110458587A CN110458587A CN201910783978.2A CN201910783978A CN110458587A CN 110458587 A CN110458587 A CN 110458587A CN 201910783978 A CN201910783978 A CN 201910783978A CN 110458587 A CN110458587 A CN 110458587A
- Authority
- CN
- China
- Prior art keywords
- customer service
- data
- insurance
- client
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Technology Law (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a kind of method and system of analysis insurance electric business platform customer service work, which is characterized in that is collected, is stored in data warehouse including step 1, to source data；Step 2 carries out first time cleaning to source data, to filter ineffective dialog and test invalidation data；Step 3, according to different business scenarios, carry out second of data and clean, the enterprising onestep extraction of data after the integration can analyze the index of customer service work；Step 4, design QlikView module, show customer service job analysis scheme.Can achieve can reduce as far as possible customer churn for enterprise, increase user's stickiness, build better Marketing Environment for the quantitative analysis of customer service work；Also help the effect of transition of the enterprise from product competition to service competition.
Description
Technical field
The present invention relates to e-commerce field, a kind of method more particularly to analysis insurance electric business platform customer service work and
System.
Background technique
With the rapid development of internet economy, e-commerce platform it is increasingly competitive, client become it is a kind of very
Important mobility digital asset, customer relation management are increasingly becoming the focus gazed at.Social education is popularized, knowledge level
The variation for leading to consumption concept is promoted, client also increasingly payes attention to product bring value of services, this is exactly also enterprise operation
The pain spot of client, customer service also just play the key player for strengthening customer relation management.
Internet insures as the Newly Sprouted Things in electronic commerce affair, is increasingly favored by client, and client can be certainly
Main selection insurance products, insurance consulting are more convenient, it is only necessary to be operated on the net.In order to comply with insurance industry production and marketing separation
Trend, for customer provide conveniently, the method and system of transparent, inexpensive internet insurance service, need a kind of based on artificial
Management function of feeling at ease based on intellectual technology makes declaration form management more intelligent, and user group is being stablized in customer service, increases in traffic transformation
It plays an important role, therefore the present invention is analyzed for the work of insurance electric business platform customer service, it is quasi- based on oracle database
Standby back-end data, then using QlikView report tool as front end way of presentation, multi-target analysis customer service work.
Summary of the invention
The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a kind of analysis insurance electric business platform
The method and system of customer service work.
In order to solve the above technical problems, the present invention provides a kind of method of analysis insurance electric business platform customer service work, it is special
Sign is, includes the following steps:
Step 1 is collected source data, is stored in data warehouse；
Step 2 carries out first time cleaning to source data, to filter ineffective dialog and test invalidation data；
Step 3, according to different business scenarios, carry out second of data and clean, further mentioned in data after the integration
Take the index that can analyze customer service work；
Step 4, design QlikView module, show customer service job analysis scheme.
In the step 1, the source data includes customer service daily record data and website client data, the customer service daily record data
Including customer service session log data and customer service work log data, the website client data are transaction class data, the customer service
Session log data include per hour client messages' number, response time, reply message number, talk time length, customer satisfaction,
Product link generation record, opening record and recommend successfully to record after customer evaluation, consulting, seek advice from the date according to client as cutting
Point, with customer service ID, customer service title, customer service type is storage cell, generates the wide table of customer service session log data；The customer service work
Daily record data includes connection customer service, connection client, connection time, job category and action, is according to the trade connection date
Cut-off is storage ID with customer service ID, and member is with member ID storage, and non-member is with order ID storage.
In the step 2, described to source data progress, cleaning includes: for the first time
Step 2-1, customer service session log data are tentatively cleaned, for filtering ineffective dialog and test invalidation number
According to；
Step 2-2, customer service work log data are tentatively cleaned, for filtering invalid data；
In the step 3, based on the wide table that step 1 generates, customer service working index is calculated, integrated data collection is stored in data
Further include following steps in fairground:
Step 3-1, customer service session log data are directed to, in Effective Dialogue, send messages to customer service for the first time from visitor,
In the time for replying client for the first time to customer service, actively by customer service response time, dialog length, Satisfaction index marking, client
Review number generates product link number, opens record number and recommend successfully to record number, is recorded as table dm_xyz_tm_user_cnt_
New, the table dm_xyz_tm_user_cnt_new are online customer service overall target collection；
Step 3-2, statistics 10 points to 10 points of the same day of unsuccessful order customer service return visit situation of the previous day, original singly continue to pay
The case where, the new single-turn rate of customer service and conversion detailed data, be recorded as table dm_xyz_app_revisit, the table dm_xyz_
App_revisit is unsuccessful order data；
Step 3-3, be based on step 3, extract the data that customer service work log continuation of insurance is paid a return visit, customer service after return visit in 90 days,
There is order identical with former declaration form warrantee's certificate number to generate, is considered as and continues insurance successfully, conversion data is recorded as table dm_
Customer_xvbao_detl, the table dm_customer_xvbao_detl are continuation of insurance conversion data.
