DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
The following FINAL Office Action is in response to Applicant’s communication filed 01/20/2026 regarding Application 18/754,657.
Status of Claim(s)
Claim(s) 1-18 is/are currently pending and are rejected as follows.
Response to Arguments – 101 Rejection
Applicant’s arguments in regards to the previously applied 101 rejection have been fully considered and deemed persuasive.
Examiner accordingly withdraws the previously applied 101 rejection.
Response to Arguments – 102/103 Rejection
Applicant’s arguments in regards to the previously applied 103 rejection are rendered moot in view of the amended prior art rejection below.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1-6, 8-10, 12, and 14-18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Beck (US 2021/0089979 A1) in view of Ramachandran (US 2019/0373071 A1)
Claim(s) 1 –
Beck discloses the following limitations:
one or more memories that store computer program codes; (Beck: Paragraph 115, "An exemplary system for implementing the inventions, which is not illustrated, includes a general purpose computing device in the form of a conventional computer, including a processing unit, a system memory, and a system bus that couples various system components including the system memory to the processing unit. The computer typically include one or more magnetic hard disk drives (also called "data stores" or "data storage" or other names) for reading from and writing to. The drives and their associated computer-readable media provide nonvolatile storage of computer-executable instructions, data structures, program modules, and other data for the computer. Although the exemplary environment described herein employs a magnetic hard disk, a removable magnetic disk, removable optical disks, other types of computer readable media for storing data can be used, including magnetic cassettes, flash memory cards, digital video disks (DVDs), Bernoulli cartridges, RAMs, ROMs, and the like.")
and one or more processors operable to execute processes based on the computer program codes, wherein (Beck: Paragraph 114, "Those skilled in the art will also appreciate that the invention may be practiced in network computing environments with many types of computer system configurations, including personal computers, hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, networked PCs, minicomputers, mainframe computers, and the like. The invention is practiced in distributed computing environments where tasks are performed by local and remote processing devices that are linked ( either by hardwired links, wireless links, or by a combination of hardwired or wireless links) through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.")
the one or more processors are operable to execute a categorizing process for categorizing users into segments, (Beck: Paragraph 49, "The average frustration score can then be utilized to determine and (4) assess company-level vulnerability, 176 by segmenting customers on probability of attrition 180. The company vulnerability score can also be on a 1-10 scale, while probability of attrition may be used the determined vulnerability scores to determine the segments ( employees or customers) that have a high, medium or low risk of attrition. For customers, the system may also determine a segment-level average revenue per customer, for each segment that has a high, medium or low risk of attrition, as indicated in 182."; Paragraph 50, "In at least one embodiment, the computerized system uses the calculated probabilities of attrition for customer segments analysis, and calculated revenue averages for each segment to (5) automatically determining Business Value at Risk 150 and 850 in FIGS. 1 and 8, respectively. It may also include a process step of identifying key expectations 830 and industry benchmarking 111 and 811, which may include in-category benchmarks 822 and out-of-category benchmarks 824.")
and an inferring process for inferring, from attributes of the users, one or more transition attributes that are correspond to a transitioning user, (Beck: Paragraph 54, "Industry Benchmarking, 111 and 811, may include automatically processing (by a computer processor executing computer instructions stored in computer memory) the private and publicly available data on companies to determine current and historical movements in market share, customer expectation, and satisfaction metrics, as well as, or in addition to, the ARPU (average revenue per unit) and employee turnover. The out-of-category benchmarks 824 for identified areas of value creation may also be examined and evaluated. The following aspects of an industry are considered as part of the benchmarking exercise: (1) Average customer/employee relationship tenure; (2) Average customer/relationship revenue; (3) New customer/employee growth in industry; ( 4) Trends in the above over time (where available)."; Paragraph 88, "In the example on FIG. 2, the weights that may be used for each frustration metric are: Frequency 210=1, Uniqueness 220=1. Sharing 230=2, Impact 240=3, and Switching 250=3. Weights by definition for all components should add up to 10. The method used to arrive at the specific weights is automated and involves execution of a regression analysis and accounts for wider industry trends in helping to automatically predict switching behavior.": Paragraph 90, "The process step of Assessing Company-Level Vulnerability by Segmenting Customers on Probability of Attrition 180 is done after the frustrations have been tested and vulnerability scores are calculated for each individual frustration. The results are automatically modeled to determine:"; Paragraph 96, "Individual results are utilized to segment a company's customer/employee population into three groups based on likelihood to attrite:"; Paragraph 97, "High Risk of Attrition 310: highly frustrated group that is considered to be "high risk" of switching/cancelling. These customers have also likely indicated they will be leaving their provider/employer in the next 12 months."; Paragraph 98, "Medium Risk of Attrition 320: frustrated group that is considered to be "medium risk" of leaving their current provider/employer. These customer/employees may have also indicated they have considered leaving their current provider/employer in the last 12 months and are still considering action."; Paragraph 98, "Low Risk of Attrition 330: group of customers that are happy on at "low risk" of leaving their current provider as informed by their overall vulnerability scores."; Paragraph 102, "In at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320) may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity).")
the calculation result includes a result of a prediction of whether each of the users will become the transitioning user, and importance of each of the attributes in the prediction. (Beck: Paragraph 65, "The Recent Switchers 116 analysis is an automated and computerized quantitative analysis performed by the computer processor that executes specific computer instructions that cause the processor to determine and process the reasons why customers or employees of a particular company have recently switched. The software system may also create benchmarks and uncover the key reasons that have driven actual switching by customers or the employee attrition behavior in the past. For example, it could determine which employees asked for a raise or bonus and switched to another company when they did not get the expected or favorable response."; Paragraph 67, "Referring to FIGS. 1 and 8, the identification of key frustrations 140 or 840 (that are tested and evaluated in one or more embodiments) may utilize a computerized process for automatically analyzing customer pain points and determine/identify the key or most important individual frustrations to test in research by leveraging industry benchmarking, primary qualitative research, client internal voice of the customer data, social media posts, media coverage of industry and specific companies being analyzed. This approach helps create an in-depth analysis into the reasons why recent customers/employees have switched based at least in part on industry benchmarks and comparison of those benchmarks with the actual data."; Paragraph 68, "In at least one embodiment, the present invention calculates and determines (using at least one computer that executes computer instructions stored in memory) what factors and constrains define the customer's or employee's expectation in a given industry, and/or expectations with a given company. Factors are extracted (i.e., derived) using a principal component analysis and are then rotated to improve interpretability using an automated statistical software."; Paragraph 85, "Analysis can be completed on each individual frustration metric, but those metrics may be weighted and combined to create a score for each respondent, and thus provide an average score for each company in the analysis. In at least one embodiment, the present invention may utilize the weighted scheme of automatically calculated values and elements by at least one computer processor (executing specific computer instructions), to arrive at the most valid or best suited industry-specific solution."; Paragraph 86, "Therefore, in at least one embodiment, the vulnerability score may be a weighted average of the frequency with which each frustration is experienced (0 to 100 scale), the perceived uniqueness of each frustration (0 to 100 scale), the frequency with which customers/employees voice/share that frustration with colleagues, friends, or on social media (0 to 100 scale), the impact each frustration has on customers'/employees' likelihood to deepen or commit more to that relationship (0 to 100 scale), and the impact that frustration has on switching (0 to 100 scale).")
Beck does not explicitly disclose the use of a machine learning model calculation, however, in analogous art of behavior monitoring, Ramachandran discloses the following:
wherein the users use one or more services provided by a business entity, (Ramachandran: Paragraph 3, “In some implementations, a server device can segment users based on user engagement with an application on a user device. For example, server device can receive user event data from many user devices indicating user activities with respect to the application and/or content presented by the application. The server device can generate user engagement segments based on the received user event data. The server device can generate predictive models for each user engagement segment based on the received user event data. The server device can determine which particular user engagement segments a particular user is associated with based on user event data associated with the particular user and/or the predictions generated by the predictive models. The application on the user device associated with the particular user can then be configured according to the user engagement segments associated with the particular user.”; Paragraph 30, “FIG. 1 is a block diagram of an example system 100 for segmenting users based on user engagement with an application on a user device. For example, system 100 can analyze user event data to determine user engagement segments to which the user should be associated. A user engagement segment can correspond to a group of users that are grouped together based on behavioral activities and/or behavioral predictions with respect to an application (e.g., a news application) on a user device and/or content presented by the application. In other words, the user engagement segments for a user can be determine based on the user's historical engagement and/or predicted engagement with the application and/or content presented by the application.”)
wherein the users are categorized into segments, (Ramachandran: Paragraph 30, “FIG. 1 is a block diagram of an example system 100 for segmenting users based on user engagement with an application on a user device. For example, system 100 can analyze user event data to determine user engagement segments to which the user should be associated. A user engagement segment can correspond to a group of users that are grouped together based on behavioral activities and/or behavioral predictions with respect to an application (e.g., a news application) on a user device and/or content presented by the application. In other words, the user engagement segments for a user can be determine based on the user's historical engagement and/or predicted engagement with the application and/or content presented by the application.”; Paragraph 50, “In the descriptions herein, the terms “anonymous user identifier” and “user identifier” both refer to the anonymous user identifier unless otherwise defined. Moreover, the terms “user” and “user identifier” may be used interchangeably. For example, the computing devices of system 100, other than user device 102, may analyze and/or generate data for a user but do so with reference to the anonymous user identifier as described above. Thus, for example, when describing determining a user engagement segment for a user, the user engagement segment is actually done with reference to the anonymous user identifier and the user event data associated therewith. User device 102 is the only computing device that can translate the anonymous user identifier to an actual identifier that can be used to identify the user of user device 102.”)
