US 8,209,218 B1Grant
Apparatus, system and method for processing, analyzing or displaying data related to performance metrics
Issue Date:2012-06-26
•19 Claims
•23 Drawing Sheets
Abstract
Embodiments of the present invention relates to a method, apparatus and computer program product for processing, analyzing and/or display data related to performance metrics, for example, predicted performance metrics. In some embodiments, the apparatus includes one or more computers, a user input device(s), and a display device(s). In some embodiments, the one or more computers are operative, to compute, from historical KPI-related data, predicted future KPI-related behavior. In some embodiments, the display device is operative to display indications of predicted KPI values and/or predicted influencer values and/or a predicted future influencer-ranking-hierarchy and/or calculated deviating influencer values and/or one or more of the deviating future KPI values. In some embodiments, one or more user input devices are configured to receive indications of user-proposed influencer values and/or any other relevant data.
Metadata
Assignee
- DataInfoCom Inc.
Inventors
- Atanu Basu
- Joydeep Biswas
- Avijit Roy
- Bappaditya Mondal
Application Information
Application Number:US 12/404,568
Filing Date:2009-03-16
Priority Date:2008-03-14
Art Unit:3623
Classifications
IPC:
G06Q10/10
Patent Drawings (23 sheets)
Description
Cross-Reference to Related Applications
[0001] This patent application claims the benefit of U.S. Provisional Patent Application No. 61/036,483, filed on Mar. 14, 2008.
Field of the Invention
[0002] Embodiments relate to apparatus and methods for computing and displaying data related to predicted KPI values.
Background and Related Art
[0003] Global Fortune 1000 corporations and many smaller businesses manage their business processes by carefully monitoring a set of key performance indicators (KPIs) or metrics, at different time intervals (from near-real-time to daily/weekly/monthly/quarterly/etc., depending on the metric in question). These metrics fall under several categories, depending on the particular business process. Some of these categories include but are not limited to productivity, customer satisfaction, customer loyalty, problem resolution, cost, revenue, sales conversion rate, timeliness, etc.
[0004] The categories of business processes for which information about predicted KPI values is useful, include but are not limited to, contact/call centers (e.g., customer service), transaction services (e.g., insurance claims), finance & accounting (e.g., accounts payable/receivable), manufacturing (e.g., supply and demand chains), field service, marketing, logistics, sales, etc.
Summary of Embodiments
[0005] The present disclosure relates to tools and techniques for computing data related to predicted KPI values, and to novel user interfaces (for example, interactive user interfaces) for displaying information related to predicted KPI values. The following published patent documents provide potentially relevant background art, and are each incorporated herein by reference in their entirety: U.S. Pat. No. 7,373,633, U.S. Pat. No. 6,876,988, US 20090064025, US 20060010164, US 20060242035, US 20060242033, US 20080195440, US 20080183786 and US 20080201397. All cited documents are incorporated herein by reference in their entirety.
[0006] Although the apparatus and methods disclosed herein apply to any set of KPIs, certain techniques will be explained, throughout the document, with reference to a non-limiting Use Case 1 associated with KPIs and influencers (explained below).
[0007] In the present section, two non-limiting Use Cases are described. The skilled artisan will appreciate that these use cases are presented as specific examples, and may be modified. The skilled artisan will also appreciate that these use cases are not intended as limiting whatsoever, and are presented only as a small set of examples to which presently-disclosed techniques may be applied.
[0008] The appearance of material in this background section should not be construed as an admission that this material is prior art.
Use Case 1: Contact Center
[0009] SearchCRM.com defines a contact center as follows:
[0010] “A contact center (also referred to as a customer interaction center or e-contact center) is a central point in an enterprise from which all customer contacts are managed. The contact center typically includes one or more online call centers but may include other types of customer contact as well, including e-mail newsletters, postal mail catalogs, Web site inquiries and chats, and the collection of information from customers during in-store purchasing.”
[0011] The same source, SearchCRM.com, defines a call center as follows:
[0012] “A call center is a physical place where customer and other telephone calls are handled by an organization, usually with some amount of computer automation. Typically, a call center has the ability to handle a considerable volume of calls at the same time, to screen calls and forward them to someone qualified to handle them, and to log calls. Call centers are used by mail-order catalog organizations, telemarketing companies, computer product help desks, and any large organization that uses the telephone to sell or service products and services.”
[0013] For a firm, a contact center handles both inbound and outbound communication with its customers and other constituencies.
[0014] The present use case (i.e., Use Case 1) relates on a Technical Support Call Center as an example of a scenario where there is a need for tools that compute and display predictions of KPI-related data (for example, interactive tools). In the present use case, the customers call the manufacturer, whose products are serviced by the Technical Support Call Center. When these customers call the manufacturer, the customers receive help for the technical problems they are experiencing with the products that they have purchased from this manufacturer.
[0015] A Discussion of Contact Center ‘Raw’ Data—Before discussing various KPIs for this use case, a high level view of ‘raw’ data available from this particular Technical Support Call Center is presented. As will be discussed below, the relevant KPIs for this use case may be derived from this ‘raw data.’
[0016] In the present example, the pertinent ‘raw’ data from this contact center includes, but is not limited to: (A) phone switch data; (B) customer relationship management (CRM) data; (C) dispatch data; (D) survey data; (E) human resources (and/or payroll) data; (F) marketing data; (G) market data; and (H) workforce management data.
(A) Phone Switch Data
[0017] Every time a customer calls this Technical Support Call center, a new record is automatically (i) captured in the phone switch (e.g. an Automatic Call Distributor) and (ii) stored in the phone switch database. The call first waits in the phone switch wait queue. If the call waits too long in the queue the customer may hang up and thus the call abandonment occurs. If the call succeeds, a few fields regarding the call are captured. Information about both the successful and abandoned calls are recorded. Example fields recorded include:
-
- [0018] (i) Talk Time—How long the call was in a talking state.
- [0019] (ii) Hold Time—How long the call was on hold.
- [0020] (iii) Call Duration—How long it took for the complete processing of the call.
- [0021] (iv) Work Time—Represents the after-call work time associated with the call.
After-call work includes post-call activities such as completing paperwork or consulting with associates, etc.
[0022] Sometimes the data generated in the phone switch is transformed to generate new fields/datasets. For example, the call-by-call data can be clubbed (through averaging, summation, or other transformations) in a half-hourly or hourly granularity to form time-series metrics. Examples for half-hourly granularity include:
-
- [0023] (i) Average Queue Time—Average delay in the queue for calls to the service during the half-hour interval
- [0024] (ii) Call Abandonment Rate—Number of calls abandoned in queue for the service during the half-hour interval
- [0025] (iii) Service Level—% of calls belonging to a half-hour interval, answered within a time, e.g. 30 Seconds
(B) Customer Relationship Management (CRM) Data
[0026] The technical Support Call Center may use CRM software to store and capture information about customers, interactions, issues, and products. Each time a customer has a contact with this call center regarding a problem (old or new) in the customer's system, the agent creates a new entry about the issue in the CRM software application.
(C) Dispatch Data
[0027] This Technical Support Call Center stores detailed information about each part dispatch, corresponding to a part replacement in the customers system.
(D) Survey Data
[0028] To know the customers' views on service and products, contact centers sometimes organize for customer feedback programs. Survey methods include Email, Post-Call IVR (i.e., Interactive Voice Response), Mail, Telephone, etc. This Technical Support Call Center conducts email surveys to get the customers' perspectives on the service being provided by this Call Center. Examples of the data these surveys gather include:
-
- [0029] (i) Customer Satisfaction: Customer's overall satisfaction with the call to the Technical Support
- [0030] (ii) Agent Satisfaction: Customer's satisfaction with the agent's communication skill, expertise in problem troubleshooting and problem resolution, ownership of the issue, etc.
- [0031] (iii) Problem resolution: Whether the customer considers the problem to be resolved.
- [0032] (iv) Brand Satisfaction: Satisfaction with the manufacturer's brand.
- [0033] (v) Repurchase Likelihood: How likely would the customer be to repurchase from the same manufacturer.
- [0034] (vi) Recommend Likelihood: How likely would the customer be to recommend to a friend or colleague the manufacturer from which the customer bought his system.
- [0035] (vii) Number of Contacts: The number of times the customer had to contact this Technical Support Call Center to get the same problem resolved completely.
- [0036] (viii) First Contact Resolution: Whether the customer's problem was resolved in the first contact to this Technical Support Call Center. This answer is typically derived from Problem Resolution and Number of Contacts.
(E) HR (and/or Payroll) Data
[0037] This data source contains information about the agents and a few items can be derived from it such as:
-
- [0038] (i) Agent Tenure: Denotes how long the agent has been in this center/profession.
- [0039] (ii) Agent Salary: Agents hourly rate of pay and benefits.
- [0040] (iii) Agent Education Level: The number of years of schooling the agent has.
(F) Marketing Data
[0041] Such as data on discounts, promotions, advertising campaigns, etc.
(G) Market Data
[0042] Includes data on competitive products, benchmarks, econometrics, etc.
(H) Workforce Management (WFM) Data
[0043] Workforce Management software is employed by this Technical Support Call Center to plan, schedule, and manage the staffing of agents to effectively handle inbound call volume, in coordination with the phone switch. Example of WFM data includes Schedule Adherence and Occupancy.