In the step 4, table dm_xyz_tm_user_cnt_new, dm_xyz_app_revisit, dm_customer_
Xvbao_detl shows three global modules respectively, and the work feelings of customer service are shown according to special topic, product, time and customer service various dimensions
Condition and rush single-turn income.
In the step 3-2, further includes:
Step 3-2-1, according to unsuccessful order data, cell-phone number of insuring is extracted, is extracted if cell-phone number of insuring is sky
Member's cell-phone number；
Step 3-2-2. according to customer service work log work_type=3, extract in customer service work log data member with it is non-
The return visit of member records, the client for excluding continuation of insurance, verifying and having accepted insurance；Extract member's cell-phone number of member and insuring for non-member
Cell-phone number；
Step 3-2-3. compares step 3-2-1 and the extracted cell-phone number of step 3-2-2, judges whether cell-phone number is identical,
If cell-phone number is identical, it is considered as client's original Dan Zhifu；If customer service contacts in client 7 days, under the extracted cell-phone number of step 3-2-2
It is single to pay successfully, then it is new Dan Zhifu in 7 days；
7 day data of step 3-2-4. fallback repeats step 3-2-2 to 3-2-3, and the crawl same day places an order or delay lower odd number
According to；
In the step 3-3, further includes:
Step 3-3-1 is based on step 2-2, extracts the continuation of insurance follow-up work record of customer service work log work_type=4,
Including customer service, the date is contacted, client identity, i.e., will expire order ID, warrantee's certificate number；Member pays a return visit with member ID identification visitor
Family, non-member, which pays a return visit to be distinguished according to order platform, identifies client with insurer's certificate or member ID；According to will expiring for extraction
Order ID identifies warrantee's certificate number；
If being protected again in warrantee 90 days of step 3-3-2. step 3-3-1 identification, to continue insurance successfully；
Step 3-3-3. be based on step 3-3-2, by the continuation of insurance order repeated update the customer service contacted recently under one's name；
Update of the step 3-3-4. based on data 3-3-3 repeats step 3-2-2 to step 3- as a result, 90 day data of fallback
2-3, the crawl same day continue insurance or delay continuation of insurance data, the continuation of insurance order repeated is updated in the customer service name contacted recently
Under.
In the step 3-1, the time in Effective Dialogue is accurate to the second, and the customer service response time refers specifically to client's initiation
Customer service response time after advice window, the dialog length refer specifically to customer service to the boot time of answering questions of client, the satisfaction
Index marking refers specifically to the marking of common industry evaluation, and client's active review number refers specifically to service terminates client and have whether to evaluate
Selectivity, it is described generate product link number and refer specifically to customer service linked to the product of lead referral, openings record number and push away
It recommends successfully record number and refers specifically to the number that client opens the recommended products link of customer service, be considered as successful referral if client's purchase；
In the step 3-2, the unsuccessful order refers specifically to not pay successful order, and the customer service return visit situation can not refer specifically to
The customer information of function order, the original that the original singly continues in the case where payment singly refer specifically to continue in the non-time horizon to failure of order
Original order is paid, the client's cancellation that newly singly refers specifically in the new single-turn rate of customer service and conversion detailed data is filled in originally
Order in addition newly check；In the step 3-2-3, the time of policy purchase that new single payment refers specifically to new list is later than customer service connection
The time of client；In the step 3-3, the customer service work log type is continuation of insurance；In the step 3-2-2, the customer service
Work log type is to pay a return visit；In the step 3-3-1, the customer service work log type is continuation of insurance；
A kind of system of analysis insurance electric business platform customer service work, which is characterized in that including the data source mould being sequentially connected
Block, data preprocessing module and QlikView data display module, the data source modules include access log module and business
System module, for providing data to subsequent module；It include customer service session log module and customer service in the access log module
Logging module；The operation system module is for storing order information；The data preprocessing module is in Oracle data
Library generates table dm_xyz_tm_user_cnt_new, table dm_xyz_app_revisit, table dm_ by storing process
customer_xvbao_detl；The QlikView data display module shows customer service job analysis for generating global view
Scheme carries out the data linkage between module, including trend module, seniority among brothers and sisters module, data module and thematic index comparison
Module；The trend module be used for as unit of day, week, the moon, season and year, statistics customer service Effective Dialogue quantity, dialogue duration,
Client actively evaluates quantity, satisfaction rank, the unsuccessful order of customer service and pays a return visit conversion ratio and customer service continuation of insurance return visit conversion ratio；Institute
It states seniority among brothers and sisters module to be used for as unit of day, week, the moon, season and year, seniority among brothers and sisters is effectively seeked advice from customer service to statistics, Products Show is successfully arranged
Row, the unsuccessful order of customer service pay a return visit conversion seniority among brothers and sisters and conversion seniority among brothers and sisters is paid a return visit in customer service continuation of insurance；The data module be used for day,
Week, the moon, season and year are unit, and statistics customer service, which is effectively seeked advice from, pays a return visit detailed data at single-turn detailed data, unsuccessful order
And detailed data is paid a return visit in customer service continuation of insurance；The special topic index contrast module is used for the whole index across comparisons of each special topic.