wherein the categorization is based on discrete values converted from numerical values of segmenting indicators and further based on boundary values set so that data loss and data overlapping do not occur between adjacent segments (Ramachandran: Paragraph 61, “In some implementations, user engagement module 122 can analyze user event data 402 to determine user engagement segments 420 related to various stages of user engagement lifecycle 406. For example, the engagement lifecycle segments can include new user segment 428, activated user segment 430, adopted user segment 432, churned user segment 434, and/or reactivated user segment 436, low engaged user segment 438, medium engaged user segment 440, and highly engaged user segment 442. These user engagement lifecycle-based user engagement segments 420 (e.g., lifecycle stages) are defined further below with reference to FIG. 5.”; Paragraph 73, “In some implementations, user engagement module 122 can determine the level of engagement a user has with respect to news application 104 based on various engagement categories. For example, a score can be determined for each engagement category. The scores determined for each category can be combined (e.g., summed) to generate an engagement level score. The engagement level score can be compared to threshold values for each engagement level (e.g., high, medium, low, etc.) to determine with which engagement level the user should be associated.”; Paragraph 74, “In some implementations, user engagement module 122 can determine the level of engagement a user has with respect to news application 104 based on a recency engagement category. For example, a score generated or determined for the recency engagement category can correspond to a number of days since the user last viewed a content item, as determined by user engagement module 122 based on user event data 402. For example, if the recency category is scored 1-5, then a score of 5 can correspond to 1 day since the user last viewed a content item. A score of 4 can correspond to 2 days since the user last viewed a content item. A score of 3 can correspond to 3-6 days since the user last viewed a content item. A score of 2 can correspond to 7-12 days since the user last viewed a content item. A score of 1 can correspond to greater than 12 days since the user last viewed a content item. These scores are merely one example of how scores can be generated for the recency engagement category; other scoring schemes can be used and/or implemented.”; Paragraph 111, “When user event data 802 is processed by prediction module 126 using segment model 806, prediction module 126 can generate a prediction score 808. For example, the prediction score can indicate the likelihood that the user will perform the user engagement segment transition modeled, or predicted by, segment model 806. For example, prediction score 808 can generate a score that indicates the likelihood that the user is likely to adopt, likely to migrate up, likely to migrate down, and/or likely to churn. When the prediction score is above a threshold value (e.g., 9 on a scale of 1-10), then prediction module 126 can determine that the user is associated with the likely to adopt 444, likely to migrate up 446, likely to migrate down 448, and/or likely to churn 450 user engagement segments. As described above, other predictive models can be generated and other predictions can be made by prediction module 126 using the predictive models. The above description merely provides some specific examples of some predictions that can be made by prediction module 126 using segment model 806.”; Paragraph 119, “In some implementations, the application configuration data properties 1030 can include a publisher threshold score. For example, each content publisher can be scored according to popularity, brand awareness, reputation, and/or other factors. Users associated with the new user engagement segment 428 may be more interested in well-known publishers and therefore configuration data associated with new user engagement segment 428 may have a high publisher threshold score (e.g., above a threshold value). Users associated with the high user engagement segment 442 may be more interested in a diverse selection of publishers and therefore configuration data associated with new high engagement segment 422 may have a low publisher threshold score (e.g., below a threshold value).”)
who will transition to another segment as time elapses (Ramachandran: Paragraph 61, “In some implementations, user engagement module 122 can analyze user event data 402 to determine user engagement segments 420 related to various stages of user engagement lifecycle 406. For example, the engagement lifecycle segments can include new user segment 428, activated user segment 430, adopted user segment 432, churned user segment 434, and/or reactivated user segment 436, low engaged user segment 438, medium engaged user segment 440, and highly engaged user segment 442. These user engagement lifecycle-based user engagement segments 420 (e.g., lifecycle stages) are defined further below with reference to FIG. 5.”; Paragraph 73, “In some implementations, user engagement module 122 can determine the level of engagement a user has with respect to news application 104 based on various engagement categories. For example, a score can be determined for each engagement category. The scores determined for each category can be combined (e.g., summed) to generate an engagement level score. The engagement level score can be compared to threshold values for each engagement level (e.g., high, medium, low, etc.) to determine with which engagement level the user should be associated.”; Paragraph 74, “In some implementations, user engagement module 122 can determine the level of engagement a user has with respect to news application 104 based on a recency engagement category. For example, a score generated or determined for the recency engagement category can correspond to a number of days since the user last viewed a content item, as determined by user engagement module 122 based on user event data 402. For example, if the recency category is scored 1-5, then a score of 5 can correspond to 1 day since the user last viewed a content item. A score of 4 can correspond to 2 days since the user last viewed a content item. A score of 3 can correspond to 3-6 days since the user last viewed a content item. A score of 2 can correspond to 7-12 days since the user last viewed a content item. A score of 1 can correspond to greater than 12 days since the user last viewed a content item. These scores are merely one example of how scores can be generated for the recency engagement category; other scoring schemes can be used and/or implemented.”; Paragraph 111, “When user event data 802 is processed by prediction module 126 using segment model 806, prediction module 126 can generate a prediction score 808. For example, the prediction score can indicate the likelihood that the user will perform the user engagement segment transition modeled, or predicted by, segment model 806. For example, prediction score 808 can generate a score that indicates the likelihood that the user is likely to adopt, likely to migrate up, likely to migrate down, and/or likely to churn. When the prediction score is above a threshold value (e.g., 9 on a scale of 1-10), then prediction module 126 can determine that the user is associated with the likely to adopt 444, likely to migrate up 446, likely to migrate down 448, and/or likely to churn 450 user engagement segments. As described above, other predictive models can be generated and other predictions can be made by prediction module 126 using the predictive models. The above description merely provides some specific examples of some predictions that can be made by prediction module 126 using segment model 806.”; Paragraph 119, “In some implementations, the application configuration data properties 1030 can include a publisher threshold score. For example, each content publisher can be scored according to popularity, brand awareness, reputation, and/or other factors. Users associated with the new user engagement segment 428 may be more interested in well-known publishers and therefore configuration data associated with new user engagement segment 428 may have a high publisher threshold score (e.g., above a threshold value). Users associated with the high user engagement segment 442 may be more interested in a diverse selection of publishers and therefore configuration data associated with new high engagement segment 422 may have a low publisher threshold score (e.g., below a threshold value).”)
wherein the inferring process includes an inputting process for inputting user data including the attributes of the users to a learning model, (Ramachandran: Paragraph 52, “In some implementations, system 100 can include analytics server 120. For example, analytics server 120 can be a computing device accessible to user device 102 through a network (e.g., local area network, wide area network, the Internet, etc.). Analytics server 120 can be configured or programmed to analyze user event data 110 received from multiple user devices 102 to generate user engagement segments and/or predictive models (e.g., machine learning models) for particular user engagement segments, as described further below. In some implementations, analytics server 120 can analyze user event data 110 corresponding to a particular user (e.g., user identifier) to determine to which user engagement segments the particular user should be associated, as described further below.”; Paragraph 54, “In some implementations, analytics server 120 can include model generator 124. For example, model generator 124 can generate predictive machine learning models (e.g., random forest models) for each user engagement segment based on user event data received from (e.g., thousands of, millions of, etc.) user devices 102. For example, the predictive models can be configured to predict whether a user will adopt (e.g., regularly or consistently use) news application 104 as a source of content, churn (e.g., stop using news application 104), migrate up (e.g., the user will increase their engagement with news application 104), or migrate down (e.g., the user will decrease their engagement with news application 104).”)
the learning model outputs a calculation result when the user data is input, and (Ramachandran: Paragraph 52, “In some implementations, system 100 can include analytics server 120. For example, analytics server 120 can be a computing device accessible to user device 102 through a network (e.g., local area network, wide area network, the Internet, etc.). Analytics server 120 can be configured or programmed to analyze user event data 110 received from multiple user devices 102 to generate user engagement segments and/or predictive models (e.g., machine learning models) for particular user engagement segments, as described further below. In some implementations, analytics server 120 can analyze user event data 110 corresponding to a particular user (e.g., user identifier) to determine to which user engagement segments the particular user should be associated, as described further below.”; Paragraph 54, “In some implementations, analytics server 120 can include model generator 124. For example, model generator 124 can generate predictive machine learning models (e.g., random forest models) for each user engagement segment based on user event data received from (e.g., thousands of, millions of, etc.) user devices 102. For example, the predictive models can be configured to predict whether a user will adopt (e.g., regularly or consistently use) news application 104 as a source of content, churn (e.g., stop using news application 104), migrate up (e.g., the user will increase their engagement with news application 104), or migrate down (e.g., the user will decrease their engagement with news application 104).”)
Beck discloses a method of identifying customer opportunities from predicted behavior data. Ramachandran discloses a method for categorizing users into segments and predicting their transitions. At the time of Applicant's filed invention, one of ordinary skill in the art would have deemed it obvious to combine the methods of Beck with the teachings of Ramachandran in order to improve the efficiency and engagement of users as disclosed by Ramachandran (Ramachandran: Paragraph 6, “By reevaluating user event data and user engagement segments over time, the application configuration and content presented can evolve with the interests and behaviors of the user to keep the user engaged with the application.”)