A Discussion of KPIs for Use Case 1
[0044] The contact centers often categorize their KPIs into different groups. This particular Technical Support Call Center uses the following major groupings: Customer Experience & Loyalty KPIs, Resolution KPIs, Productivity KPIs, Dispatch KPIs, and Cost & Revenue KPIs.
[0045] (A) Customer Experience & Loyalty KPIs—These are KPIs whose values are obtained from the customer feedback (i.e., email survey) system. The email survey for this Technical Support Call Center provides the customers with a scale system (such as 9-point, 5-point, 3-point) to respond to the questions in this group. These KPIs are typically categorical (ordinal) response variables in nature. Examples include:
-
- [0046] 1. Customer Satisfaction
- [0047] 2. Brand Satisfaction
- [0048] 3. Recommend Likelihood
- [0049] 4. Repurchase Likelihood
- [0050] 5. Agent Satisfaction
[0051] (B) Resolution KPIs—These KPIs are also usually derived from customer feedback (i.e., email surveys). For this Technical Support Call Center, they represent whether or not (or within how many contacts) a customer's particular problem was resolved. These KPIs are typically binary response variables in nature. Examples include:
-
- [0052] 1. Problem Resolution: Whether or not the problem was resolved.
- [0053] 2. Resolved-in-one-Contact (Ri1): Also called First Call Resolution. This KPI is derived from two other metrics as follows: it is ‘True’ if Problem Resolution is ‘True’ and the number of times the customer had to call for the same issue is 1.
- [0054] 3. Resolved-within-two-Contacts (Ri2): This is also a derived KPI. It is a ‘True’ if Problem Resolution is ‘True’ and the number of times the customer had to call for the same issue is <=2.
[0055] (C) Productivity KPIs—These KPIs measure contact center productivity. These KPIs are typically metrical response variables in nature and are mostly derived from the phone switch (i.e., ACD) and workforce management system. Examples include:
-
- [0056] 1. Abandonment Rate
- [0057] 2. Average Hold Time
- [0058] 3. Average Talk Time
- [0059] 4. Average After Call Work Time
- [0060] 5. Average Queue Time
- [0061] 6. Average Handle Time
- [0062] 7. Service Level
- [0063] 8. Occupancy
- [0064] 9. Schedule Adherence
[0065] (D) Dispatch KPIs—This Technical Support Call Center uses Dispatch KPIs to keep track of the parts being dispatched by its agents to resolve customers' issues. Examples of Dispatch KPIs include:
-
- [0066] 1. Repeat Dispatch Rate: Average rate of repeat dispatches within a specified time.
- [0067] 2. Material Value Per Dispatch: Average part cost per dispatch.
- [0068] 3. Dispatch Rate: Average number of dispatches per call.
- [0069] 4. Parts Per Dispatch: Average number of parts per dispatch.
[0070] (D) Cost & Revenue KPIs—Examples for this Technical Support Call Center include:
-
- [0071] 1. Cost Per Call: Total operational costs divided by the total number of calls for a given period of time.
- [0072] 2. Revenue Per Call: Total revenue divided by the total number of calls for a given period of time.
- [0073] 3. Sales Conversion Rate: Number of sales made relative to the number of attempts made.
[0074] As will be discussed throughout this disclosure, the KPIs may be influenced by one or more ‘influencers.’ For the present example, the influencers may include: Number of Contacts, Product Description, Brand Description, Reason Codes (i.e., issue description), Shipping Delay, Age of the System, Dispatch Sent, Hold Time, Local Queue Time, etc.
Use Case 2: Field Service
[0075] Field service can be defined as the maintenance and repair of equipment for a customer at the customer's location, according to warranty or other contractual agreements.
[0076] In this particular example for this use case, when a customer contacts a contact center for a service of his/her system, the contact center agent works with the customer to diagnose the problem. If the service requires a part to be replaced, the technician creates a dispatch order. The purpose of the dispatch is to send the part to a service provider to replace the failing component. When the dispatch order is processed and parts are available, the parts are sent to the service provider in the customer's location. A field technician from the service provider reaches the customer's site and replaces the parts and tests the customer's system for functionality. After the field technician completes the service, the service provider transmits notification back to the contact center and the service call is completed.
[0077] Feedback about the service provided is solicited from some customers who are serviced by the field delivery process. Information about a service call and corresponding feedback is accumulated in database records from which a set of KPIs and their influencers are identified.
Example Field Service KPIs Include:
-
- [0078] (i) Customer Satisfaction: Customer's overall satisfaction with the onsite service.
- [0079] (ii) On-Time Arrival Satisfaction: Customer's satisfaction regarding the on-time arrival of the technician at the customer's location.
- [0080] (iii) Professionalism Satisfaction: Customer's satisfaction regarding the field technician's professionalism.
- [0081] (iv) Expertise Satisfaction: Customer's satisfaction regarding the field technician's expertise to resolve the customer's problem
- [0082] (v) Service Time Satisfaction: Customer's satisfaction regarding the time required for the onsite service.
- [0083] (vi) Status Update Satisfaction: Customer's satisfaction regarding the field technician's effectiveness in updating the customer on the status of the customer's problem
- [0084] (vii) Problem Resolution: Customer's impression of whether his issue has been resolved by this technician visit to the customer's location
- [0085] (viii) Repurchase Likelihood: How likely would the customer be to repurchase from the company from which he bought his system.
- [0086] (ix) Recommend Likelihood: How likely would the customer be to recommend to a friend or colleague the company from which he bought his system.
- [0087] (x) Resolved-in-one (Ri1): It denotes the percentage of cases the customer's issue was resolved in one visit of the field technician to the customer's site.
- [0088] (xi) Resolved-within-two (Ri2): It denotes the percentage of cases the customer's issue was resolved within two visits by the field technicians to the customer's site.
[0089] Some examples of self-explanatory KPIs in field service include:
-
- [0090] 1. Time to close the call (i.e., from the creation to closure)
- [0091] 2. Length of onsite visit (i.e., time to service the equipment at the customer's site)
- [0092] 3. On-time-first-time fixes
- [0093] 4. On-time responses
- [0094] 5. First-time fixes
- [0095] 6. Open Calls (i.e., calls that are open after a predetermined period)
[0096] Examples of influencers for this example (i.e. Use Case 2) include:
-
- [0097] 1. Number of Onsite Visits
- [0098] 2. Issue Description
- [0099] 3. Tenure of the Technician
- [0100] 4. Certification Status of the Technician
- [0101] 5. Service Time (to resolve the customer issue onsite)
- [0102] 6. Metro Statistical Areas
- [0103] 7. Vendor Employing the Technician
- [0104] 8. Technician's Arrival Day and Time (at the customer site)
- [0105] 9. Deferred Status (of the customer visit)
- [0106] 10. Parts Delay, etc.
[0107] It is now disclosed for the first time a method of displaying KPI-related data on a display device in accordance with information received via a user input device, the display device and the user input device being operatively coupled to one or more digital computers. The method comprises the steps of:
-
- [0108] a) for each dashboard future time frame of a plurality of dashboard future time frames, for each KPI of a plurality of KPIs, displaying, on the display device, a respective indication of a respective predicted KPI value (in one non-limiting example, this includes displaying a predictive dashboard);
- [0109] b) for an influencer-ranking future time frame, for a root-cause KPI whose predicted value in the influencer-ranking future time frame is predicted to be influenced by behavior of a set of influencers, displaying, on the display device, a predicted future influencer ranking-hierarchy including identifiers of a first plurality of influencers of the set of influencers, the first plurality of influencers being displayed in accordance with predicted relative influence contributions to the predicted future value of the root-cause KPI in the influencer-ranking future time frame (in one non-limiting example, this relates to ‘root-cause analysis’);
- [0110] c) for each influencer of a second plurality of influencers of the set of influencers, receiving, via the user input device, a respective user proposed deviating influencer value; and
- [0111] d) in accordance with: i) an influencer action future time frame for user-proposed influencer values; ii) a deviating-KPI-viewing time frame for user-proposed influencers; and iii) the received user-proposed deviating influencer values, displaying, on the display device, for one or more of KPIs, descriptions of one or more deviating future KPI values derived from the user-proposed deviating influencer values (in one non-limiting example, step (c) includes receiving input for a ‘what-if interface’, and step (d) includes displaying output for a ‘what-if interface’).
[0112] In some embodiments, the first plurality and the second plurality of influencers are identical or are different.
[0113] In some embodiments, the identifiers of the influencer-ranking-hierarchy (for example, in ‘root-cause analysis’) are ordered and/or sized on the display device in accordance with the predicted relative influencer contributions to the predicted future value of the root-cause KPI for a future time frame.
[0114] In some embodiments, the first plurality of influencers (for example, used in root cause analysis) are selected from the set of influencers in accordance with predicted relative influence strength in the influencer-ranking future time frame.
[0115] In some embodiments (for example, some embodiments relating to ‘what-if analysis’), i) the method further comprises: e) for each given influencer of the second plurality of influencers, displaying on the display device a respective user control (for example, a slider or a text box or any other user control) that is visually associated with a respective identifier of the given influencer such that the user controls and the influencer identifiers are displayed in accordance with predicted relative influence contributions to one or more KPIs in a future time frame; and ii) for each given influencer, the respective receiving of the respective user-proposed deviating influencer value is carried out in response to a respective one or more user engagements of the respective displayed user controls using the user input device.