Advantageous effects of the invention:
1. reducing customer churn as far as possible for the quantitative analysis of customer service work for enterprise, increasing user's stickiness, build more
Good Marketing Environment；
2. being conducive to transition of the enterprise from product competition to service competition.
Detailed description of the invention
Fig. 1 is the method flow schematic diagram of exemplary embodiment of the present invention；
Fig. 2 is that the method for exemplary embodiment of the present invention pays a return visit unsuccessful order the flow diagram of processing；
Fig. 3 is that the method for exemplary embodiment of the present invention pays a return visit continuation of insurance the flow diagram of processing；
Fig. 4 is the system structure diagram of exemplary embodiment of the present invention.
Specific embodiment
A method of analysis insurance electric business platform customer service work, which comprises the steps of:
Step 1 is collected source data, is stored in data warehouse；
Step 2 carries out first time cleaning to source data, to filter ineffective dialog and test invalidation data；
Step 3, according to different business scenarios, carry out second of data and clean, further mentioned in data after the integration
Take the index that can analyze customer service work；
Step 4, design QlikView module, show customer service job analysis scheme.
In the step 1, the source data includes customer service daily record data and website client data, the customer service daily record data
Including customer service session log data and customer service work log data, the website client data are transaction class data, the customer service
Session log data include per hour client messages' number, response time, reply message number, talk time length, customer satisfaction,
Product link generation record, opening record and recommend successfully to record after customer evaluation, consulting, seek advice from the date according to client as cutting
Point, with customer service ID, customer service title, customer service type is storage cell, generates the wide table of customer service session log data；The customer service work
Daily record data includes connection customer service, connection client, connection time, job category and action, is according to the trade connection date
Cut-off is storage ID with customer service ID, and member is with member ID storage, and non-member is with order ID storage.
In the step 2, described to source data progress, cleaning includes: for the first time
Step 2-1, customer service session log data are tentatively cleaned, for filtering ineffective dialog and test invalidation number
According to；
Step 2-2, customer service work log data are tentatively cleaned, for filtering invalid data；
In the step 3, based on the wide table that step 1 generates, customer service working index is calculated, integrated data collection is stored in data
Further include following steps in fairground:
Step 3-1, customer service session log data are directed to, in Effective Dialogue, send messages to customer service for the first time from visitor,
In the time for replying client for the first time to customer service, actively by customer service response time, dialog length, Satisfaction index marking, client
Review number generates product link number, opens record number and recommend successfully to record number, is recorded as table dm_xyz_tm_user_cnt_
New, the table dm_xyz_tm_user_cnt_new are online customer service overall target collection；
Step 3-2, statistics 10 points to 10 points of the same day of unsuccessful order customer service return visit situation of the previous day, original singly continue to pay
The case where, the new single-turn rate of customer service and conversion detailed data, be recorded as table dm_xyz_app_revisit, the table dm_xyz_
App_revisit is unsuccessful order data；
Step 3-3, be based on step 3, extract the data that customer service work log continuation of insurance is paid a return visit, customer service after return visit in 90 days,
There is order identical with former declaration form warrantee's certificate number to generate, is considered as and continues insurance successfully, conversion data is recorded as table dm_
Customer_xvbao_detl, the table dm_customer_xvbao_detl are continuation of insurance conversion data.
In the step 4, table dm_xyz_tm_user_cnt_new, dm_xyz_app_revisit, dm_customer_
Xvbao_detl shows three global modules respectively, and the work feelings of customer service are shown according to special topic, product, time and customer service various dimensions
Condition and rush single-turn income.