Claim(s) 2 –
Beck in view of Ramachandran disclose the limitations of claim 1
Beck further discloses the following:
wherein the inferring process further includes extracting, from at least one of the segments, a transitional user having the one or more transition attributes. (Beck: Paragraph 65, "The Recent Switchers 116 analysis is an automated and computerized quantitative analysis performed by the computer processor that executes specific computer instructions that cause the processor to determine and process the reasons why customers or employees of a particular company have recently switched. The software system may also create benchmarks and uncover the key reasons that have driven actual switching by customers or the employee attrition behavior in the past. For example, it could determine which employees asked for a raise or bonus and switched to another company when they did not get the expected or favorable response."; Paragraph 67, "Referring to FIGS. 1 and 8, the identification of key frustrations 140 or 840 (that are tested and evaluated in one or more embodiments) may utilize a computerized process for automatically analyzing customer pain points and determine/identify the key or most important individual frustrations to test in research by leveraging industry benchmarking, primary qualitative research, client internal voice of the customer data, social media posts, media coverage of industry and specific companies being analyzed. This approach helps create an in-depth analysis into the reasons why recent customers/employees have switched based at least in part on industry benchmarks and comparison of those benchmarks with the actual data."; Paragraph 68, "In at least one embodiment, the present invention calculates and determines (using at least one computer that executes computer instructions stored in memory) what factors and constrains define the customer's or employee's expectation in a given industry, and/or expectations with a given company. Factors are extracted (i.e., derived) using a principal component analysis and are then rotated to improve interpretability using an automated statistical software."; Paragraph 85, "Analysis can be completed on each individual frustration metric, but those metrics may be weighted and combined to create a score for each respondent, and thus provide an average score for each company in the analysis. In at least one embodiment, the present invention may utilize the weighted scheme of automatically calculated values and elements by at least one computer processor ( executing specific computer instructions), to arrive at the most valid or best suited industry-specific solution."; Paragraph 86, "Therefore, in at least one embodiment, the vulnerability score may be a weighted average of the frequency with which each frustration is experienced (0 to 100 scale), the perceived uniqueness of each frustration (0 to 100 scale), the frequency with which customers/employees voice/share that frustration with colleagues, friends, or on social media (0 to 100 scale), the impact each frustration has on customers'/employees' likelihood to deepen or commit more to that relationship (0 to 100 scale), and the impact that frustration has on switching (0 to 100 scale).")
Claim(s) 3 –
Beck in view of Ramachandran disclose the limitations of claim 1
Beck further discloses the following:
the one or more services include one or more of a point-program service, a credit card service, an electronic payment service, a commercial transaction service, a travel business service, a communication service, a banking service, a securities trading service, and an insurance service, and the user data includes a usage history of the one or more services. (Beck: Paragraph 7, "An observation of current systems and methods indicates a need for a deeper understanding of customer behavior, and more importantly the drivers of that behavior, If, for example, currently used survey methodology is applied to assess and quantify customer attrition for a retail bank (e.g., for a given bank project in year 2010), the results would indicate that there is a lack of connection between the survey methodology (known and used in the industry) and the actual results (quantifiable losses of customers and impact)."; Paragraph 54, "Industry Benchmarking, 111 and 811, may include automatically processing (by a computer processor executing computer instructions stored in computer memory) the private and publicly available data on companies to determine current and historical movements in market share, customer expectation, and satisfaction metrics, as well as, or in addition to, the ARPU (average revenue per unit) and employee turnover. The out-of-category benchmarks 824 for identified areas of value creation may also be examined and evaluated. The following aspects of an industry are considered as part of the benchmarking exercise: (1) Average customer/employee relationship tenure; (2) Average customer/relationship revenue; (3) New customer/employee growth in industry; (4) Trends in the above over time (where available)."; Paragraph 57, "n at least one embodiment, the sample areas of value creation for consumers may be the following: (1) Deals and financial benefits; (2) Strong customer service; (3) Product upgrades; ( 4) Ease of access; (5) Knowledge of staff; (6) Timeliness of requests/service; (7) Convenience of service; (8) Ethical/honesty of the company. These value-creation factors may be utilized in the evaluation process and calculations to computationally assess which companies would benefit most from the business risk of others.")
Claim(s) 4 –
Beck in view of Ramachandran disclose the limitations of claims 1 and 3
Beck further discloses the following:
the segmenting indicators include a frequency indicator related to usage frequency of each of the services, and a quantitative indicator related to an expenditure on each of the services (Beck: Paragraph 49, "The average frustration score can then be utilized to determine and ( 4) assess company-level vulnerability, 176 by segmenting customers on probability of attrition 180. The company vulnerability score can also be on a 1-10 scale, while probability of attrition may be used the determined vulnerability scores to determine the segments ( employees or customers) that have a high, medium or low risk of attrition. For customers, the system may also determine a segment-level average revenue per customer, for each segment that has a high, medium or low risk of attrition, as indicated in 182."; Paragraph 54, "Industry Benchmarking, 111 and 811, may include automatically processing (by a computer processor executing computer instructions stored in computer memory) the private and publicly available data on companies to determine current and historical movements in market share, customer expectation, and satisfaction metrics, as well as, or in addition to, the ARPU (average revenue per unit) and employee turnover. The out-of-category benchmarks 824 for identified areas of value creation may also be examined and evaluated. The following aspects of an industry are considered as part of the benchmarking exercise: (1) Average customer/employee relationship tenure; (2) Average customer/relationship revenue; (3) New customer/employee growth in industry; ( 4) Trends in the above over time (where available)."; Paragraph 57, "n at least one embodiment, the sample areas of value creation for consumers may be the following: (1) Deals and financial benefits; (2) Strong customer service; (3) Product upgrades; ( 4) Ease of access; (5) Knowledge of staff; (6) Timeliness of requests/service; (7) Convenience of service; (8) Ethical/honesty of the company. These value-creation factors may be utilized in the evaluation process and calculations to computationally assess which companies would benefit most from the business risk of others."; Paragraph 96, "Individual results are utilized to segment a company's customer/employee population into three groups based on likelihood to attrite:"; Paragraph 97, "High Risk of Attrition 310: highly frustrated group that is considered to be "high risk" of switching/cancelling. These customers have also likely indicated they will be leaving their provider/employer in the next 12 months."; Paragraph 98, "Medium Risk of Attrition 320: frustrated group that is considered to be "medium risk" of leaving their current provider/employer. These customer/employees may have also indicated they have considered leaving their current provider/employer in the last 12 months and are still considering action."; Paragraph 98, "Low Risk of Attrition 330: group of customers that are happy on at "low risk" of leaving their current provider as informed by their overall vulnerability scores."; Paragraph 102, "In at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320) may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity).")
Claim(s) 5 –
Beck in view of Ramachandran disclose the limitations of claim 1
Beck further discloses the following:
wherein the segmenting indicators include indicators related to different services.(Beck: Paragraph 16, "In certain instances of the present invention, industry benchmarking data includes out-of-category benchmark data that comprises determining the average tenure relationship with the individual associated with the company, the average revenue derived by the company from the relationship with the company, new relationship growth in an associated industry, and trends data."; Paragraph 20, "The inventive system of automatic identification of key consumer frustrations from the received data involve, in certain instances, evaluating select factors, such as (1) the strength of the company's current relationship with the consumers; (2) the consumer engagement with industry; (3) the consumer satisfaction with the company; (3) the out-of-category expectation setting, including evaluation of an income-based category engagement levels with other products, and which organizations define an industry's role; and ( 4) the identity of the primary relationship owner, including identification of the primary company or product manufacturer of the consumer product."; Paragraph 22, "The inventive modeling and evaluation of the key frustrations from the received frustration data comprises automatically assigning a Vulnerability Score for each individual frustration, for individual customers or employees, for one or more companies, and for the industry overal. The Vulnerability Score may be calculated as a weighted average of the frustration factors, with specific weights assigned to the evaluated frustration factors. It could utilize a binominal logistic regression analysis to translate one or more individual-level Vulnerability Score into a probability of attrition for that individual with respect to company employment, or use of company products or services. The individual probability of attrition results for a plurality of individual can be segmented into groups, based on the determined individual Vulnerability Scores, and the determined Vulnerability Scores for different groups can be utilized to determine a Business Value at Risk for the company, including a calculation of revenue or value shift from the company or overall industry."; Paragraph 96, "Individual results are utilized to segment a company's customer/employee population into three groups based on likelihood to attrite:"; Paragraph 97, "High Risk of Attrition 310: highly frustrated group that is considered to be "high risk" of switching/cancelling. These customers have also likely indicated they will be leaving their provider/employer in the next 12 months."; Paragraph 98, "Medium Risk of Attrition 320: frustrated group that is considered to be "medium risk" of leaving their current provider/employer. These customer/employees may have also indicated they have considered leaving their current provider/employer in the last 12 months and are still considering action."; Paragraph 98, "Low Risk of Attrition 330: group of customers that are happy on at "low risk" of leaving their current provider as informed by their overall vulnerability scores."; Paragraph 102, "In at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320) may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity).")
Claim(s) 6 –
Beck in view of Ramachandran disclose the limitations of claim 1
Beck further discloses the following:
wherein the segments include at least one of a segment of new users having no usage history of a service or a segment of dormant users who have not used a service over a certain period (Beck: Paragraph 16, "In certain instances of the present invention, industry benchmarking data includes out-of-category benchmark data that comprises determining the average tenure relationship with the individual associated with the company, the average revenue derived by the company from the relationship with the company, new relationship growth in an associated industry, and trends data."; Paragraph 54, "Industry Benchmarking, 111 and 811, may include automatically processing (by a computer processor executing computer instructions stored in computer memory) the private and publicly available data on companies to determine current and historical movements in market share, customer expectation, and satisfaction metrics, as well as, or in addition to, the ARPU (average revenue per unit) and employee turnover. The out-of-category benchmarks 824 for identified areas of value creation may also be examined and evaluated. The following aspects of an industry are considered as part of the benchmarking exercise: (1) Average customer/employee relationship tenure; (2) Average customer/relationship revenue; (3) New customer/employee growth in industry; ( 4) Trends in the above over time (where available)."; Paragraph 65, "The Recent Switchers 116 analysis is an automated and computerized quantitative analysis performed by the computer processor that executes specific computer instructions that cause the processor to determine and process the reasons why customers or employees of a particular company have recently switched. The software system may also create benchmarks and uncover the key reasons that have driven actual switching by customers or the employee attrition behavior in the past. For example, it could determine which employees asked for a raise or bonus and switched to another company when they did not get the expected or favorable response."; Paragraph 67, "Referring to FIGS. 1 and 8, the identification of key frustrations 140 or 840 (that are tested and evaluated in one or more embodiments) may utilize a computerized process for automatically analyzing customer pain points and determine/identify the key or most important individual frustrations to test in research by leveraging industry benchmarking, primary qualitative research, client internal voice of the customer data, social media posts, media coverage of industry and specific companies being analyzed. This approach helps create an indepth analysis into the reasons why recent customers/employees have switched based at least in part on industry benchmarks and comparison of those benchmarks with the actual data.")