[0116] In some embodiments (for example, some embodiments relating to ‘what-if analysis’), the user controls and the influencer identifiers are sized and/or ordered on the display device in accordance with predicted relative influence contributions to one or more KPIs in a future time frame.
[0117] In some embodiments for example, some embodiments relating to ‘what-if analysis’, for a given influencer of the second plurality of influencers: i) the displaying of the corresponding influencer identifier for the given influencer by the display device; and ii) the displaying of the corresponding user control for the given influencer by the display device are contingent upon a relationship between: i) a magnitude of a predicted relative future influence, over the one or more KPIs, by the given influencer over the one or more KPIs; and ii) magnitudes of predicted relative future influences, over the one or more KPIs, by one or more other influencers, wherein: i) the one or more other influencers are members of the set of influencers; and ii) the one or more influencers do not belong to the second plurality of influencers. Thus, it is possible to order the influencer descriptions and/or user controls and/or GUI widgets by estimated contribution or importance—visual elements associated with the ‘more important’ influencers may be bigger or brighter or displayed earlier according to some visual ordering scheme.
[0118] In some embodiments (for example, related to ‘what-if’ analysis), i) the method further comprises: e) for each given influencer of the second plurality of influencers, displaying a respective user control that is visually associated with a respective identifier of the given influencer; and ii) for each given influencer, the respective receiving of the respective user-proposed deviating influencer value is carried out in response to a respective one or more user engagements of the respective displayed user controls using the user input device; and iii) the display device and the user input device are configured such that each user control is associated with a respective influencer which is associated with a default value that is determined in accordance with a respective predicted future value for the influencer-action time frame for the respective influencer.
[0119] In some embodiments (for example, related to ‘what-if’ analysis), i) KPI value descriptions for a plurality of deviating future KPI values derived from the user-proposed influencer values are displayed, each deviating future KPI value being mathematically associated with a respective deviation magnitude respectively describing by how much each said future KPI value derived from the user-proposed influencer values deviates from its respective predicted KPI value for a future deviating-KPI-viewing time frame; and ii) the displaying of the descriptions of KPI values derived from user-proposed influencer values is carried out in accordance with the deviation magnitudes.
[0120] In some embodiments (for example, related to ‘what-if’ analysis), i) KPI value descriptions for a plurality of deviating future KPI values derived from the user-proposed influencer values are displayed, each deviating future KPI value being mathematically associated with a respective deviation magnitude respectively describing by how much each said future KPI value derived from the user-proposed influencer values deviates from its respective predicted KPI value for a future deviating-KPI-viewing time frame; and ii) the display of a given displayed KPI value description derived from the user-proposed influencer values is contingent upon a respective deviation magnitude for the given displayed KPI value. Thus, in some embodiments, only GUI controls and/or information about ‘more important’ influencers may be presented for soliciting user-proposed influencer values.
[0121] In some embodiments (for example, in some embodiments related to optimization and/or the computer(s) suggesting preferred deviating influencer values), the method further comprises the step of: e) for a KPI-moving future time frame, one or more target KPIs, and an influencer-action-time frame for calculated influencers, displaying (for example, by selecting and/or sizing and/or ordering and/or font-selection of any visual element) on the display device, a set of calculated deviating influencer values which: i) satisfy one or more provided constraints; ii) deviate from predicted influencer values for the KPI-moving future time frame; and iii) if adopted at the influencer-action-time frame, are predicted to move a future value of the one or more target KPIs, at the KPI-moving future time frame, from predicted values to a target value or range.
[0122] In some embodiments (for example, in some embodiments related to optimization and/or the computer(s) suggesting preferred deviating influencer values), the method further comprises the step of: f) for a specified deviating KPI-viewing time frame for the calculated influencer values, the specified deviating KPI-viewing time frame for the calculated influencer values ending after the influencer-action-time frame for calculated influencers, displaying, on the display device, indications of one or more deviating future KPI values determined for: i) the calculated deviating influencers for the influencer-action time-frame of the calculated influencers; and ii) the deviating KPI-viewing time frame.
[0123] In some embodiments (for example, in some embodiments related to optimization and/or the computer(s) suggesting preferred deviating influencer values), i) KPI value descriptions for a plurality of deviating future KPI values determined for the calculated deviating influencer values are displayed, each deviating future KPI value determined for the calculated influencer values being mathematically associated with a respective deviation magnitude respectively describing by how much each said future KPI value determined for the calculated deviating influencer values deviates from its respective predicted KPI value for a future deviating-KPI-viewing time frame; and ii) the displaying of the descriptions of KPI values determined for the calculated influencer values is carried out in accordance with the deviation magnitudes.
[0124] In some embodiments (for example, in some embodiments related to optimization and/or the computer(s) suggesting preferred deviating influencer values), i) KPI value descriptions for a plurality of deviating future KPI values determined for the calculated influencer values are displayed, each deviating future KPI value determined for the calculated influencer values being mathematically associated with a respective deviation magnitude respectively describing by how much each said future KPI value determined for the calculated influencer values deviates from its respective predicted KPI value for a future deviating-KPI-viewing time frame; and ii) the display of a given displayed KPI value description determined from the calculated influencer values is contingent upon a respective deviation magnitude for the given displayed KPI value.
[0125] In some embodiments, the constraint includes at least one of a KPI constraint (i.e. a constraint involving one or more KPI(s)) and a cost constraint.
[0126] In some embodiments, at least one of: i) a specification of the root-cause KPI; ii) a specification of the influencer-ranking future time frame; iii) a specification of the influencer action future time frame; and iv) a specification of the deviating-KPI-viewing future time frame is received via one or more user input devices.
[0127] In some embodiments, at least one user input device is selected from the group consisting of a mouse, a keyboard, a touch-screen, and a microphone.
[0128] In some embodiments, at least one influencer and/or KPI is indicative of at least one of: i) a length of a telephone conversation; ii) an amount of time that a telephone device is in a hold-state; and iii) a number of distinct telephone conversations; and iv) whether or not a telephone conversation is an incoming conversation or an outgoing conversation.
[0129] It is now disclosed for the first time an apparatus for displaying KPI-related data, the apparatus comprising: a) one or more computers (for example, digital computers); b) a display device operatively coupled to one or more of the computers; and c) a user input device operatively coupled to one or more of the computers; wherein: (i) the one or more computers are operative, to compute, from historical KPI-related data, predicted future KPI-related behavior including: A) predicted behavior of one or more KPIs; B) predicted future behavior of one more influencers, and C) predicted future relationships between one or more of the influencers and one or more of the KPIs; (ii) the display device is operative to display, in accordance with the predicted future KPI-related behavior: A) for each dashboard future time frame of a plurality of dashboard future time frames, for each KPI of a plurality of KPIs, a respective indication of a respective predicted KPI value; B) for an influencer-ranking future time frame, for a root-cause KPI whose predicted value, is predicted, in accordance with the predicted future KPI-related behavior, to be influenced by behavior of a set of influencers, a predicted future influencer-ranking-hierarchy including identifiers of a first plurality of influencers of the set of influencers, the first plurality of influencers being displayed in accordance with predicted relative influence contributions to the predicted future value of the root-cause KPI in the influencer-ranking future time frame; iii) the user input device is operative, for each influencer of a second plurality of influencers of the set influencers, to receive, a respective user proposed deviating influencer value; iv) the one or more computers are further operative, to compute, from historical KPI-related data and from the user-proposed deviating influencer values, one or more deviating future KPI values; v) the display device is further operative, in accordance with: A) an influencer action future time frame for user-proposed influencer values; B) a deviating-KPI-viewing time frame for user-proposed influencers; and C) the computed deviating future KPI values, to display, on the display device, for one or more of KPIs, descriptions of one or more of the deviating future KPI values.
[0130] In some embodiments, the apparatus may be configured to effect calculations related to ‘optimization’ and/or display indications of calculated deviating influencer values.
[0131] It will be appreciated that in different embodiments, any presently-disclosed apparatus (or any presently disclosed computer product including a computer usable medium) may be configured and/or operative to any presently disclosed method or technique.
[0132] It is now disclosed for the first time a computer program product, comprising a computer usable medium having a computer readable program code embodied therein, said computer readable program code adapted to be executed to implement a method of computing and displaying KPI-related data on a display device in accordance with information received via a user input device, the display device and the user input device being operatively coupled to one or more digital computers, the method comprising: a) for each dashboard future time frame of a plurality of dashboard future time frames, for each KPI of a plurality of KPIs, displaying, on the display device, a respective indication of a respective predicted KPI value; b) for an influencer-ranking future time frame, for a root-cause KPI whose predicted value in the influencer-ranking future time frame is predicted to be influenced by behavior of a set of influencers, displaying, on the display device, a predicted future influencer-ranking-hierarchy including identifiers of a first plurality of influencers of the set of influencers, the first plurality of influencers being displayed in accordance with predicted relative influence contributions to the predicted future value of the root-cause KPI in the influencer-ranking future time frame; c) for each influencer of a second plurality of influencers of the set influencers, receiving, via the user input device, a respective user proposed deviating influencer value; and d) in accordance with: i) an influencer action future time frame for user-proposed influencer values; ii) a deviating-KPI-viewing time frame for user-proposed influencers; and iii) the received user-proposed deviating influencer values, displaying, on the display device, for one or more of KPIs, descriptions of one or more deviating future KPI values derived from the user-proposed deviating influencer values.