In the step 3-2, further includes:
Step 3-2-1, according to unsuccessful order data, cell-phone number of insuring is extracted, is extracted if cell-phone number of insuring is sky
Member's cell-phone number；
Step 3-2-2. according to customer service work log work_type=3, extract in customer service work log data member with it is non-
The return visit of member records, the client for excluding continuation of insurance, verifying and having accepted insurance；Extract member's cell-phone number of member and insuring for non-member
Cell-phone number；
Step 3-2-3. compares step 3-2-1 and the extracted cell-phone number of step 3-2-2, judges whether cell-phone number is identical,
If cell-phone number is identical, it is considered as client's original Dan Zhifu；If customer service contacts in client 7 days, under the extracted cell-phone number of step 3-2-2
It is single to pay successfully, then it is new Dan Zhifu in 7 days；
7 day data of step 3-2-4. fallback repeats step 3-2-2 to 3-2-3, and the crawl same day places an order or delay lower odd number
According to；
In the step 3-3, further includes:
Step 3-3-1 is based on step 2-2, extracts the continuation of insurance follow-up work record of customer service work log work_type=4,
Including customer service, the date is contacted, client identity, i.e., will expire order ID, warrantee's certificate number；Member pays a return visit with member ID identification visitor
Family, non-member, which pays a return visit to be distinguished according to order platform, identifies client with insurer's certificate or member ID；According to will expiring for extraction
Order ID identifies warrantee's certificate number；
If being protected again in warrantee 90 days of step 3-3-2. step 3-3-1 identification, to continue insurance successfully；
Step 3-3-3. be based on step 3-3-2, by the continuation of insurance order repeated update the customer service contacted recently under one's name；
Update of the step 3-3-4. based on data 3-3-3 repeats step 3-2-2 to step 3- as a result, 90 day data of fallback
2-3, the crawl same day continue insurance or delay continuation of insurance data, the continuation of insurance order repeated is updated in the customer service name contacted recently
Under.
In the step 3-1, the time in Effective Dialogue is accurate to the second, and the customer service response time refers specifically to client's initiation
Customer service response time after advice window, the dialog length refer specifically to customer service to the boot time of answering questions of client, the satisfaction
Index marking refers specifically to the marking of common industry evaluation, and client's active review number refers specifically to service terminates client and have whether to evaluate
Selectivity, it is described generate product link number and refer specifically to customer service linked to the product of lead referral, openings record number and push away
It recommends successfully record number and refers specifically to the number that client opens the recommended products link of customer service, be considered as successful referral if client's purchase；
In the step 3-2, the unsuccessful order refers specifically to not pay successful order, and the customer service return visit situation can not refer specifically to
The customer information of function order, the original that the original singly continues in the case where payment singly refer specifically to continue in the non-time horizon to failure of order
Original order is paid, the client's cancellation that newly singly refers specifically in the new single-turn rate of customer service and conversion detailed data is filled in originally
Order in addition newly check；In the step 3-2-3, the time of policy purchase that new single payment refers specifically to new list is later than customer service connection
The time of client；In the step 3-3, the customer service work log type is continuation of insurance；In the step 3-2-2, the customer service
Work log type is to pay a return visit；In the step 3-3-1, the customer service work log type is continuation of insurance；
A kind of system of analysis insurance electric business platform customer service work, which is characterized in that including the data source mould being sequentially connected
Block, data preprocessing module and QlikView data display module, the data source modules include access log module and business
System module, for providing data to subsequent module；It include customer service session log module and customer service in the access log module
Logging module；The operation system module is for storing order information；The data preprocessing module is in Oracle data
Library generates table dm_xyz_tm_user_cnt_new, table dm_xyz_app_revisit, table dm_ by storing process
customer_xvbao_detl；The QlikView data display module shows customer service job analysis for generating global view
Scheme carries out the data linkage between module, including trend module, seniority among brothers and sisters module, data module and thematic index comparison
Module；The trend module be used for as unit of day, week, the moon, season and year, statistics customer service Effective Dialogue quantity, dialogue duration,
Client actively evaluates quantity, satisfaction rank, the unsuccessful order of customer service and pays a return visit conversion ratio and customer service continuation of insurance return visit conversion ratio；Institute
It states seniority among brothers and sisters module to be used for as unit of day, week, the moon, season and year, seniority among brothers and sisters is effectively seeked advice from customer service to statistics, Products Show is successfully arranged
Row, the unsuccessful order of customer service pay a return visit conversion seniority among brothers and sisters and conversion seniority among brothers and sisters is paid a return visit in customer service continuation of insurance；The data module be used for day,
Week, the moon, season and year are unit, and statistics customer service, which is effectively seeked advice from, pays a return visit detailed data at single-turn detailed data, unsuccessful order
And detailed data is paid a return visit in customer service continuation of insurance；The special topic index contrast module is used for the whole index across comparisons of each special topic.