Claim(s) 8 –
Beck in view of Ramachandran disclose the limitations of claim 1
Beck further discloses the following:
acquiring user data including data of the users, the user data including classification data indicating to which of the segments each of the users belongs during each of multiple periods, and attribute data indicating the attributes of the users, (Beck: Paragraph 49, "The average frustration score can then be utilized to determine and ( 4) assess company-level vulnerability, 176 by segmenting customers on probability of attrition 180. The company vulnerability score can also be on a 1-10 scale, while probability of attrition may be used the determined vulnerability scores to determine the segments ( employees or customers) that have a high, medium or low risk of attrition. For customers, the system may also determine a segment-level average revenue per customer, for each segment that has a high, medium or low risk of attrition, as indicated in 182."; Paragraph 52, "Referring to FIG. 1, the process step of obtaining various types of data related to frustrations, 110 (used for understanding value creation and value degradation in accordance with at least one embodiment) may utilize such data points as industry benchmarking 111, qualitative research 112, voice of the customer data 113, social media listening 114, a review of media/news coverage 115, and an in-depth analysis into customers or employees who have recently switched 116 to understand areas of value creation and value degradation across an industry's players. In at least one embodiment it may be applied to customers, while in other embodiments it may be applied to employees. In some embodiments, it may be applied to both, the customers and employees for a particular business or industry."; Paragraph 65, "The Recent Switchers 116 analysis is an automated and computerized quantitative analysis performed by the computer processor that executes specific computer instructions that cause the processor to determine and process the reasons why customers or employees of a particular company have recently switched. The software system may also create benchmarks and uncover the key reasons that have driven actual switching by customers or the employee attrition behavior in the past. For example, it could determine which employees asked for a raise or bonus and switched to another company when they did not get the expected or favorable response.")
an inferring process for inferring, from the attributes, one or more transition attributes that are characteristic to the transitioning user who will transition to another segment as time elapses, and (Beck: Paragraph 65, "The Recent Switchers 116 analysis is an automated and computerized quantitative analysis performed by the computer processor that executes specific computer instructions that cause the processor to determine and process the reasons why customers or employees of a particular company have recently switched. The software system may also create benchmarks and uncover the key reasons that have driven actual switching by customers or the employee attrition behavior in the past. For example, it could determine which employees asked for a raise or bonus and switched to another company when they did not get the expected or favorable response."; Paragraph 67, "Referring to FIGS. 1 and 8, the identification of key frustrations 140 or 840 (that are tested and evaluated in one or more embodiments) may utilize a computerized process for automatically analyzing customer pain points and determine/identify the key or most important individual frustrations to test in research by leveraging industry benchmarking, primary qualitative research, client internal voice of the customer data, social media posts, media coverage of industry and specific companies being analyzed. This approach helps create an indepth analysis into the reasons why recent customers/employees have switched based at least in part on industry benchmarks and comparison of those benchmarks with the actual data."; Paragraph 68, "In at least one embodiment, the present invention calculates and determines (using at least one computer that executes computer instructions stored in memory) what factors and constrains define the customer's or employee's expectation in a given industry, and/or expectations with a given company. Factors are extracted (i.e., derived) using a principal component analysis and are then rotated to improve interpretability using an automated statistical software.")
an extracting process for extracting one or more similar users from at least one of the segments, the one or more similar users each have one or more attributes similar to the one of more transition attributes. (Beck: Paragraph 83, "The process of testing via automated, computerized market research with customers or employees of the companies being analyzed. The customer or employee frustrations and levels of frustrations may be tested for and account for various characteristics, including without limitation the following: (a) Frequency 161, indicating how frequent this frustration occurs; (b) Uniqueness 162, indicating how unique the frustration is to a given provider or employer; (c) Sharing 163, indicating how often the frustration is shared with friends or family; ( d) Impact 164, indicating how much the frustration impacts the depth of the relationship with a provider or employer; and (d) Switching 165, indicating how much the frustration prompts switching away from this provider/employer."; Paragraph 87, "In at least one embodiment, sample weights and calculations of individual frustrations can be calculated as illustrated in FIG 2, which illustrates the Sample Vulnerability Score Calculations at the Individual Frustration and Company Level with Corresponding Weights. In at least one embodiment, the Vulnerability Scores are calculated to range from 0 to 10, with 0 signifying a frustration does not occur at all and has no impact on perceptions of uniqueness, or on actual behaviors around sharing, deepening and switching. A score of 10 signifies that a frustration occurs frequently and has significant impact on perceptions of uniqueness and on actual behaviors around sharing, relationship deepening and switching."; Paragraph 88, "In the example on FIG. 2, the weights that may be used for each frustration metric are: Frequency 210=1, Uniqueness 220=1. Sharing 230=2, Impact 240=3, and Switching 250=3. Weights by definition for all components should add up to 10. The method used to arrive at the specific weights is automated and involves execution of a regression analysis and accounts for wider industry trends in helping to automatically predict switching behavior."; Paragraph 90, "The process step of Assessing Company-Level Vulnerability by Segmenting Customers on Probability of Attrition 180 is done after the frustrations have been tested and vulnerability scores are calculated for each individual frustration. The results are automatically modeled to determine:")
Claim(s) 9 –
Beck in view of Ramachandran disclose the limitations of claims 1 and 8
Beck further discloses the following:
wherein the extracting process further includes extracting one or more transitional users from at least one of the segments, and the one or more similar users are extracted from users remaining in the at least one of the segments from which the one or more transitional users are extracted. (Beck: Paragraph 83, "The process of testing via automated, computerized market research with customers or employees of the companies being analyzed. The customer or employee frustrations and levels of frustrations may be tested for and account for various characteristics, including without limitation the following: (a) Frequency 161, indicating how frequent this frustration occurs; (b) Uniquenessl62, indicating how unique the frustration is to a given provider or employer; (c) Sharing 163, indicating how often the frustration is shared with friends or family; (d) Impact 164, indicating how much the frustration impacts the depth of the relationship with a provider or employer; and (d) Switching 165, indicating how much the frustration prompts switching away from this provider/employer."; Paragraph 87, "In at least one embodiment, sample weights and calculations of individual frustrations can be calculated as illustrated in FIG 2, which illustrates the Sample Vulnerability Score Calculations at the Individual Frustration and Company Level with Corresponding Weights. In at least one embodiment, the Vulnerability Scores are calculated to range from 0 to 10, with 0 signifying a frustration does not occur at all and has no impact on perceptions of uniqueness, or on actual behaviors around sharing, deepening and switching. A score of 10 signifies that a frustration occurs frequently and has significant impact on perceptions of uniqueness and on actual behaviors around sharing, relationship deepening and switching."; Paragraph 88, "In the example on FIG. 2, the weights that may be used for each frustration metric are: Frequency 210=1, Uniqueness 220=1. Sharing 230=2, Impact 240=3, and Switching 250=3. Weights by definition for all components should add up to 10. The method used to arrive at the specific weights is automated and involves execution of a regression analysis and accounts for wider industry trends in helping to automatically predict switching behavior."; Paragraph 90, "The process step of Assessing Company-Level Vulnerability by Segmenting Customers on Probability of Attrition 180 is done after the frustrations have been tested and vulnerability scores are calculated for each individual frustration. The results are automatically modeled to determine:"; Paragraph 101, "The step of Automatically Determining Business Value at Risk 50 and 850, as illustrated in FIGS. 1 and 8, may be based on and include calculations of a company, industry, and individual respondent financial data, as well as share of vulnerable populated projected to switch providers/employers. The model for the business-value-at-risk of being lost could be developed for either the individual companies or industry overall. For the employer-focused study, the present invention may determine and provide a connection between employee attrition and customer value lost.")
Claim(s) 10 –
Beck in view of Ramachandran disclose the limitations of claims 1 and 8-9
Beck further discloses the following:
further execute a target outputting process for outputting a target group, and the target group includes the one or more transitional users and the one or more similar users. (Beck: Paragraph 102, "n at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320)may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity)."; Paragraph 103, "This phase of the methodology can also act as a diagnostic tool for individual companies, to determine a strategy for either capturing new value or preventing value erosion by addressing the key identified and quantified frustrations causing attrition. The predictive analysis and remedial measures may come from either the firm itself or from major competitors projected to lose the most value in a particular industry.")