Brief Description of the Drawings
[0133] FIG. 1 is a block diagram of an exemplary computer operatively coupled to a display device and a user input device.
[0134] FIG. 2 is a flow chart of an exemplary technique for providing one or more interface(s) via which a user may interactively explore relationships between predicted KPI(s) and influencer(s).
[0135] FIG. 3 is a screen shot of an exemplary predictive dashboard for a plurality of KPIs and a plurality of future time-frames in accordance with some embodiments.
[0136] FIGS. 4A-4D are screen shots of a “ranked influencer” interface for a particular KPI (in this example “recommended likelihood”) and a particular future timeframe (in this example, next 7 days”), in accordance with some embodiments.
[0137] FIGS. 5A-5C are screen shots of an exemplary “optimization” interface for displaying a set of calculated deviating influencer values which are predicted to move one or more KPI(s) from predicted values to target values or ranges, in accordance with some embodiments.
[0138] FIG. 5D is a flow chart of an exemplary routine for providing an interface whereby a user specifies one or more KPI(s) and a future time frame and one or more calculated deviating influencer values are displayed.
[0139] FIGS. 6A-6B are time lines illustrating relationships between a time period for action for deviating influencers, a deviating KPI display time frame, and a KPI-moving future time frame.
[0140] FIGS. 7A-7B are screen shots of a “what-if” interface including both plurality of user-controls via which the user specifies one or more user-suggested deviating influencers and KPI display area where indications of values of one or more affected KPIs are displayed.
[0141] FIGS. 7C-7E are flow charts of exemplary routines for providing an ‘what-if’ interface.
[0142] FIG. 8 is a block diagram of multiple interfaces and describes an exemplary technique for navigating between these interfaces.
[0143] FIGS. 9A-9D are flowcharts of computational techniques for generating data related to KPI predictions.
[0144] FIG. 10 is a block diagram illustrating one embodiment of a topology which may be used in conjunction with an implementation of embodiments of the present invention.
[0145] FIG. 11 is a block diagram illustrating one embodiment of a performance metric analysis system.
Detailed Description of Embodiments
[0146] Reference is now made in detail to the exemplary embodiments of the invention, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts (elements). The description, embodiments and figures are not to be taken as limiting the scope of the claims. It should be understood that not every feature of the presently disclosed methods and apparatuses is necessary in every implementation. It should also be understood that throughout this disclosure, where a process or method is shown or described, the steps of the method may be performed in any order or simultaneously, unless it is clear from the context that one step depends on another being performed first. As used throughout this application, the word “may” is used in a permissive sense (i.e., meaning “having the potential to’), rather than the mandatory sense (i.e. meaning “must”).
[0147] Before discussing specific embodiments, embodiments of a hardware architecture for implementing certain embodiments is described herein. One embodiment can include one or more computers communicatively coupled to a network. As is known to those skilled in the art, the computer can include a central processing unit (“CPU”), at least one read-only memory (“ROM”), at least one random access memory (“RAM”), at least one hard drive (“HD”), and one or more input/output (“I/O”) device(s). The I/O devices can include a keyboard, monitor, printer, electronic pointing device (such as a mouse, trackball, stylist, etc.), or the like. In various embodiments, the computer has access to at least one database over the network.
[0148] ROM, RAM, and HD are computer memories for storing computer-executable instructions executable by the CPU. Within this disclosure, the term “computer-readable medium” is not limited to ROM, RAM, and HD and can include any type of data storage medium that can be read by a processor. In some embodiments, a computer-readable medium may refer to a data cartridge, a data backup magnetic tape, a floppy diskette, a flash memory drive, an optical data storage drive, a CD-ROM, ROM, RAM, HD, or the like.
[0149] At least portions of the functionalities or processes described herein can be implemented in suitable computer-executable instructions. The computer-executable instructions may be stored as software code components or modules on one or more computer readable media (such as non-volatile memories, volatile memories, DASD arrays, magnetic tapes, floppy diskettes, hard drives, optical storage devices, etc. or any other appropriate computer-readable medium or storage device). In one embodiment, the computer-executable instructions may include lines of complied C++, Java, HTML, or any other programming or scripting code.
[0150] Additionally, the functions of the disclosed embodiments may be implemented on one computer or shared/distributed among two or more computers in or across a network. Communications between computers implementing embodiments can be accomplished using any electronic, optical, radio frequency signals, or other suitable methods and tools of communication in compliance with known network protocols.
[0151] As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having” or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a process, article, or apparatus that comprises a list of elements is not necessarily limited only those elements but may include other elements not expressly listed or inherent to such process, article, or apparatus. Further, unless expressly stated to the contrary, “or” refers to an inclusive or and not to an exclusive or. For example, a condition A or B is satisfied by any one of the following: A is true (or present) and B is false (or not present), A is false (or not present) and B is true (or present), and both A and B are true (or present).
[0152] Additionally, any examples or illustrations given herein are not to be regarded in any way as restrictions on, limits to, or express definitions of, any term or terms with which they are utilized. Instead, these examples or illustrations are to be regarded as being described with respect to one particular embodiment and as illustrative only. Those of ordinary skill in the art will appreciate that any term or terms with which these examples or illustrations are utilized will encompass other embodiments which may or may not be given therewith or elsewhere in the specification and all such embodiments are intended to be included within the scope of that term or terms. Language designating such nonlimiting examples and illustrations includes, but is not limited to: “for example”, “for instance”, “e.g.”, “in one embodiment”.
Introductory Remarks
[0153] Embodiments of the present invention provide an apparatus and method whereby even a user who is relatively untrained in mathematics can interactively and intuitively (i) investigate predicted future behavior of KPIs, (ii) investigate what are predicted to be the causes of this predicted behavior; and (iii) probe hypothetical scenarios whereby, in response to modified organizational behavior, the predicted KPI “outcomes” change.
[0154] Towards this end, as will be discussed below with reference to FIG. 2 , a tool is disclosed whereby the user may investigate answers to the following four questions:
-
- [0155] 1. What will happen, when? (i.e., by using a ‘predictive dashboard’) (Question 1 or Q1);
- [0156] 2. (ii) Why will it happen? (i.e., by using a ‘root cause’ interface); (Question 2 or Q2);
- [0157] 3. If something bad (or good) is predicted to happen, how can we avoid (or benefit from) it? (i.e., by using an ‘optimization’ interface) (Question 3 or Q3);
- [0158] 4. How will user-suggested changes affect the future? (i.e., by using a ‘what-if’ interface) (Question 4 or Q4).
Definitions
[0159] For convenience, in the context of the description herein, various terms are presented here. To the extent that definitions are provided, explicitly or implicitly, here or elsewhere in this application, such definitions are understood to be consistent with the usage of the defined terms by those of skill in the pertinent art(s). Furthermore, such definitions are to be construed in the broadest possible sense consistent with such usage.
[0160] A “key performance indicator” (KPI) is a metric describing performance of a business (or business process) during a period of time.
[0161] A “future time frame” is a period of time (i.e. having a starting time and an ending time) where at least the ‘ending time’ is in the future. If a time frame A is defined as an ‘earlier time frame’ than time frame B, this means that at least the ending time of time frame A is before the ending time of time frame B—however, this does not impose a constraint on the starting times of time frames A and B.
[0162] “Historical KPI-related data” relates to data describing KPI(s) and/or influencer(s) for a time frame which has already been completed—i.e. not a future time frame.
[0163] In the present disclosure, reference is made to (i) ‘dashboard future time frames’ (i.e. with reference, for example, to 120 of FIG. 3 ), (ii) ‘influencer-ranking future time frames,’ (i.e. with reference, for example, to 120A of FIGS. 4A-4C ), (iii) ‘influencer action future time frames’ 196; and (iv) deviating KPI-viewing time frames 198.
[0164] The “dashboard future time frames” are future time frames for which it is desired to display predicted future KPI values. The ‘influencer-ranking future time frame’ is a future time frame for which it is desired to display an indication of a predicted future influencer-ranking hierarchy. ‘Influencer action future time frames’ and ‘deviating KPI-viewing time frames’ are defined below with reference to FIGS. 7A-7B . The ‘KPI-moving future time frame’ is discussed with reference to FIGS. 5A-5B , and refers to the time period for which it is desired to display calculated deviating influencer values which move one or more KPI(s) from an ‘out-of-range value’ (i.e. as predicted for the ‘KPI-moving future time frame’) to a target value or target range.
[0165] Typically, the behavior of KPIs are influenced or governed, at least in part, by one or more ‘influencers’—influencers are discussed below.
[0166] When the ‘influencer action future time frames’ 196 are discussed in the context of calculated deviating influencer values (see FIGS. 5A-5B ), then the ‘influencer action future time frame 196’ may be referred to as ‘influencer action future time frames 196 for calculated influencers.’
[0167] When the ‘influencer action future time frames’ 196 are discussed in the context of user-proposed deviating influencer values (see FIGS. 7A-7B ), then the ‘influencer action future time frame 196’ may be referred to as ‘influencer action future time frames 196 for user-proposed influencers.’
[0168] When the ‘deviating KPI-viewing time frames 198’ are discussed in the context of calculated deviating influencer values (see FIGS. 5A-5B ), then the ‘deviating KPI-viewing time frames 198’ may be referred to as the ‘deviating KPI-viewing time frames 198 for calculated influencers.’