The present invention is further illustrated with exemplary embodiment with reference to the accompanying drawing:
The method of insurance electric business Platform Analysis customer service work as shown in Figure 1, including
Step 1: source data is collected, is stored in data warehouse, including transaction data, client's browsing behavior, visitor
Logging etc. is taken, seeks advice from the date according to client as cut-off, with customer service ID, customer service title, customer service type is storage cell, raw
It include message count, response time, dialog length, the indexs such as satisfaction at the wide table of customer service session log data；It works customer service and remembers
Record carries out data collection, is cut-off according to the trade connection date, is storage ID with customer service ID, and member is non-with member ID storage
Member is with order ID storage；It is cut-off according to the order time of payment, using member as storage cell to transaction data；
Step 2. carries out first time cleaning to source data；
Step 21. cleans customer service session log, filters ineffective dialog or test invalidation data；
Step 22. cleans customer service work log, filters invalid data；
Second of the cleaning of step 3. data, parameter, integrated data collection are stored in Data Mart；
Step 31. is extracted daily customer service active session detailed data (according to business rule limiting session state), and visitor is calculated
The indexs such as session number, session duration, client's number, review number are taken, specifically includes and calculates each customer service session number, session duration is rung
The session number of time-out is answered, the session number timely responded to, client's number, evaluation score, (satisfaction is unsatisfied with evaluation sum, very full
Meaning, the summation of general evaluation)；
Step 32. calculates the unsuccessful order of customer service and pays a return visit conversion index；
Step 4. designs QlikView module based on service display requirement, shows the scheme of analysis customer service working method.
As shown in Fig. 2, unsuccessful order data and the corresponding cell-phone number of step 321. extraction stipulated time section, specifically
To extract cell-phone number of insuring according to 10 points to 10 points of the same day of unsuccessful order data of the previous day, cell-phone number of such as insuring is sky
Then extract member's cell-phone number；
Step 322. extracts customer service work_type=3 working record data, extracts member and non-member in logging
Return visit record, excludes continuation of insurance, verifies, and has accepted insurance and has waited clients；Member's cell-phone number and non-member is extracted to insure cell-phone number；
Whether identical step 323. judges cell-phone number according to step 321 and step 322, if cell-phone number is identical, for client
Former Dan Zhifu；It otherwise, is then new in 7 days if the corresponding cell-phone number of step 322, which places an order, pays success 7 days after customer service connection client
Single payment (client is later than the customer service connection time time of policy purchase)；
7 day data of step 324. fallback repeats step 322 to 323, and the crawl same day places an order or delay lower forms data；
As shown in figure 3, step 331. according to step 22, extract the continuation of insurance of customer service work_type=4 working record data with
Into logging (customer service work log type be continuation of insurance), including customer service, connection date, client identity, declaration form i.e. will be overdue
Order ID, warrantee's certificate number；Identify member and non-member's identity, matching declaration form demonstrate,proves the corresponding warrantee of overdue order
Piece number, member, which pays a return visit, identifies that client, non-member pay a return visit and distinguish according to order platform with member ID, and day cat ant is with insurer's certificate
Identify client, non-day cat ant identifies client with the member ID to place an order；
Step 332. extracts the order transaction data that above-mentioned warrantee is protected again, if the warrantee 90 that step 331 identifies
It is protected in it again, then to continue insurance successfully；
Step 333. be based on step 332, by the continuation of insurance order repeated update the customer service contacted recently under one's name；
Update of the step 334. based on data 333 repeats step 332 and arrives step 333, crawl as a result, 90 day data of fallback
The same day continues insurance or delays continuation of insurance data, by the continuation of insurance order repeated update the customer service contacted recently under one's name；
The system of insurance electric business Platform Analysis customer service work as shown in Figure 4, including:
Module 1: data source modules, this module provide data to subsequent module；
Module 11: access log module (customer service daily record data): customer service session log module, customer service logging；
Module 12: operation system module (website client data): this module stores new station order information, including order
Id, order pay status, time of payment, member id, warrantee's certificate number etc.；
Module 2: data preprocessing module, this module generate table dm_xyz_tm_ by storing process in oracle database
User_cnt_new, table dm_xyz_app_revisit, table dm_customer_xvbao_detl, data prediction rule are shown in reality
Apply a process, step 1 to step 4.
Module 3:QlikView data display module, this module generate global view, show customer service job analysis scheme, mould
Data linkage between block；
Module 31: trend module: can according to day, week, the moon, season, year, observation customer service Effective Dialogue quantity, dialogue duration,
Client actively evaluates quantity, satisfaction rank, and the unsuccessful order of customer service pays a return visit conversion ratio and conversion ratio is paid a return visit in customer service continuation of insurance；
Module 32: seniority among brothers and sisters module: can be according to day, in week, the moon, in season, year extracts customer service and effectively seeks advice from seniority among brothers and sisters, Products Show
Success is ranked, and the unsuccessful order of customer service pays a return visit conversion seniority among brothers and sisters and conversion seniority among brothers and sisters is paid a return visit in customer service continuation of insurance；
Module 33: data module: can be according to day, in week, the moon, in season, in year, observation customer service is effectively seeked advice from into single-turn detail number
According to unsuccessful order pays a return visit detailed data and detailed data is paid a return visit in customer service continuation of insurance；
Module 34: the whole index across comparisons of each special topic are highlighted thematic bring income effect by thematic index contrast module
Fruit.