Claim(s) 12 –
Beck in view of Ramachandran disclose the limitations of claim 1
Beck further discloses the following:
forecasting process for forecasting an economic performance when user transition occurs, the forecasting process includes calculating a transition performance coefficient that indicates the economic performance when transition from one of the segments to another segment occurs, and (Beck: Paragraph 94, "The output of this modeling is a vulnerability 'score' 290, compiled from the above metrics relating to individual frustrations. This score can be applied and evaluated horizontally-i.e. what is the most intense frustration-and vertically-i.e. which firm or company is the most vulnerable. This is further accentuated by determining and predicting which consumers or employees will definitely leave their provider or employer in the next 12 months."; Paragraph 102, "In at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320) may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity)."; Paragraph 106, "FIG. 5 illustrates the Individual Firm Analysis Output based on the Customer Attrition process for a specific bank ( e.g., Chase bank) in the retail banking industry. It shows the calculated "at risk frustrations" 510 and particularly the Frustration Score ( calculated in accordance with at least one embodiment) versus the peer group mean value. As indicated, the highest frustration score is attributed to the frustration factor of having inconsistent experience across branches, offline and over the phone (-36%) 520. The next frustration involves not offering a service to allow automatic and simple money transfers (-29%) 530, followed by having problems with the online or mobile banking tools (-29%) 540. The last frustration illustrated in FIG. 5 involves having to deal with staff that is not empowered to resolve a particular issue or frustration (-25%) 550."; Paragraph 107, "FIG. 6 illustrates the Competitive Vulnerability Financial Impacts, and customer attrition output for the cable industry. It indicates the financial impact for the customers in the cable industry, both in terms of savings for individual customers and industry-wide impacts. It also shows a loss 650 of $5.5 B to the industry due to customers' frustrations and dropping their cable services as a result of their frustrations."; Paragraph 108, "FIG. 7 illustrates an alternative Competitive Vulnerability process in accordance with at least one embodiment. This diagram indicates how the total population 710 is cut down to determine the Vulnerable population 720, is the portion that is frustrated 720. From it, the system and method determines the Switching subpopulation 730. From the Switching subpopulation 730, the Value at Risk 740 is calculated. The Value at Risk 740 calculation may include predicting the actual losses of customers or employees by a company or an overall industry in accordance with at least one embodiment. The Value at Risk 740 may also include quantified financial losses to the company or losses in value shift of employees or customers to a competitor. It also indicate the overall quantified predicted losses for the overall industry.")
comparing the transition performance coefficient of different transition paths and outputting a transition path having a higher economic performance than other ones of the transition paths. (Beck: Paragraph 94, "The output of this modeling is a vulnerability 'score' 290, compiled from the above metrics relating to individual frustrations. This score can be applied and evaluated horizontally-i.e. what is the most intense frustration-and vertically-i.e. which firm or company is the most vulnerable. This is further accentuated by determining and predicting which consumers or employees will definitely leave their provider or employer in the next 12 months."; Paragraph 102, "In at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320) may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity)."; Paragraph 106, "FIG. 5 illustrates the Individual Firm Analysis Output based on the Customer Attrition process for a specific bank ( e.g., Chase bank) in the retail banking industry. It shows the calculated "at risk frustrations" 510 and particularly the Frustration Score (calculated in accordance with at least one embodiment) versus the peer group mean value. As indicated, the highest frustration score is attributed to the frustration factor of having inconsistent experience across branches, offline and over the phone (-36%) 520. The next frustration involves not offering a service to allow automatic and simple money transfers (-29%) 530, followed by having problems with the online or mobile banking tools (-29%) 540. The last frustration illustrated in FIG. 5 involves having to deal with staff that is not empowered to resolve a particular issue or frustration (-25%) 550."; Paragraph 107, "FIG. 6 illustrates the Competitive Vulnerability Financial Impacts, and customer attrition output for the cable industry. It indicates the financial impact for the customers in the cable industry, both in terms of savings for individual customers and industry-wide impacts. It also shows a loss 650 of $5.5 B to the industry due to customers' frustrations and dropping their cable services as a result of their frustrations."; Paragraph 108, "FIG. 7 illustrates an alternative Competitive Vulnerability process in accordance with at least one embodiment. This diagram indicates how the total population 710 is cut down to determine the Vulnerable population 720, is the portion that is frustrated 720. From it, the system and method determines the Switching subpopulation 730. From the Switching subpopulation 730, the Value at Risk 740 is calculated. The Value at Risk 740 calculation may include predicting the actual losses of customers or employees by a company or an overall industry in accordance with at least one embodiment. The Value at Risk 740 may also include quantified financial losses to the company or losses in value shift of employees or customers to a competitor. It also indicate the overall quantified predicted losses for the overall industry.")
Claim(s) 14 –
Beck in view of Ramachandran disclose the limitations of claim 1
Beck further discloses the following:
execute a graph outputting process for outputting one or more graphs indicating distribution of users included in each of the segments (Beck: Paragraph 12, "The calculated business value at risk may be displayed on a display screen, printed ( e.g., via a printer) or transmitted through a computer network to the company management, where it then can be utilized by company management to quantify monetary losses caused by the predicted attrition among the individuals related to the company. Company management may further utilize the quantified and calculated Business Value at Risk (monetary losses) to implement a set of remedial measures to be carried out by the company in order to prevent company value erosion or to capture the value shift from one or more competing companies."; Paragraph 104, "In at least one embodiment, the Vulnerability Study's final output is a rank list of frustrations across the industry and company, a comparative ranking of most vulnerable firms, projected attrition, business value at risk for firms, and potential capturers of that value being lost by most vulnerable firms."; Paragraph 106, "FIG. 5 illustrates the Individual Firm Analysis Output based on the Customer Attrition process for a specific bank (e.g., Chase bank) in the retail banking industry. It shows the calculated "at risk frustrations" 510 and particularly the Frustration Score ( calculated in accordance with at least one embodiment) versus the peer group mean value. As indicated, the highest frustration score is attributed to the frustration factor of having inconsistent experience across branches, offline and over the phone (-36%) 520. The next frustration involves not offering a service to allow automatic and simple money transfers (-29%) 530, followed by having problems with the online or mobile banking tools (-29%) 540. The last frustration illustrated in FIG. 5 involves having to deal with staff that is not empowered to resolve a particular issue or frustration (-25%) 550."; Paragraph 107, "FIG. 6 illustrates the Competitive Vulnerability Financial Impacts, and customer attrition output for the cable industry. It indicates the financial impact for the customers in the cable industry, both in terms of savings for individual customers and industrywide impacts. It also shows a loss 650 of $5.5 B to the industry due to customers' frustrations and dropping their cable services as a result of their frustrations."; Paragraph 113, "Those skilled in the art will understand the features and aspects of a suitable computing environment in which aspects of the invention may be implemented. Although not required, the inventions are described in the general context of computer-executable instructions, such as program modules or engines, as described earlier, being executed by computers in networked environments. Such program modules are often reflected and illustrated by flow charts, sequence diagrams, exemplary displays, and other techniques used by those skilled in the art to communicate how to make and use such computer program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types, within the computer. Computer-executable instructions, associated data structures, and program modules represent examples of the program code for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represent examples of corresponding acts for implementing the functions described in such steps.")
Claim(s) 15 –
Beck in view of Ramachandran discloses the limitations of claims 1 and 14
Beck further discloses the following:
wherein the one or more graphs includes representation of a transition history of the users, and the transition history includes a transition path indicating transition from one of the segments to another one of the segments (Beck: Paragraph 12, "The calculated business value at risk may be displayed on a display screen, printed ( e.g., via a printer) or transmitted through a computer network to the company management, where it then can be utilized by company management to quantify monetary losses caused by the predicted attrition among the individuals related to the company. Company management may further utilize the quantified and calculated Business Value at Risk
(monetary losses) to implement a set of remedial measures to be carried out by the company in order to prevent company value erosion or to capture the value shift from one or more competing companies."; Paragraph 104, "In at least one embodiment, the Vulnerability Study's final output is a rank list of frustrations across the industry and company, a comparative ranking of most vulnerable firms, projected attrition, business value at risk for firms, and potential capturers of that value being lost by most vulnerable firms."; Paragraph 106, "FIG. 5 illustrates the Individual Firm Analysis Output based on the Customer Attrition process for a specific bank (e.g., Chase bank) in the retail banking industry. It shows the calculated "at risk frustrations" 510 and particularly the Frustration Score ( calculated in accordance with at least one embodiment) versus the peer group mean value. As indicated, the highest frustration score is attributed to the frustration factor of having inconsistent experience across branches, offline and over the phone (-36%) 520. The next frustration involves not offering a service to allow automatic and simple money transfers (-29%) 530, followed by having problems with the online or mobile banking tools (-29%) 540. The last frustration illustrated in FIG. 5 involves having to deal with staff that is not empowered to resolve a particular issue or frustration (-25%) 550."; Paragraph 107, "FIG. 6 illustrates the Competitive Vulnerability Financial Impacts, and customer attrition output for the cable industry. It indicates the financial impact for the customers in the cable industry, both in terms of savings for individual customers and industrywide impacts. It also shows a loss 650 of $5.5 B to the industry due to customers' frustrations and dropping their cable services as a result of their frustrations."; Paragraph 113, "Those skilled in the art will understand the features and aspects of a suitable computing environment in which aspects of the invention may be implemented. Although not required, the inventions are described in the general context of computer-executable instructions, such as program modules or engines, as described earlier, being executed by computers in networked environments. Such program modules are often reflected and illustrated by flow charts, sequence diagrams, exemplary displays, and other techniques used by those skilled in the art to communicate how to make and use such computer program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types, within the computer. Computer-executable instructions, associated data structures, and program modules represent examples of the program code for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represent examples of corresponding acts for implementing the functions described in such steps.")