[0169] When the ‘deviating KPI-viewing time frames 198’ are discussed in the context of user-proposed deviating influencer values (see FIGS. 7A-7B ), then the ‘deviating KPI-viewing time frames 198’ may be referred to as the ‘deviating KPI-viewing time frames 198 for user-proposed influencers.’
[0170] In the present disclosure, certain routines for calculating data and/or displaying data may be carried out ‘in accordance with’ a certain factor—for example, influencer identifiers may be ordered or selected in accordance with magnitudes of one or more deviations. It is understood, in the present disclosure, that when a routine is carried out ‘in accordance with’ a certain factor or factors, that it is possible that the routine will also be carried out in accordance with one or more ‘additional factors’ or additional business rules (for example, displayed identifiers may be ordered both according to magnitudes of deviations as well as whether or not the influencer is designated as an ‘external influencer’ or a ‘controllable influencer.’)
[0171] Thus, in the present disclosure, the phrase ‘in accordance with’ (or ‘according to’) should be interpreted as ‘at least in part in accordance with.’
Discussion of FIG. 1
[0172] Before discussing to questions (Q1, Q2, Q3 and Q4) explicitly with reference to FIG. 2 and latter figures, an exemplary computer environment for implementing the presently-disclosed tool and techniques is described—this computer environment is intended as illustrative and not as limiting.
[0173] FIG. 1 illustrates one embodiment of a digital computer 1200 including a processor 1180. Processor 1180 is shown coupled to a memory 1160, a non-volatile storage 1210 (for example, a flash memory or a hard disk), a video display device 1184, one or more user input device(s) 1190, and a network interface 1150 such as a network interface card. It is appreciated that not every element is required in every embodiment.
[0174] Processor 1180 may be configured to execute instructions and to process data according to a particular instruction set architecture (ISA). In one embodiment, processor 1180 may be configured to implement an x86 compatible ISA, although in other embodiments it is contemplated that any desired ISA may be employed, such as the SPARC V9 ISA, PowerPC compatible ISAs, or MIPS compatible ISAs, for example. (SPARC is a registered trademark of Sun Microsystems, Inc.; PowerPC is a registered trademark of International Business Machines Corporation; MIPS is a registered trademark of MIPS Computer Systems, Inc.).
[0175] In various embodiments, memory 1160 may comprise any suitable type of system memory as described above, such as FB-DIMM, DDR/DDR2, SDRAM, or RDRAM, for example or any other types of volatile memory. Memory 1160 may include multiple discrete banks of memory. Also, in some embodiments, memory 1160 may include multiple different types of memory.
[0176] In some embodiments, computer 1200 may include more than one instance of the devices shown, such as more than one processor 1180, for example. In various embodiments, computer 1200 may be configured as a rack-mountable server system, a standalone system, or in any other suitable form factor. In different embodiments, computer 1200 may be configured as a client system or as a server system.
[0177] In one embodiment, processor 1180 may be configured to run operating system software such as Microsoft Windows, IBM AIX or Sun Microsystems Solaris. Operating system software may in turn provide an environment in which processor 1180 may execute additional software modules in the form of applications, programs, or processes designed to perform specific functions. Running operating system software or software modules may comprise executing instructions that are stored in memory 1160. As shown in FIG. 1 , a client or server application process(es) 1162 (for example, whose execution by processor(s) 1180 is operative to cause computer(s) 1200, display device 1184 to carry out any presently-disclosed routine in accordance with user input received by user input device(s) 1190) resides in the memory 1160.
[0178] Although a single computer is illustrated in FIG. 1 , it is appreciated that any presently disclosed technique may be carried out using multiple computer(s), for example, operatively linked over a LAN or a WAN or in any other manner.
[0179] In a networked deployment, the computer system may operate in the capacity of a server or as a client user computer in a server-client user network environment, or as a peer computer system in a peer-to-peer (or distributed) network environment. The computer system can also be implemented as or incorporated into various devices, such as a personal computer (PC), a tablet PC, a set-top box (STB), a personal digital assistant (PDA), a mobile device, a palmtop computer, a laptop computer, a desktop computer, a communications device, a wireless telephone, a land-line telephone, a control system, a camera, a scanner, a facsimile machine, a printer, a pager, a personal trusted device, a web appliance, a network router, switch or bridge, or any other machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine. In a particular embodiment, the computer system can be implemented using electronic devices that provide voice, video or data communication. Further, while a single computer system is illustrated, the term “system” shall also be taken to include any collection of systems or sub-systems that individually or jointly execute a set, or multiple sets, of instructions to perform one or more computer functions.
[0180] As shown, the computer system may further include and/or be operatively linked to a display device 1184, such as a liquid crystal display (LCD), an organic light emitting diode (OLED), a flat panel display, a solid state display, a cathode ray tube (CRT), a projector, a printer or other now known or later developed display device for outputting determined information. The display may act as an interface for the user to see the functioning of the processor, or specifically as an interface with the software stored in the memory or in the drive unit. In some embodiments, display device 1184 resides in a common housing with computer 1200 (for example, in a laptop computer chassis). Alternatively or additionally, display device 1184 may be provided in a housing separate from the housing of computer 1200.
[0181] Examples of user input device(s) 1190 include but are not limited to mouse devices, keyboard(s), joysticks, and touch screens, and microphones (i.e. for embodiments related to voice-commands). In some embodiments, more than one user input device(s) 1190 are provided—for example, a keyboard and a mouse. In some embodiments, user input device 1190 resides in a common housing with computer 1200 (for example, in a mouse pad of a laptop). Alternatively or additionally, user input device 1190 may be provided in a housing separate from the housing of computer 1200—for example, coupled via a USB or PS/2 interface.
A Discussion of FIG. 2
[0182] FIG. 2 is a flow chart of an exemplary routine for providing one or more interface(s) via which a user may interactively explore relationships between predicted KPI(s) and influencer(s). As will be explained below, when a user interacts with the interface(s) provided by the routine of FIG. 2 , information will be displayed on display device 1184—this information may help the user to answer four questions about future KPI behavior (both predicted actual behavior as well as hypothetical behavior): (i) Q1: what will happen, and when? [see step S211 and FIG. 3 ]; (ii) Q2: why will it happen? [see step S219 and FIGS. 4A-4D ] (iii) Q3: If something bad (or, good) is predicted to happen, how can we avoid (or, benefit from) it? [in see step S225 and FIGS. 5A-5C ]; (iv) Q4: how will user-suggested changes affect the future [in step S229 where user-suggested changes are received via input device 1190, and step S233 where an indication of predicted consequences of the user suggested changes are displayed on display device 1184]?
[0183] Throughout the present disclosure, Q1 is an abbreviation for “question 1,” Q2 is an abbreviation for “question 2,” etc.
[0184] The steps associated with Q1 (step S211) and Q2 (step S219) facilitate diagnosis of the situation, while the steps associated with Q3 (step S225) and Q4 (steps S229, S233) provide decision support for a user who wishes to investigate options for modifying the situation—i.e. in order to prevent (or minimize) a problem from occurring and/or in order to explore how one may take advantage of an opportunity.
[0185] As will be evident from the description below, the presently-disclosed user-interfaces including ‘interactive’ interfaces allow even a “layperson” to explore these questions without any need to understand the underlying mathematics.
[0186] In some embodiments, one or more digital computer(s) 1200 carry out the steps of the routine described below with reference to FIG. 9A in order to provide information displayed in step S211 (see, for example, the screenshot of FIG. 3 ), one or more digital computer(s) 1200 carry out the steps of the routine described below with reference to FIG. 9B in order to provide information displayed in step S219 (see, for example, the screenshots of FIGS. 4A-4D ), one or more digital computer(s) 1200 carry out the steps of the routine described below with reference to FIG. 9C in order to provide information displayed in step S225 (see, for example, the screenshots of FIGS. 5A-5C ), one or more digital computer(s) 1200 carry out the steps of the routine described below with reference to FIG. 9D in order to provide information displayed in step S233 (see, for example, the screenshots of FIGS. 7A-7E ).
Predictive Dashboard: A Discussion of Step S211 of FIG. 2 and the Screen Shot of FIG. 3
[0187] In step S211 of FIG. 2 , a “predictive” dashboard (for example, see FIG. 3 ) is presented to the user on display device 1184. For the example of FIG. 3 , 11 KPIs are presented: Event satisfaction 110A, recommend likelihood 110B, satisfaction with agent 110C, brand satisfaction 110D, repurchase likelihood 110E, external problem resolution 110F resolved within one contact 110G, resolved within two contacts 110H, abandon rate 110I, hold time 110J and talk time 110K.
[0188] It is appreciated that presently-disclosed techniques may be applied to any set of KPIs and influencers, and are not restricted to the example KPIs and influencers presented in the background section. Nevertheless, for the sake of brevity, only certain use cases where described in the background section of this disclosure, and only the KPIs and influencers of only “Use Case 1” will be described in detail with reference to FIGS. 3-7 .