Present invention is mainly used for a kind of method and system of insurance electric business Platform Analysis customer service work are provided, can not only be directed to
The quantitative analysis of customer service work reduces customer churn for enterprise as far as possible, increases user's stickiness, build better Marketing Environment；
Also help transition of the enterprise from product competition to service competition.
Above embodiments do not limit the present invention in any way, all to be made in a manner of equivalent transformation to above embodiments
Other improvement and application, belong to protection scope of the present invention.
Claims (9)
1. a kind of method of analysis insurance electric business platform customer service work, which comprises the steps of:
Step 1 is collected source data, is stored in data warehouse；
Step 2 carries out first time cleaning to source data, to filter ineffective dialog and test invalidation data；
Step 3, according to different business scenarios, carry out second of data and clean, the enterprising onestep extraction energy of data after the integration
Enough indexs of analysis customer service work；
Step 4, design QlikView module, show customer service job analysis scheme.
2. a kind of method of analysis insurance electric business platform customer service work as described in claim 1, which is characterized in that the step
In 1, the source data includes customer service daily record data and website client data, and the customer service daily record data includes customer service session log
Data and customer service work log data, the website client data are transaction class data, and the customer service session log data include
Per hour client messages' number, response time, reply message number, talk time length, customer satisfaction, customer evaluation, consulting after
Product link generation record opens record and recommends successfully to record, and seeks advice from the date for cut-off, with customer service ID, visitor according to client
Title is taken, customer service type is storage cell, generates the wide table of customer service session log data；The customer service work log data include connection
It is customer service, connection client, connection time, job category and action, is cut-off according to the trade connection date, with customer service ID
To store ID, member is with member ID storage, and non-member is with order ID storage.
3. a kind of method of analysis insurance electric business platform customer service work as claimed in claim 2, which is characterized in that the step
In 2, described to source data progress, cleaning includes: for the first time
Step 2-1, customer service session log data are tentatively cleaned, for filtering ineffective dialog and test invalidation data；
Step 2-2, customer service work log data are tentatively cleaned, for filtering invalid data.
4. a kind of method of analysis insurance electric business platform customer service work as claimed in claim 3, which is characterized in that the step
In 3, based on the wide table that step 1 generates, calculate customer service working index, integrated data collection is stored in Data Mart, further include as
Lower step:
Step 3-1, customer service session log data are directed to, in Effective Dialogue, customer service are sent messages to for the first time from visitor, to visitor
Clothes were replied for the first time in the time of client, and customer service response time, dialog length, Satisfaction index marking, client are actively evaluated
Number generates product link number, opens record number and recommend successfully to record number, is recorded as table dm_xyz_tm_user_cnt_new,
The table dm_xyz_tm_user_cnt_new is online customer service overall target collection；
Step 3-2, statistics 10 points to 10 points of the same day of unsuccessful order customer service return visit situation of the previous day, the former feelings for singly continuing payment
Condition, the new single-turn rate of customer service and conversion detailed data, are recorded as table dm_xyz_app_revisit, the table dm_xyz_app_
Revisit is unsuccessful order data；
Step 3-3, be based on step 3, extract the data that customer service work log continuation of insurance is paid a return visit, customer service after return visit in 90 days, have with
The identical order of former declaration form warrantee certificate number generates, and is considered as and continues insurance successfully, conversion data is recorded as table dm_customer_
Xvbao_detl, the table dm_customer_xvbao_detl are continuation of insurance conversion data.
5. a kind of method of analysis insurance electric business platform customer service work as claimed in claim 4, which is characterized in that the step
In 4, table dm_xyz_tm_user_cnt_new, dm_xyz_app_revisit, dm_customer_xvbao_detl are opened up respectively
Show three global modules, shows the working condition of customer service according to special topic, product, time and customer service various dimensions and promote single-turnization and receive
Benefit.
6. a kind of method of analysis insurance electric business platform customer service work as claimed in claim 5, which is characterized in that the step
In 3-2, further includes:
Step 3-2-1, according to unsuccessful order data, cell-phone number of insuring is extracted, extracts member if cell-phone number of insuring is sky
Cell-phone number；
Step 3-2-2. extracts member and non-member in customer service work log data according to customer service work log work_type=3
Return visit record, exclude continuation of insurance, verify and the client that has accepted insurance；Extract member's cell-phone number of member and the mobile phone of insuring of non-member
Number；
Step 3-2-3. compares step 3-2-1 and the extracted cell-phone number of step 3-2-2, judges whether cell-phone number is identical, if hand
Machine number is identical, then is considered as client's original Dan Zhifu；If customer service contacts in client 7 days, the extracted cell-phone number of step 3-2-2 places an order branch
It pays successfully, is then new Dan Zhifu in 7 days；
7 day data of step 3-2-4. fallback repeats step 3-2-2 to 3-2-3, and the crawl same day places an order or delay lower forms data.