Claim(s) 16 and 18 –
Beck disclose the following:
A non-transitory computer-readable medium storing a computer program, (Beck: Paragraph 115, "An exemplary system for implementing the inventions, which is not illustrated, includes a general purpose computing device in the form of a conventional computer, including a processing unit, a system memory, and a system bus that couples various system components including the system memory to the processing unit. The computer typically include one or more magnetic hard disk drives (also called "data stores" or "data storage" or other names) for reading from and writing to. The drives and their associated computer-readable media provide nonvolatile storage of computer-executable instructions, data structures, program modules, and other data for the computer. Although the exemplary environment described herein employs a magnetic hard disk, a removable magnetic disk, removable optical disks, other types of computer readable media for storing data can be used, including magnetic cassettes, flash memory cards, digital video disks (DVDs), Bernoulli cartridges, RAMs, ROMs, and the like.")
categorizing users into segments; and (Beck: Paragraph 49, "The average frustration score can then be utilized to determine and ( 4) assess company-level vulnerability, 176 by segmenting customers on probability of attrition 180. The company vulnerability score can also be on a 1-10 scale, while probability of attrition may be used the determined vulnerability scores to determine the segments ( employees or customers) that have a high, medium or low risk of attrition. For customers, the system may also determine a segment-level average revenue per customer, for each segment that has a high, medium or low risk of attrition, as indicated in 182."; Paragraph 50, "In at least one embodiment, the computerized system uses the calculated probabilities of attrition for customer segments analysis, and calculated revenue averages for each segment to (5) automatically determining Business Value at Risk 150 and 850 in FIGS. 1 and 8, respectively. It may also include a process step of identifying key expectations 830 and industry benchmarking 111 and 811, which may include in-category benchmarks 822 and out-of-category benchmarks 824.")
inferring, from attributes of the users, one or more transition attributes that are characteristic to a transitioning user who will transition to another segment as time elapses, wherein the inferring includes predicting whether each of the users (Beck: Paragraph 54, "Industry Benchmarking, 111 and 811, may include automatically processing (by a computer processor executing computer instructions stored in computer memory) the private and publicly available data on companies to determine current and historical movements in market share, customer expectation, and satisfaction metrics, as well as, or in addition to, the ARPU (average revenue per unit) and employee turnover. The out-of-category benchmarks 824 for identified areas of value creation may also be examined and evaluated. The following aspects of an industry are considered as part of the benchmarking exercise: (1) Average customer/employee relationship tenure; (2) Average customer/relationship revenue; (3) New customer/employee growth in industry; (4) Trends in the above over time (where available)."; Paragraph 88, "In the example on FIG. 2, the weights that may be used for each frustration metric are: Frequency 210=1, Uniqueness 220=1. Sharing 230=2, Impact 240=3, and Switching 250=3. Weights by definition for all components should add up to 10. The method used to arrive at the specific weights is automated and involves execution of a regression analysis and accounts for wider industry trends in helping to automatically predict switching behavior.": Paragraph 90, "The process step of Assessing Company-Level Vulnerability by Segmenting Customers on Probability of Attrition 180 is done after the frustrations have been tested and vulnerability scores are calculated for each individual frustration. The results are automatically modeled to determine:"; Paragraph 96, "Individual results are utilized to segment a company's customer/employee population into three groups based on likelihood to attrite:"; Paragraph 97, "High Risk of Attrition 310: highly frustrated group that is considered to be "high risk" of switching/cancelling. These customers have also likely indicated they will be leaving their provider/employer in the next 12 months."; Paragraph 98, "Medium Risk of Attrition 320: frustrated group that is considered to be "medium risk" of leaving their current provider/employer. These customer/employees may have also indicated they have considered leaving their current provider/employer in the last 12 months and are still considering action."; Paragraph 98, "Low Risk of Attrition 330: group of customers that are happy on at "low risk" of leaving their current provider as informed by their overall vulnerability scores."; Paragraph 102, "In at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320) may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity).")
calculating importance of each of the attributes in the prediction. (Beck: Paragraph 65, "The Recent Switchers 116 analysis is an automated and computerized quantitative analysis performed by the computer processor that executes specific computer instructions that cause the processor to determine and process the reasons why customers or employees of a particular company have recently switched. The software system may also create benchmarks and uncover the key reasons that have driven actual switching by customers or the employee attrition behavior in the past. For example, it could determine which employees asked for a raise or bonus and switched to another company when they did not get the expected or favorable response."; Paragraph 67, "Referring to FIGS. 1 and 8, the identification of key frustrations 140 or 840 (that are tested and evaluated in one or more embodiments) may utilize a computerized process for automatically analyzing customer pain points and determine/identify the key or most important individual frustrations to test in research by leveraging industry benchmarking, primary qualitative research, client internal voice of the customer data, social media posts, media coverage of industry and specific companies being analyzed. This approach helps create an in-depth analysis into the reasons why recent customers/employees have switched based at least in part on industry benchmarks and comparison of those benchmarks with the actual data."; Paragraph 68, "In at least one embodiment, the present invention calculates and determines (using at least one computer that executes computer instructions stored in memory) what factors and constrains define the customer's or employee's expectation in a given industry, and/or expectations with a given company. Factors are extracted (i.e., derived) using a principal component analysis and are then rotated to improve interpretability using an automated statistical software."; Paragraph 85, "Analysis can be completed on each individual frustration metric, but those metrics may be weighted and combined to create a score for each respondent, and thus provide an average score for each company in the analysis. In at least one embodiment, the present invention may utilize the weighted scheme of automatically calculated values and elements by at least one computer processor (executing specific computer instructions), to arrive at the most valid or best suited industry-specific solution."; Paragraph 86, "Therefore, in at least one embodiment, the vulnerability score may be a weighted average of the frequency with which each frustration is experienced (0 to 100 scale), the perceived uniqueness of each frustration (0 to 100 scale), the frequency with which customers/employees voice/share that frustration with colleagues, friends, or on social media (0 to 100 scale), the impact each frustration has on customers'/employees' likelihood to deepen or commit more to that relationship (0 to 100 scale), and the impact that frustration has on switching (0 to 100 scale).")
Beck does not explicitly disclose the use of a machine learning model calculation, however, in analogous art of behavior monitoring, Ramachandran discloses the following:
wherein the users are categorized into segments, (Ramachandran: Paragraph 30, “FIG. 1 is a block diagram of an example system 100 for segmenting users based on user engagement with an application on a user device. For example, system 100 can analyze user event data to determine user engagement segments to which the user should be associated. A user engagement segment can correspond to a group of users that are grouped together based on behavioral activities and/or behavioral predictions with respect to an application (e.g., a news application) on a user device and/or content presented by the application. In other words, the user engagement segments for a user can be determine based on the user's historical engagement and/or predicted engagement with the application and/or content presented by the application.”; Paragraph 50, “In the descriptions herein, the terms “anonymous user identifier” and “user identifier” both refer to the anonymous user identifier unless otherwise defined. Moreover, the terms “user” and “user identifier” may be used interchangeably. For example, the computing devices of system 100, other than user device 102, may analyze and/or generate data for a user but do so with reference to the anonymous user identifier as described above. Thus, for example, when describing determining a user engagement segment for a user, the user engagement segment is actually done with reference to the anonymous user identifier and the user event data associated therewith. User device 102 is the only computing device that can translate the anonymous user identifier to an actual identifier that can be used to identify the user of user device 102.”)
wherein the categorization is based on discrete values converted from numerical values of segmenting indicators and further based on boundary values set so that data loss and data overlapping do not occur between adjacent segments (Ramachandran: Paragraph 61, “In some implementations, user engagement module 122 can analyze user event data 402 to determine user engagement segments 420 related to various stages of user engagement lifecycle 406. For example, the engagement lifecycle segments can include new user segment 428, activated user segment 430, adopted user segment 432, churned user segment 434, and/or reactivated user segment 436, low engaged user segment 438, medium engaged user segment 440, and highly engaged user segment 442. These user engagement lifecycle-based user engagement segments 420 (e.g., lifecycle stages) are defined further below with reference to FIG. 5.”; Paragraph 73, “In some implementations, user engagement module 122 can determine the level of engagement a user has with respect to news application 104 based on various engagement categories. For example, a score can be determined for each engagement category. The scores determined for each category can be combined (e.g., summed) to generate an engagement level score. The engagement level score can be compared to threshold values for each engagement level (e.g., high, medium, low, etc.) to determine with which engagement level the user should be associated.”; Paragraph 74, “In some implementations, user engagement module 122 can determine the level of engagement a user has with respect to news application 104 based on a recency engagement category. For example, a score generated or determined for the recency engagement category can correspond to a number of days since the user last viewed a content item, as determined by user engagement module 122 based on user event data 402. For example, if the recency category is scored 1-5, then a score of 5 can correspond to 1 day since the user last viewed a content item. A score of 4 can correspond to 2 days since the user last viewed a content item. A score of 3 can correspond to 3-6 days since the user last viewed a content item. A score of 2 can correspond to 7-12 days since the user last viewed a content item. A score of 1 can correspond to greater than 12 days since the user last viewed a content item. These scores are merely one example of how scores can be generated for the recency engagement category; other scoring schemes can be used and/or implemented.”; Paragraph 111, “When user event data 802 is processed by prediction module 126 using segment model 806, prediction module 126 can generate a prediction score 808. For example, the prediction score can indicate the likelihood that the user will perform the user engagement segment transition modeled, or predicted by, segment model 806. For example, prediction score 808 can generate a score that indicates the likelihood that the user is likely to adopt, likely to migrate up, likely to migrate down, and/or likely to churn. When the prediction score is above a threshold value (e.g., 9 on a scale of 1-10), then prediction module 126 can determine that the user is associated with the likely to adopt 444, likely to migrate up 446, likely to migrate down 448, and/or likely to churn 450 user engagement segments. As described above, other predictive models can be generated and other predictions can be made by prediction module 126 using the predictive models. The above description merely provides some specific examples of some predictions that can be made by prediction module 126 using segment model 806.”; Paragraph 119, “In some implementations, the application configuration data properties 1030 can include a publisher threshold score. For example, each content publisher can be scored according to popularity, brand awareness, reputation, and/or other factors. Users associated with the new user engagement segment 428 may be more interested in well-known publishers and therefore configuration data associated with new user engagement segment 428 may have a high publisher threshold score (e.g., above a threshold value). Users associated with the high user engagement segment 442 may be more interested in a diverse selection of publishers and therefore configuration data associated with new high engagement segment 422 may have a low publisher threshold score (e.g., below a threshold value).”)