[0189] The “predictive dashboard interface” of FIG. 3 allows a user to view predicted trends for multiple KPIs by presenting, to the user, an indication of each KPI value KPI_PREij, 130ij (where PRE is an abbreviation for “predicted”) for a plurality of different “dashboard” future time frames 120. In the example of FIG. 3 , the predicted KPI values are presented for three “dashboard” future time frames: “next 7 days” (120A), “this month” (120B), and “this quarter” (120C)—it is noted that because the time frames “this month” and “this quarter” have not passed (i.e. they are still ‘occurring’ or ‘in-progress’) at the time of displaying the predictive dashboard, they are considered ‘future time frames.’ Furthermore, it is understood that for KPI_PREij, 130ij that the subscript i designates the ith future time period 120i and j designates the jth KPI 110j.
[0190] As discussed below with reference to FIG. 9A , the predicted KPI values KPI_PREij may be determined by one or more computers 1200 in accordance with historical KPI-related data using one or more disclosed computational techniques or any other computational technique. The “historical KPI-related data” may describe the behavior, during a historical time period, of (i) one or more KPIs; and/or (ii) one or more “influencer(s)” known to or suspected to influence KPI value(s). The historical KPI-related data may be provided to digital computer 1200 in any manner, for example, via a batch and/or continuous data feed or in any other manner.
[0191] Because indications 130ij of predicted KPI values KPI_PREij are displayed for multiple time frames (i.e. in the example of FIG. 3 , for the next 7 days 120A, this month 120B, and this quarter 120C), it is possible for the user to visualize, on display device 1184, a prediction of when particular KPI-related events (e.g., KPI(s) having certain values or being in certain value range(s)) are predicted to or forecast to happen.
[0192] Although explicit numerical values are displayed for each of the predicted KPIs for each time frame KPI_PREij, this is not a limitation—in some embodiments, some indication other than a numerical value may be displayed for one or more predicted KPIs values KPI_PREij—for example, an alphabetical indication and/or a color indication and/or a font size indication and/or any other indication of the numerical value.
[0193] It is noted that certain KPIs may be associated with preferred or target values or ranges. A predicted departure of a KPI value from one of these ranges may be indicative of a problem to be avoided (or an opportunity). Thus, in the non-limiting example of FIG. 3 which relates to the “Customer Contact Center” use case, it is desirable, for example, for the “Customer Experience” KPIs (i.e. the KPIs labeled as 110A-110E) to be above 80%—this indicates a “positive” customer experience. Any value below 80% might be a cause of concern (see the bottom of FIG. 3 under ‘charts’).
[0194] Thus, in some embodiments, the display device 1184 presents (in the ‘predictive dashboard;—see for example, FIG. 3 ) an indication (i.e. numerical or otherwise—for example, a color indication) of a predicted relationship between the KPI value and the target value or range. In the event that the KPI is predicted to depart from the range at a given future time period, the dashboard interface may indicate this predicted KPI value departure.
[0195] Although this indication of a predicted ‘out-of-range-KPI value can be useful, it is not always clear to the user what parameters or “influencers” are responsible for the predicted KPI value (i.e. for a particular future time frame), and the relative relevance of each of these parameters—i.e. the “root cause” of the KPI-related problem/opportunity. Access to this information in step S219 (see the discussion below with reference to FIGS. 4A-4D ) can assist the manager in targeting specific parameters or “influencers” for modification in a future time frame.
A First Discussion of Step S219 and FIGS. 4A-4D
[0196] FIGS. 4A-4D are screenshots of an exemplary “root-cause” interface for displaying an indication of a “predicted future influencer ranking hierarchy” 220 for plurality of influencers 150A-150J for a target KPI (in the example of FIG. 4A , for recommend likelihood 110B) for a target future time frame (in the example of FIG. 4A , for the next 7 days 120A).
[0197] Thus, in the example of FIG. 4A , indications of relative magnitudes of ‘influence contributions’ (i.e. the relative influence that the predicted value of each influencer has upon the predicted KPI value) of 10 influencers 150 upon the predicted future KPI value (i.e. in the case of ‘recommended likelihood’ and ‘next 7 days’ the predicted future value is 79.06 as displayed in FIG. 3 ) for the “recommended likelihood” KPI 110B for the ‘next seven days’ 120B time period are displayed. These influence magnitudes may be determined, for example, by computer 1200 and/or one or more additional computers in communication with computer 1200 for example via network interface 1150—(for example, using the routine described in FIG. 9B or any other routine). Identifiers of these influencers are displayed by display device 1184 as described below.
[0198] In the example of FIG. 4A , the ranked influencers 150 are visually ordered in accordance with relative influence, where the influencer with the greatest influence appears first. By ordering the influencer identifiers 150 according to predicted influence strengths (i.e. in the example of FIG. 4A , the predicted influence strengths for the particular KPI 110B and the particular future time frame 120A are 25.6, 18.6, etc), it is possible to allow the user to visualize the relative importance of the influencers, and to allow a user to focus upon “more important influencers.”
[0199] For the present disclosure, a “predicted future influencer ranking hierarchy” 220 is a plurality of influencer identifiers 150 presented, on display device 1184, in a manner determined according to predicted influence strength (or relative contribution to the total influence) on a predicted value of a KPI at a future time period. Thus, in the non-limiting example of FIG. 4A , the influencer with the greatest amount of influence (i.e. “Average Number of Contacts” 150A) has a relative contribution of 25.6, the influencer with the second greatest amount of influence (i.e. “Brand Description” 150B) has a relative contribution of 18.6, etc.
[0200] In the non-limiting example of FIG. 4A , the influencer identifiers 150 are ordered according to the influence strength/relative contributions. This is not the only way that the influencer identifiers 150 may be presented “in a manner determined to predicted influence strength.” Alternatively or additionally, the influencer identifiers may be presented in a manner so that “more important” influencers (i.e. with larger relative contributions) are displayed in a larger font, a brighter color, or in any other manner indicative of a “larger contribution.”
[0201] The interface displayed in step S219 of FIG. 2 (see, for example, FIG. 4 ) allows the user to answer Q2 of FIG. 2 : “Why will it happen?”—the answer is provided by the graphical indication of which influencers are more “important” in determining why it will happen.
[0202] Optionally, as in the case of FIG. 4A , a numerical indication of the magnitude of each contribution (i.e. 25.6, 18.6, 18.2, etc) is also displayed.
[0203] It is noted that the ranked influencers are specific for the selected KPI 110B and the selected future time period 120A (in the case of FIG. 4A , it is for the predicted value of 79.06 displayed in FIG. 3 ). In some embodiments, in response to a user selection of the KPI 110B and/or time period 120A (for example, by using user input device 1190 to engage the dropdown list for 110B or 120A or to engage any other visual element such as a “GUI control”), a new “predicted future influencer ranking hierarchy” 220 is displayed for the newly selected KPI (for example, selected using the list box associated with graphically element 110A of FIG. 3 or using any other GUI widget or in any other manner) and/or time period (for example, selected using the list box associated with graphically element 120A of FIG. 3 or using any other GUI widget or in any other manner) corresponding to a different KPI_PREij (for example, also displayed in FIG. 3 ).
Histograms of FIGS. 4A-4D
[0204] In some embodiments, in response to a user engagement within the GUI of a particular influencer identifier 150 (for example, using user input device 1190), a corresponding histogram describing the frequency of occurrences of various “predicted outcomes” for particular ‘user-selected’ or ‘user-engaged’ influencer for the time frame selected in 120A—is displayed. Thus, in FIG. 4A (where the influencer “contacts” is displayed)
[0205] By displaying the predicted outcome frequency for a given influencer, this allows the user to visualize not only which influencers are most “important,” but also how the influencer is predicted to behave in the time frame selected in 120.
[0206] In FIG. 4A , the histogram 230A for the “Contacts” influencer (identified by identifier 150A) is displayed. In addition to the histogram, some sort of predicted central tendency (for example, an average) is displayed 232A—for the example of FIG. 4A , the average predicted number of contacts 232A (i.e. this is a predicted influencer value) for the next 7 days is 1.7.
[0207] In FIG. 4B , in response to a user engagement (for example, using user input device 1190) of an identifier 150H for the “Hold Time” influencer, the corresponding “Hold Time” histogram 230H is displayed to the user. In FIG. 4C , in response to a user engagement (for example, using user input device 1190) of an identifier 150B for the “Brand Description” influencer, the corresponding “Brand Description” histogram 230B is displayed to the user. In FIG. 4D , in response to a user engagement (for example, using user input device 1190) of an identifier 150D for the “reason code” influencer, the corresponding “reason code” histogram 230D is displayed to the user.
“Controllable Influencers” and “External Influencers”
[0208] It is recognized that some influencers may be at least partially controllable by the management of the organization whose performance is being measured by the KPIs, while other influencers may be “external” influencers derived from external factors over which the organization his little or no control.
[0209] For Use Case 1, the following influencers are designated as “controllable”—average number of contacts 150A (i.e. because if the agents do a “good job,” the customer will be less likely to repeatedly contact the Call Center for the same issue because the issue will be solved in a ‘low’ number of calls), shipping delay 150E, dispatch sent 150G (i.e., whether or not a dispatch is sent for a call), hold time 150H, local queue time (minutes) 150I, and answered within service level (%) 150J. Definitions of these influencers are available in FIG. 7A .
[0210] For Use Case 1, the following influencers are designated as:“external”—brand description 150B (i.e., because the Call Center has no substantial control over which brands require serving using the Call Center), product description 150C (i.e., because the Call Center has substantially no control over which products require serving using the Call Center), reason code 150D (i.e., because the Call Center has substantially no control over the type of problem for which a customer calls), and age of the system 150F.