7. a kind of method of analysis insurance electric business platform customer service work as claimed in claim 6, which is characterized in that the step
In 3-3, further includes:
Step 3-3-1 is based on step 2-2, extracts the continuation of insurance follow-up work record of customer service work log work_type=4, including
Customer service contacts the date, and client identity, i.e., will expire order ID, warrantee's certificate number；Member, which pays a return visit, identifies client with member ID,
Non-member, which pays a return visit to be distinguished according to order platform, identifies client with insurer's certificate or member ID；According to the order that will expire of extraction
ID identifies warrantee's certificate number；
If being protected again in warrantee 90 days of step 3-3-2. step 3-3-1 identification, to continue insurance successfully；
Step 3-3-3. be based on step 3-3-2, by the continuation of insurance order repeated update the customer service contacted recently under one's name；
Update of the step 3-3-4. based on data 3-3-3 repeats step 3-2-2 to step 3-2-3 as a result, 90 day data of fallback,
Continuation of insurance data are delayed in crawl same day continuation of insurance, by the continuation of insurance order repeated update the customer service contacted recently under one's name.
8. a kind of method of analysis insurance electric business platform customer service work as claimed in claim 7, which is characterized in that the step
In 3-1, the time in Effective Dialogue is accurate to the second, and the customer service response time refers specifically to customer service after client initiates advice window and answers
The time is answered, the dialog length refers specifically to customer service to the boot time of answering questions of client, and the Satisfaction index marking refers specifically to often
See that industry evaluation is given a mark, client's active review number, which refers specifically to service end client, the selectivity whether evaluated, the life
Customer service is referred specifically at product link number to link to the product of lead referral, openings record number and recommends successfully to record number specific
Refer to that client opens the number of the recommended products link of customer service, is considered as successful referral if client's purchase；In the step 3-2, institute
It states unsuccessful order to refer specifically to not pay successful order, the customer service return visit situation refers specifically to client's letter of unsuccessful order
Breath, the original that the original singly continues in the case where payment singly refer specifically to continue to pay original order in the non-time horizon to failure of order
Single, to cancel the order filled in originally in addition new by the client that newly singly refers specifically in the new single-turn rate of customer service and conversion detailed data
It checks；In the step 3-2-3, the time of policy purchase that new single payment refers specifically to new list is later than the time of customer service connection client；
In the step 3-3, the customer service work log type is continuation of insurance；In the step 3-2-2, the customer service work log type
To pay a return visit；In the step 3-3-1, the customer service work log type is continuation of insurance.
9. a kind of system for the analysis insurance electric business platform customer service work that one of -8 the methods according to claim 1 carry out,
It is characterized in that, including data source modules, data preprocessing module and the QlikView data display module being sequentially connected, the number
It include access log module and operation system module according to source module, for providing data to subsequent module；The access log mould
It include customer service session log module and customer service logging module in block；The operation system module is for storing order information；
The data preprocessing module generates table dm_xyz_tm_user_cnt_new, table by storing process in oracle database
DM_XYZ_APP_REVISIT, table dm_customer_xvbao_detl；The QlikView data display module is for generating
Global view shows customer service job analysis scheme, carries out the data linkage between module, including trend module, seniority among brothers and sisters mould
Block, data module and thematic index contrast module；The trend module was used for as unit of day, week, the moon, season and year, statistics
Customer service Effective Dialogue quantity, dialogue duration, client actively evaluate quantity, satisfaction rank, the unsuccessful order of customer service and pay a return visit conversion
Conversion ratio is paid a return visit in rate and customer service continuation of insurance；The seniority among brothers and sisters module was used for as unit of day, week, the moon, season and year, and statistics customer service has
Effect consulting seniority among brothers and sisters, Products Show are successfully ranked, the unsuccessful order of customer service pays a return visit conversion seniority among brothers and sisters and conversion row is paid a return visit in customer service continuation of insurance
Row；The data module was used for as unit of day, week, the moon, season and year, and statistics customer service is effectively seeked advice from into single-turn detail number