who will transition to another segment as time elapses (Ramachandran: Paragraph 61, “In some implementations, user engagement module 122 can analyze user event data 402 to determine user engagement segments 420 related to various stages of user engagement lifecycle 406. For example, the engagement lifecycle segments can include new user segment 428, activated user segment 430, adopted user segment 432, churned user segment 434, and/or reactivated user segment 436, low engaged user segment 438, medium engaged user segment 440, and highly engaged user segment 442. These user engagement lifecycle-based user engagement segments 420 (e.g., lifecycle stages) are defined further below with reference to FIG. 5.”; Paragraph 73, “In some implementations, user engagement module 122 can determine the level of engagement a user has with respect to news application 104 based on various engagement categories. For example, a score can be determined for each engagement category. The scores determined for each category can be combined (e.g., summed) to generate an engagement level score. The engagement level score can be compared to threshold values for each engagement level (e.g., high, medium, low, etc.) to determine with which engagement level the user should be associated.”; Paragraph 74, “In some implementations, user engagement module 122 can determine the level of engagement a user has with respect to news application 104 based on a recency engagement category. For example, a score generated or determined for the recency engagement category can correspond to a number of days since the user last viewed a content item, as determined by user engagement module 122 based on user event data 402. For example, if the recency category is scored 1-5, then a score of 5 can correspond to 1 day since the user last viewed a content item. A score of 4 can correspond to 2 days since the user last viewed a content item. A score of 3 can correspond to 3-6 days since the user last viewed a content item. A score of 2 can correspond to 7-12 days since the user last viewed a content item. A score of 1 can correspond to greater than 12 days since the user last viewed a content item. These scores are merely one example of how scores can be generated for the recency engagement category; other scoring schemes can be used and/or implemented.”; Paragraph 111, “When user event data 802 is processed by prediction module 126 using segment model 806, prediction module 126 can generate a prediction score 808. For example, the prediction score can indicate the likelihood that the user will perform the user engagement segment transition modeled, or predicted by, segment model 806. For example, prediction score 808 can generate a score that indicates the likelihood that the user is likely to adopt, likely to migrate up, likely to migrate down, and/or likely to churn. When the prediction score is above a threshold value (e.g., 9 on a scale of 1-10), then prediction module 126 can determine that the user is associated with the likely to adopt 444, likely to migrate up 446, likely to migrate down 448, and/or likely to churn 450 user engagement segments. As described above, other predictive models can be generated and other predictions can be made by prediction module 126 using the predictive models. The above description merely provides some specific examples of some predictions that can be made by prediction module 126 using segment model 806.”; Paragraph 119, “In some implementations, the application configuration data properties 1030 can include a publisher threshold score. For example, each content publisher can be scored according to popularity, brand awareness, reputation, and/or other factors. Users associated with the new user engagement segment 428 may be more interested in well-known publishers and therefore configuration data associated with new user engagement segment 428 may have a high publisher threshold score (e.g., above a threshold value). Users associated with the high user engagement segment 442 may be more interested in a diverse selection of publishers and therefore configuration data associated with new high engagement segment 422 may have a low publisher threshold score (e.g., below a threshold value).”)
Beck discloses a method of identifying customer opportunities from predicted behavior data. Ramachandran discloses a method for categorizing users into segments and predicting their transitions. At the time of Applicant's filed invention, one of ordinary skill in the art would have deemed it obvious to combine the methods of Beck with the teachings of Ramachandran in order to improve the efficiency and engagement of users as disclosed by Ramachandran (Ramachandran: Paragraph 6, “By reevaluating user event data and user engagement segments over time, the application configuration and content presented can evolve with the interests and behaviors of the user to keep the user engaged with the application.”)
Claim(s) 17 –
Beck in view of Ramachandran discloses the limitations of claims 16
Beck further discloses the following:
outputting a target group from at least one of the segments, wherein the target group includes one or more transitional users each having the one or more transition attributes, and one or more similar users each having one or more attributes similar to the transition attributes; (Beck: Paragraph 102, "In at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320) may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity)."; Paragraph 103, "This phase of the methodology can also act as a diagnostic tool for individual companies, to determine a strategy for either capturing new value or preventing value erosion by addressing the key identified and quantified frustrations causing attrition. The predictive analysis and remedial measures may come from either the firm itself or from major competitors projected to lose the most value in a particular industry.")
forecasting a transition performance coefficient indicating economic performance when transition from one of the segments to another segment occurs; and (Beck: Paragraph 102, "In at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320) may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity)."; Paragraph 104, "In at least one embodiment, the Vulnerability Study's final output is a rank list of frustrations across the industry and company, a comparative ranking of most vulnerable firms, projected attrition, business value at risk for firms, and potential capturers of that value being lost by most vulnerable firms."; Paragraph 105, "FIG. 4 illustrates Industry Analysis Output based on the Customer Attrition process applied in accordance with at least one embodiment for the retail banking industry. The table in FIG. 4 illustrates the industry-level output for several retail banks 410. It shows ranking of each bank in the retail banking industry, projected losses (financial and in the customer base), projected gains based on switching indicators, and banking-specific revenue to the deposit factors."; Paragraph 108, "FIG. 7 illustrates an alternative Competitive Vulnerability process in accordance with at least one embodiment. This diagram indicates how the total population 710 is cut down to determine the Vulnerable population 720, is the portion that is frustrated 720. From it, the system and method determines the Switching subpopulation 730. From the Switching subpopulation 730, the Value at Risk 740 is calculated. The Value at Risk 740 calculation may include predicting the actual losses of customers or employees by a company or an overall industry in accordance with at least one embodiment. The Value at Risk 740 may also include quantified financial losses to the company or losses in value shift of employees or customers to a competitor. It also indicate the overall quantified predicted losses for the overall industry.")
executing one of the classifying, the inferring, the outputting, and the forecasting with the one or more computers, and then feeding back a result of the execution to one or more remaining ones of the classifying, the inferring, the outputting, and the forecasting. (Beck: Paragraph 58, "In another embodiment, the areas of value creation for employees may include the following factors: (1) Perks/benefits; (2) Timeliness of requests; (3) Career progression; ( 4) Transparency into feedback; (5) Decision empowerment; ( 6) Ease of access to answers/knowledge; (7) Ethical/honesty of the company; and (8) Perceived fairness."; Paragraph 60, "The Voice of the Customer Data, 113, may also be utilized and processed in at least one embodiment, particularly when such data is available from clients to leverage their existing customer or employee feedback to assess areas of value creation and value degradation. The data is collected from proprietary sources within the organization such as CRM systems and is electronically processed and evaluated by a computerized system of the present invention."; Paragraph 102, "In at least one embodiment, based on the self-reported value created for the business, two of the three previously identified segments (High Risk of Attrition 310 and Medium Risk of Attrition 320) may be utilized and evaluated as the financial value risk for an individual company or provider. This step may also determine and calculate the total amount of revenue or value shifting amongst players in an industry. In other words, it may quantify and evaluate whether a particular financial value shift is "up for grabs" for the other competitors (at the expense of the losing entity)."; Paragraph 104, "In at least one embodiment, the Vulnerability Study's final output is a rank list of frustrations across the industry and company, a comparative ranking of most vulnerable firms, projected attrition, business value at risk for firms, and potential capturers of that value being lost by most vulnerable firms."; Paragraph 105, "FIG. 4 illustrates Industry Analysis Output based on the Customer Attrition process applied in accordance with at least one embodiment for the retail banking industry. The table in FIG. 4 illustrates the industry-level output for several retail banks 410. It shows ranking of each bank in the retail banking industry, projected losses (financial and in the customer base), projected gains based on switching indicators, and banking-specific revenue to the deposit factors."; Paragraph 108, "FIG. 7 illustrates an alternative Competitive Vulnerability process in accordance with at least one embodiment. This diagram indicates how the total population 710 is cut down to determine the Vulnerable population 720, is the portion that is frustrated 720. From it, the system and method determines the Switching subpopulation 730. From the Switching subpopulation 730, the Value at Risk 740 is calculated. The Value at Risk 740 calculation may include predicting the actual losses of customers or employees by a company or an overall industry in accordance with at least one embodiment. The Value at Risk 740 may also include quantified financial losses to the company or losses in value shift of employees or customers to a competitor. It also indicate the overall quantified predicted losses for the overall industry.")
Claim(s) 7, 11, and 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Beck (US 2021/0089979 A1) in view of Ramachandran (US 2019/0373071 A1) and Beharie (US 2023/0419345 A1)
Claim(s) 7 –
Beck in view of Ramachandran disclose the limitations of claim 1
Beck in view of Ramachandran does not explicitly disclose the limitation below, however, in analogous art of behavior prediction, Beharie discloses the following:
wherein the importance of each of the attributes in the prediction is calculated as a SHapley Additive exPlanations (SHAP) value. (Beharie: Paragraph 20, "In one or more embodiments, the explanatory algorithm may include at least one of a Local Interpretable Model-Agnostic Explanation algorithm or a SHapley Additive exPlanations (SHAP) algorithm."; Paragraph 242, "Attribution models 438 may be generated and may be used to isolate the effect of single channels where multiple channels are in use. The attribution models 438 may also include attribution models for isolating the effect of single actions when many engagement actions with an HCP may exist. For example, attribution models 438 may isolate the effect of a channel of marketing data where multiple channels serve ads simultaneously and where a channel of marketing data could include email, phone calls, social media, television and websites. Attribution models 438 may give attribution to single action only or to multiple actions. Attribution models 438 may use Shapley Value-based Attribution, Modified Shapley ValueBased Attribution, Markov Attribution, CIU, Counterfactuals, and the like.")
Beck discloses a method of identifying customer opportunities from predicted behavior data. Ramachandran discloses a method for categorizing users into segments and predicting their transitions. Beharie discloses a method for using machine learning to derive reasons for various predicted behaviors. At the time of Applicant's filed invention, one of ordinary skill in the art would have deemed it obvious to combine the methods of Beck in view of Ramachandran with the teachings of Beharie in order to improve the ability to gain insight to take more efficient actions as disclosed by Beharie (Beharie: Paragraph 4, "They allow businesses to learn more about their target audiences and how to best cater for their needs, thus retaining customers and driving sales growth. CRM systems may be used with past, present or potential customers.")