[0211] Thus, in some embodiments, the user may wish to focus upon more “controllable” influencers in order to be able to determine how behavior modification by the organization is predicted to affect the future outcome (i.e. as indicated by a KPI value(s)) of one or more aspects of a business process.
[0212] As will be discussed below, in some embodiments, a decision needs to be made about which influencers are presented.
[0213] As will be discussed below, in the non-limiting example of FIGS. 5A-5C and 7A-7B, the only influencers that are presented to the user are those defined as ‘controllable.’
A Brief Overview of Optimization (Q3 of FIG. 2 ; also see FIG. 5 ) and “What-If” Analysis (Q4 of FIG. 2 ; also see FIG. 7 )
[0214] As noted earlier, in many situations the information displayed on the dashboard (both the information and the manner in which it is displayed—see, for example, FIG. 3 ) may provide insight as to what will (i.e. what is predicted to) happen (see Q1 of FIG. 2—dashboard), and the information (both the information and the manner in which it is displayed) displayed by the future influencer ranking hierarchy may provide insight as to why it will happen (see Q2 of FIG. 2—root cause).
[0215] This will allow a manager or user to acquire an understanding of the events that are predicted to transpire in the future and their cause—i.e. to make a diagnosis of a problem and/or opportunity—see Q1 and Q2 of FIG. 2 . Nevertheless, it may still remain unclear to the manager or user what particular steps she may take (i.e. to modify behavior of the organization—for example, by attempting to “hit a target” for one or more influencers) in order to remedy (e.g. in the case of a problematic KPI) and/or take advantage of the situation (e.g. in the case of an ‘over-performing’ KPI)—see Q3 and Q4 of FIG. 2 .
[0216] For the non-limiting example of Use Case 1 (i.e. the Call Center) presented in FIGS. 3-7 , it is noted that one example of “hitting a target” may be as follows: it is predicted that the shipping delay will be 0.3 weeks for a future time period (for example, for the current quarter which is in progress but has not yet ended)—if the organization can “bring down” shipping delay to, say, 0.2 weeks, then the organization could “hit” some hypothetical target. Another example may be as follows: if an average hold time of 45 seconds is predicted for the future time period (e.g., the current quarter)—if the organization can “bring down” the average hold time to, say 35 seconds then the organization could “hit” some hypothetical target.
[0217] Typically, there is some sort of “cost” associated with modifying an influencer from its predicted future value (which, for example, may be derived from its historical behavior and/or the historical behavior of one or more other factors) to a “deviating influencer value.” For example, in order to reduce the shipping delay, it may be necessary to pay shipping agents for overtime hours. In another example, a company may have a fixed number of employees, and reducing shipping delay may require transferring employees from the role of a “phone agent” to the role of a “shipping agent”—thus, improving the predicted performance/value(s) of a first set of KPI(s) (for example, a KPI set that includes shipping delay as an influencer) may be predicted to adversely affect the performance of a second set of KPI(s) (for example, a KPI set that includes hold time).
[0218] As will be discussed below, two additional graphical interface elements are provided in order to allow the user to explore possibilities for “moving” a KPI(s) from its predicted value (for example, a predicted value that indicates “sub-optimal” performance and is below its desired range) to a deviating (hopefully, improved) KPI value. Thus, in FIGS. 5A-5C (see also step S225—this is the FIRST additional graphical interface element), discussed below, an “optimization” interface is provided whereby the user specifies one or more KPIs that she wishes to “fix” for a given future time period (i.e. KPIs which are predicted to perform outside of a desired range for which it is desirable to modify organizational behavior in order to ‘move’ the KPI value to its desired range). In the example of FIGS. 5A-5C , the user does not need to suggest actual influencer values (i.e. there is no need to suggest a “target” influencer value to this), and “most of the work” is done by the machine, with little need for user input or customization.
[0219] And, in FIGS. 7A-7B see also steps S229 and S233 of FIG. 2—this is the SECOND additional graphical interface element, an interface is provided whereby the user can suggest actual deviating influencer values—thus, the user can interactively receive an answer to the question—“based upon the information of historical organization behavior (i.e. as expressed by historical KPI-related data), if we modify organizational behavior to “hit” the suggested list of targets (i.e. as expressed by user-proposed deviating influencer values), what will be the consequences (i.e. as expressed by deviating KPI values that are predicted for the hypothetical scenario proposed by the user)?
[0220] Thus, FIGS. 5A-5C (and step S225) relate to a ‘coarse-grained’ approach with little customization (but little need for the user to exert herself by providing particular influencer values), while FIGS. 7A-7C (and steps S229-233) relate to a ‘fine-grained’ approach with the opportunity for much more customization (at the cost of the need for the user to supply influencer values).
[0221] For the present disclosure, a “deviating value” is a value that differs from a ‘predicted value’ (for example, displayed as element 130 in FIG. 3 and as elements 232 in FIGS. 4A-4D) that is predicted only from historical behavior—i.e. the deviating value is a value predicated upon some sort of proposed change in organizational behavior. One type of “deviating value” is a deviating influencer value—examples include (i) the “machine-proposed” deviating influencer value(s) (see 250 of FIG. 5 discussed below); and (ii) the “user-proposed” deviating influencer value(s) (see 170 of FIG. 7 discussed below).
[0222] Another type of “deviating value” is a deviating KPI value (see element 260 of FIG. 5 and element 280 of FIG. 7)—the deviating KPI value(s) is derived from one or more deviating influencer values. The deviating KPI value(s) may be predicated upon either calculated ‘machine-proposed’ deviating influencer values displayed, for example, in element 160 of FIG. 5 and/or predicated upon ‘user-proposed’ deviating influencer values displayed, for example, in element 170 of FIG. 7A .
A Brief Overview of Optimization (See Step 225 and Q3 of FIG. 2 ; Also See FIG. 5 )
[0223] FIG. 5A is a screen shot of an exemplary interface for displaying a set of “calculated” deviating influencer values 250 predicted to ‘move’ a value of the selected KPI (in the example of FIG. 5 , this is recommended likelihood 110B), for a selected ‘KPI-moving’ future time period 120A (in the example of FIG. 5 , this is “next 7 days”) from its predicted value (for the example of FIG. 5 , this is 79.06 (which is displayed in the dashboard of FIG. 3 as element 130AB) to an acceptable range of 80-100 (see the bottom of FIG. 3 )).
[0224] When one or more deviating influencer values (see, for example, 160 of FIGS. 5A-5C and 170 of FIGS. 7A-7B which provide examples of deviating influencer values) are predicted to ‘move’ a KPI from its predicted value 130 (i.e. for example, displayed in the predictive dashboard—the ‘original’ prediction that is based upon historical data and not upon a proposed deviating influencer) to a deviating KPI value, this is a result of substituting predicted influencer values (for example, from the original model based upon historical behavior) with the deviating influencer values (i.e. associated with a proposed behavior modification for the organization that is not known from the original model based upon historical behavior).
[0225] The predicted influencer values are derived from historical KPI-related data while the deviating influencer values are descriptive of a ‘proposed modified’ organizational behavior (the deviating influencer values may also be derived from the historical KPI-related data).
[0226] It is noted that predicted influencer values are displayed in 162 of FIGS. 5A-5C and in 232 of FIGS. 4A-4B , while deviating KPI values are displayed in 260 of FIGS. 5A-5B and 280 of FIGS. 7A-7C .
[0227] As noted above, the calculated deviating influencer values 250 are determined by one or more computer(s) 1200 in accordance with historical data describing historical behavior of one or more KPI(s) and/or influencer(s).
[0228] Thus, in the example of FIG. 5B , it is predicted (i.e. based upon a mathematical model discussed below) that IF the organization can, within the two days (i.e. as specified in 196 for the ‘KPI-moving’ future time period 120A) hit the ‘targets’ of the: (i) achieving an average number of contacts 150A of 1.61; (ii) achieving a shipping delay 150E of 0.24 weeks; (iii) achieving a ‘dispatch sent’ 150G of 12.02%; (iv) achieving an average hold time 150H of 0.71 minutes; (v) achieving an average local queue time 150I of 1.31 minutes; and (vi) achieving an answered within service level 94.30%, THEN, the value of the KPI recommended likelihood (i.e., for ‘KPI-moving’ future time period—for example, specified in 120A) will not be 79.06 (as displayed in FIG. 3 in the predictive dashboard—this is an ‘out-of-range’ value) but rather in the range of 80-100.
[0229] It is noted that these ‘calculated deviating influencer values’ would require the organization to change its behavior in order to ‘hit the targets.’ Without such ‘behavior change’ (i.e. if the organization continues according to what is predictable from historical KPI-related data), it is predicted that the values of these influencers (see 258 and the values 162A, 162E, 162G, 162H and 162I) would be 1.67, 0.70, 16.31, 2.62, 0.17 and 93.45%. Thus, in the illustrated example, in order to improve recommend likelihood from 79.06% (i.e. out of range) to at least 80% for the ‘KPI-moving’ future time period 120A (for this example, within 7 days), it is necessary in the earlier' time period (i.e. the influencer action time period for calculated deviating influencer values—i.e. specified in 196) of ‘within 2 days from today’ to reduce shipping delay from 0.70 weeks (i.e. what it would have been, according to predictions, if determined solely from historical patterns) to 0.24 weeks (i.e. the target which is the calculated deviating influencer value).