Detailed data is paid a return visit according to, unsuccessful order and detailed data is paid a return visit in customer service continuation of insurance；The special topic index contrast module is used for will
Each whole index across comparisons of special topic.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910783978.2A CN110458587A (en) | 2019-08-23 | 2019-08-23 | A kind of method and system of analysis insurance electric business platform customer service work |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910783978.2A CN110458587A (en) | 2019-08-23 | 2019-08-23 | A kind of method and system of analysis insurance electric business platform customer service work |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110458587A true CN110458587A (en) | 2019-11-15 |
Family
ID=68488797
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910783978.2A Withdrawn CN110458587A (en) | 2019-08-23 | 2019-08-23 | A kind of method and system of analysis insurance electric business platform customer service work |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110458587A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111221953A (en) * | 2020-01-08 | 2020-06-02 | 南京新贝金服科技有限公司 | Online pre-sale customer service effect evaluation method |
| CN112767177A (en) * | 2020-12-30 | 2021-05-07 | 中国人寿保险股份有限公司上海数据中心 | Insurance customer information management system for customer grading based on random forest |
| CN113139842A (en) * | 2020-01-17 | 2021-07-20 | 阿里巴巴集团控股有限公司 | Form processing method, device and system |
| CN115907272A (en) * | 2022-11-29 | 2023-04-04 | 贝壳找房(北京)科技有限公司 | Broker evaluation method and device, electronic device, and storage medium |
| CN117236798A (en) * | 2023-11-14 | 2023-12-15 | 北京及时语智能科技有限公司 | Comprehensive management system for intelligent dialogue customer service |
-
2019
- 2019-08-23 CN CN201910783978.2A patent/CN110458587A/en not_active Withdrawn
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111221953A (en) * | 2020-01-08 | 2020-06-02 | 南京新贝金服科技有限公司 | Online pre-sale customer service effect evaluation method |
| CN111221953B (en) * | 2020-01-08 | 2024-03-08 | 南京新贝金服科技有限公司 | Online pre-sale customer service effect evaluation method |
| CN113139842A (en) * | 2020-01-17 | 2021-07-20 | 阿里巴巴集团控股有限公司 | Form processing method, device and system |
| CN112767177A (en) * | 2020-12-30 | 2021-05-07 | 中国人寿保险股份有限公司上海数据中心 | Insurance customer information management system for customer grading based on random forest |
| CN115907272A (en) * | 2022-11-29 | 2023-04-04 | 贝壳找房(北京)科技有限公司 | Broker evaluation method and device, electronic device, and storage medium |
| CN115907272B (en) * | 2022-11-29 | 2023-11-14 | 贝壳找房(北京)科技有限公司 | Method and device for evaluating brokers, electronic equipment and storage medium |
| CN117236798A (en) * | 2023-11-14 | 2023-12-15 | 北京及时语智能科技有限公司 | Comprehensive management system for intelligent dialogue customer service |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110458587A (en) | A kind of method and system of analysis insurance electric business platform customer service work | |
| US11893593B2 (en) | Sales prediction systems and methods | |
| Peters et al. | Market information scanning activities and growth in new ventures: A comparison of service and manufacturing businesses | |
| US6925442B1 (en) | Method and apparatus for evaluating vistors to a web server | |
| Baker et al. | The changing structure of the legal services industry and the careers of lawyers | |
| US20050240433A1 (en) | Method and system for an efficient fundraising campaign over a wide area network | |
| Tucker | The implications of improved attribution and measurability for antitrust and privacy in online advertising markets | |
| CN102460493A (en) | Method and apparatus for generating a media plan | |
| CN101802856A (en) | Measuring a location based advertising campaign | |
| Slof et al. | A competing risks model based on latent Dirichlet Allocation for predicting churn reasons | |
| Al‐Qirim | E‐commerce in the aerial mapping industry: A New Zealand case study | |
| CN101223519A (en) | Method, system and computer program product for executing a marketing campaign | |
| Preissl et al. | E-life after the dot com bust | |
| CN112926997A (en) | Customer relationship management system and method | |
| Dillon et al. | How important is the yellow pages? experimental evidence from Tanzania | |
| Lekhanya | The significance of emerging technologies in promoting internationalization of rural SMEs in South Africa | |
| Aker et al. | How important is the yellow pages? experimental evidence from tanzania | |
| Guzana et al. | The challenges in employing digital marketing as a tool for improving sales at selected retail stores in the transkei region | |
| Ermes et al. | Supply chain disruption propagation: A study of South African fast-moving consumer goods food and beverage manufacturers | |
| Torero et al. | Impacts of ICT on low-income rural households | |
| US10013481B2 (en) | Using a graph database to match entities by evaluating boolean expressions | |
| CN112907266A (en) | User screening model acquisition method, financial information pushing method and related device | |
| AU2014204031A1 (en) | Quota cell priority determination to match a panelist to a market research project | |
| Pol | Marketing Automation in B2B Services as a Necessary Form in Acquiring a Contemporary Client | |
| Talk et al. | Optimize the effectiveness of recruiting campaigns |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20191115 |
|
| WW01 | Invention patent application withdrawn after publication |