Claim(s) 11 –
Beck in view of Ramachandran disclose the limitations of claims 1 and 8-10
Beck in view of Ramachandran does not explicitly disclose the limitation below, however, in analogous art of behavior prediction, Beharie discloses the following:
wherein the extracting process includes a converting process for converting attributes of the one or more transitional users into vector representations, and the extracting process extracts the one or more similar users based on similarity of the vector representations. (Beharie: Paragraph 254, "A segment memberships look-alike recommendation 412 may be generated using data from historical micro-segments and static segments 410. A segment memberships look-alike recommendation 412 may be used to find a set of users that are similar in both static and dynamic features to a given set of users. A segment memberships look-alike recommendation 412 may be generated with access only to user attributes and contextual data, and no access to behavioral data of the users. For example, given the membership data of young growers in one population, a segment memberships look-alike recommendation 412 may find matching young growers in a different population."; Paragraph 255, "The segment memberships look-alike recommendation 412 may be generated through a process involving feature generation 490, followed by vector generation 492, followed by distance measurement 494a and/or semi-supervised learning 494b. The output of this process may be look-alike segments 496. Feature selection 490 may use statistical methods such as SHAP and LIME. Vector generation 492 may use methods such as embeddings. Distance measurement 494a may use methods such as NN-Search, SCANN and FAISS. Semi-supervised learning 494b may use methods such as PU Learning."; Paragraph 384, "The binary classification package 3704 receives long term trend data 3700. For a given user, the binary classification package 3704 may identify a change point and the estimate sequences that took place in a predetermined window around the change point. The estimate sequences may be estimate sequences 2704a (see FIGS. 27-35). The binary classification package 3704 may use a classifier, such as a random forest classifier, to identify a desired sequence. The desired sequence may be converted into a vector using a vector-conversion method such as SGT or count vectorizer. The output of the binary classification package 3704 may be sent to the attribution model 3706."; Paragraph 428, "The predictive user model training task 1616 generates a lookalike model 1618 for use in predicting a set of users that are similar in both static and dynamic features to a given set of subjects. The lookalike model 1618 may be validated 1620 using a split of one or more data sets. The split may be 80/20. The model validation 1620 may generate an evaluation file 1610 that describes the quality of the generated model 1618.")
Beck discloses a method of identifying customer opportunities from predicted behavior data. Ramachandran discloses a method for categorizing users into segments and predicting their transitions. Beharie discloses a method for using machine learning to derive reasons for various predicted behaviors. At the time of Applicant's filed invention, one of ordinary skill in the art would have deemed it obvious to combine the methods of Beck in view of Ramachandran with the teachings of Beharie in order to improve the ability to gain insight to take more efficient actions as disclosed by Beharie (Beharie: Paragraph 4, "They allow businesses to learn more about their target audiences and how to best cater for their needs, thus retaining customers and driving sales growth. CRM systems may be used with past, present or potential customers.")
Claim(s) 13 –
Beck in view of Ramachandran disclose the limitations of claims 1 and 12
Beck in view of Ramachandran does not explicitly disclose the limitation below, however, in analogous art of behavior prediction, Beharie discloses the following:
wherein the forecasting process includes inputting input data to a further learning model, the input data includes first input data and second input data,(Beharie: Paragraph 232, "The one or more embeddings 426 may be determined from the data sets by one or more machine learning models, include a neural network. Embeddings 426 may include vectors created from categorical features that are then used to train prediction models. For example, a location embedding may be used to replace a categorical feature such as a postal code with a four-dimensional vector."; Paragraph 242, "Attribution models 438 may be generated and may be used to isolate the effect of single channels where multiple channels are in use. The attribution models 438 may also include attribution models for isolating the effect of single actions when many engagement actions with an HCP may exist. For example, attribution models 43 8 may isolate the effect of a channel of marketing data where multiple channels serve ads simultaneously and where a channel of marketing data could include e-mail, phone calls, social media, television and websites. Attribution models 438 may give attribution to single action only or to multiple actions. Attribution models 438 may use Shapley Value-based Attribution, Modified Shapley Value-Based Attribution, Markov Attribution, CIU, Counterfactuals, and the like.")
the first input data includes a number of transitioning users transitioned through a certain transition path during a certain period, (Beharie: Paragraph 360, "The loyalty labelling output table 1406 includes examples of loyalty labels, associated trend types and associated metrics. Loyalty labels may include loyalists, churners, shrinking practice, growing practice, shrinking practice and loyalist, and growing practice and churner. The loyaltyassociated trend type may include short term or long term. The loyaltyassociated metric may include total prescription volume and new to brand prescriptions."; Paragraph 361, "The channel type labelling output table 1408 includes examples of channel type labels, attribution labels, associated trend types, associated metrics, associated objective values, secondary channel labels and tertiary channel labels. The channel type labelling output table 1508 may be a mapping table to categorize all generated labels so that the system can locate the labels for a specific capability, insight or calculation."; Paragraph 376, "An event 2602 may include an action such as a call 2602a, an advertisement 2602b, an e-mail 2602c or an outcome such as a prescription 2602d. Other events 2602 may include a learning program, a face-to-face meeting, a sample drop and a lunch and learn. The causal window estimation output 2404 may be a period of time such as 3 months. At 2406, a sequence of sales and marketing actions and physician prescriptions and a trendline depicting a metric per month are shown. The metric may include volume, share and decile."; Paragraph 448, "The segmentation evaluation diagram 4300 may be for instrumentation of the switcher function 4210 and may report on the number of entities (for example, HCPs) that have shown a trend to switch in one direction (see FIG. 43: from a competing productive to the objective brand) or in an opposite direction (see FIG. 44: from the objective brand to a competing brand)."; Paragraph 449, "In FIGS. 45 and 46, another report is shown identifying the number of entities (HCPs) who are shown to continue their trend of switching from one product set (FIG. 45) or shown to reverse their direction of switch behaviour (FIG. 46).")
the second input data includes a value indicating a quality of the transitioning users in the certain transition path, and the (Beharie: Paragraph 360, "The loyalty labelling output table 1406 includes examples of loyalty labels, associated trend types and associated metrics. Loyalty labels may include loyalists, churners, shrinking practice, growing practice, shrinking practice and loyalist, and growing practice and churner. The loyalty-associated trend type may include short term or long term. The loyalty-associated metric may include total prescription volume and new to brand prescriptions."; Paragraph 361, "The channel type labelling output table 1408 includes examples of channel type labels, attribution labels, associated trend types, associated metrics, associated objective values, secondary channel labels and tertiary channel labels. The channel type labelling output table 1508 may be a mapping table to categorize all generated labels so that the system can locate the labels for a specific capability, insight or calculation."; Paragraph 376, "An event 2602 may include an action such as a call 2602a, an advertisement 2602b, an e-mail 2602c or an outcome such as a prescription 2602d. Other events 2602 may include a learning program, a face-to-face meeting, a sample drop and a lunch and learn. The causal window estimation output 2404 may be a period of time such as 3 months. At 2406, a sequence of sales and marketing actions and physician prescriptions and a trendline depicting a metric per month are shown. The metric may include volume, share and decile."; Paragraph 448, "The segmentation evaluation diagram 4300 may be for instrumentation of the switcher function 4210 and may report on the number of entities (for example, HCPs) that have shown a trend to switch in one direction (see FIG. 43: from a competing productive to the objective brand) or in an opposite direction (see FIG. 44: from the objective brand to a competing brand)."; Paragraph 449, "In FIGS. 45 and 46, another report is shown identifying the number of entities (HCPs) who are shown to continue their trend of switching from one product set (FIG. 45) or shown to reverse their direction of switch behaviour (FIG. 46).")
learning model outputs the transition performance coefficient of the certain transition path when the input data is input. (Beharie: Paragraph 444, "The rising star function 4216 may identify (for example, HCPs) who currently have a small market but which are likely to grow to a bigger market within a future time period ( e.g. 2 years). The identification may include predicting if total market (total prescriptions for product a and for product b) grows by at least double (or another factor) compared to data in a prior period. The predicted total market (total prescriptions for product a and for product b) is at least more than the median predicted total market of all subjects (HCPs)."; Paragraph 546, "The entity ranking panel 5102 may display all relevant entities (audience members, or subjects in this case) ranked based on their score. The information displayed for each entity may include a name, user segments, a photograph, history of engagement, a numerical or categorical score and explanatory factors of the entity's score. For example, the entity ranking panel 5102 may display a list of health care providers and their audience score out of ten, including audience member or subject 5110. Audience member or subject 5110 may include the subject's name, history of engagement, user segments, and an audience score of 10. The audience score may be generated by the next best audience container 1704 (see FIG. 17) or the next best audience task 1912 (see FIG. 19). At the top of the entity ranking panel 5102 there may be provided functionalities to search for an entity, filter the ranked list of entities and download the ranked list of entities. The user may select an entity from entity ranking panel 5102.")
Beck discloses a method of identifying customer opportunities from predicted behavior data. Ramachandran discloses a method for categorizing users into segments and predicting their transitions. Beharie discloses a method for using machine learning to derive reasons for various predicted behaviors. At the time of Applicant's filed invention, one of ordinary skill in the art would have deemed it obvious to combine the methods of Beck in view of Ramachandran with the teachings of Beharie in order to improve the ability to gain insight to take more efficient actions as disclosed by Beharie (Beharie: Paragraph 4, "They allow businesses to learn more about their target audiences and how to best cater for their needs, thus retaining customers and driving sales growth. CRM systems may be used with past, present or potential customers.")
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Philip N Warner whose telephone number is (571)270-7407. The examiner can normally be reached Monday-Friday 7am-4:00pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jerry O’Connor can be reached at 571-272-6787. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Philip N Warner/Examiner, Art Unit 3624
/Jerry O'Connor/Supervisory Patent Examiner,Group Art Unit 3624
Prosecution Insights