[0230] It is noted that the influencers 150 within area 250, are ordered in accordance with their estimated “importance” or relative influence strength as determined by one or more computer(s) 1200 and are consistent with the hierarchy presented in step S219 (see element 220 of FIG. 4A ). Thus, the “most important” influencers are displayed earlier (alternatively or additionally, in the largest font and/or brightest color, etc) in the order.
[0231] In one example, the most important influencers are those which are predicted to change (or change the most)—thus, instead of “wasting the user's time” by prominently displaying calculated deviating influencers with little or no deviation (i.e. these are not important because these influencers with little or no deviation do not substantially require any change of organizational behavior), those associated with greater deviation (i.e. informing the user that there is a need to change organizational behavior) are emphasized in the visual displaying.
[0232] It is also noted that not all influencers displayed in FIG. 4A are displayed in FIG. 5A—this is because FIG. 5A is limited to “controllable” influencers and does not relate to “external influencers” (see the discussion above)—i.e. in the non-limiting example, the organization has little or no control over influencers 150B-D and 150F.
[0233] In yet another example, there may be a larger ‘set of influencers’ (for example, 10 influencers), and information only about the influencers that are more ‘important’ is shown (for example, only about a plurality of influencers selected from the influencer set—for example, 4 influencers selected from the 10 according to larger deviations from predicted values).
[0234] The interface of FIG. 5A is associated with Q3—it is noted that the answer to Q3 is not always a unique answer—i.e. more than one preemptive actions may be taken in order to “move” a KPI value from a predicted “out-of-range” value to a preferred value or range. Thus, in some embodiment, in order to reach a “unique solution,” one or more ‘constraints’ may be used (i.e. by computer(s) 1200 when determining the set(s) of calculated deviating influencer value(s) 150) to be presented in step S225 of FIG. 2 (or element 250 of FIG. 5 ).
[0235] As noted above, there is often a “cost” associated with modifying organizational behavior in order to ‘hit a target’ or ‘move an influencer value’—for example, in order to reduce shipping delay, it may be necessary to pay overtime wages to the shipping staff or transfer personnel from the customer-service department to the shipping department. In another example, to reduce hold time, it may be necessary to staff a call center with additional personnel, even if some of these personnel are sometimes ‘idle’ because the demand for personnel fluctuates in time.
[0236] Thus, in some embodiments, when the computer(s) 1200 determines the calculated deviating influencer value(s) 150 displayed in 250, this computation is carried out in accordance with one or more ‘constraints.’ One example of a constraint might be ‘minimum cost’—i.e. agents salary, the cost of losing a customer, and other factor(s) may be translated into dollars-and-cents terms, and an attempt is made to find the ‘minimum cost solution’ for moving the KPI(s) specified in 240 from an ‘out-of-range’ value (i.e. for the future time period specified in 120A) to an ‘in range’ value or a target value.
[0237] In another example, the constraint might contain a rule such as ‘we may not allow the brand satisfaction 110D KPI to leave its range. It is appreciated that combinations of such ‘constraint’ may also be used. The constraint policies may be received explicitly via user input device 1190, or may be ‘pre-programmed into the system.’
[0238] Once the calculated deviating influencer values 160 are known, it is possible to display how assuming that the ‘target will be hit’ for these influencer values affects one or more KPIS—thus, on the right hand side of FIG. 5A , indications 260 of a plurality of KPI values (i.e. these KPI values are typically ‘deviating’ KPI values because they are based upon the deviating influencers 160) are displayed for a deviating-KPI-viewing time frame 198 (i.e. indications of what the KPI values are expected to be based upon historical KPI-related and/or influencer related data and assuming that the ‘targets are hit’ for the calculated deviating influencers 160).
[0239] The interface of FIG. 5 provides several user controls, including:
-
- [0240] (i) the checkbox 240 on the left hand side of the screen—this allows the user to specify via input device 1190 for which KPIs the “problem will be solved.” In the example of
FIG. 5 , the only KPI specified is recommend likelihood 110B—however, in the event that another KPI (or multiple KPIs) are specified, then the “problem is solved” for these other KPIs, thereby ‘solving the problem’ for a different set of calculated deviating influencer values 250. - [0241] (ii) a dropdown list for selecting the time period for action 196—thus, in the example of
FIG. 5A , the set of deviating influencer values are displayed for the case where the time period for action is the next day in order to ‘move’ KPI values in the future time frame of the next 7 days 120A. - [0242] (iii) another dropdown list for selecting the deviating KPI-viewing time frame 198 for viewing one or more KPI values derived from the calculated deviating influencer values 160.
- [0240] (i) the checkbox 240 on the left hand side of the screen—this allows the user to specify via input device 1190 for which KPIs the “problem will be solved.” In the example of
[0243] Thus, in FIG. 5A , the value of 120A is set to ‘next 7 days,’ the value of 196 is set to ‘next day,’ and the value of 198 is set of ‘3 days from today.’ (note in the interface of FIGS. 5A-5C the calculated deviating influencers 160 are not affected by the value of the deviating KPI-viewing time frame 198).
[0244] FIG. 5D is a flow chart of an exemplary routine for providing an interface whereby a user specifies one or more KPI(s) and a future time frame and one or more calculated deviating influencer values are displayed.
[0245] FIGS. 6A-6B are time lines illustrating relationships between a time period for action for deviating influencers, a deviating KPI display time frame, and a KPI-moving future time frame. As illustrated in FIGS. 6A-6B , modification of organizational behavior (i.e. by changing influencer values from predicted influencer values to deviating influencer values) in an ‘earlier’ time period (i.e. a time period that ends at an earlier time than the ‘KPI-moving future time frame’ 120A) may modify a KPI at later time (i.e. t2>t1) ‘KPI-moving future time frame’ 120A (for example, as in FIGS. 5A-5C where it is desired to ‘bring into range’ the KPI ‘recommended likelihood’ for the ‘KPI-moving future time frame’ 120 of ‘Next 7 days’ 120A from its predicted value of 79.06 to the preferred range of 80-100).
[0246] It is possible to view effects of the deviating influencer values (i.e. either calculated or user-proposed deviating influencer value) at a ‘later time’ (i.e. t3>t1) defined as the ‘deviating KPI display time frame’ 120 by viewing indications of deviating KPI values (i.e. KPI values that derive from the ‘purely-historically-based’ prediction and are predicted upon the organization ‘hitting the targets’ defined by the deviating influencer values).
[0247] It is noted that in the present disclosure, the terms ‘time period’ and ‘time frame’ are synonymous, and used interchangeably.
A Brief Overview of Providing a ‘What-If’ Interface (see steps S229-233 of FIG. 2 , FIGS. 7A-7E and FIG. 8 )
[0248] FIG. 7A is a screen shot of an exemplary ‘what-if’ interface for: (i) receiving (in step S229 of FIG. 2 ) user-proposed deviating influencer values for a plurality of influencers (for example, using the sliders 170) associated with an influencer action-time frame (specified, for example, using drop-down 196—in FIG. 7A this is 3 days from today); and (ii) displaying (in step S233 of FIG. 2 and FIGS. 7C-7E ) descriptions 282 of one or more deviating future KPI values 282 derived from the user-proposed deviating influencer values for the deviating-KPI-viewing time frame specified by GUI control 198.
[0249] In contrast to the calculated deviating influencer values, indications of which are displayed as element 160 of FIGS. 5A-5C , and which are determined by one or more computers as a solution to an ‘optimization-related’ problem (i.e. moving a predicted KPI to a target value or into a target range), the user-proposed deviating influencer values are explicitly proposed by a user and are received via user input device 1190.
[0250] In the particular example of FIG. 7A , a plurality of sliders 170 is provided, each slider being associated with a respective identifier 150 (i.e. identifying the influencer which is to be modified in the time period for action 196 by the user). When the user moves the slider via user input device 1190, this modifies the corresponding user-proposed deviating influencer value. It is appreciated that the slider 170 is just one particular example of a ‘user control’ or ‘GUI control.’ Other user or GUI controls (or widgets) may include, but are not limited to, text boxes, radio buttons, list boxes, etc. As will be explained below, the values of the user-provided influencers (for example, provided by a user engagement of the slider or other GUI control) may influence the values of the displayed KPIs of 280 (i.e. the ‘answer’ to the question ‘WHAT-IF’)
[0251] In some embodiments, the displayed influencers 270 (including the influencer identifier 150 and the GUI control/widget 170) may be displayed (for example, sorted and/or sized) in accordance with estimations of their relative ‘importance’ during a future time frame (i.e. including but not limited to deviating-KPI-viewing time frame 198, time period for action 196, the time frame specified in 120A, or any other time frame). This may allow the user to ‘focus’ in the ‘more important’ influencers when conducting a ‘what-if’ investigation of one or more KPIs and/or influencers.
[0252] In the non-limiting example of FIG. 7A , the displayed influencers 170 are sorted in the same order (i.e. 150A is displayed first, then 150E, then 150G, then 150H, then 150I, then 150J) as they were sorted in FIGS. 4A-4B , giving preference to what is estimated to be more ‘important’ influencers for the time period specified in 120A. It is noted that sorting is not the only way that the display may be carried out ‘in accordance’ with ‘relative influence contributions’ or relative importance—in yet another example, the influencer identifier 150 and/or user control 170 is sized (or assigned a color) in accordance with the ‘relative influence contributions.’
[0253] In yet another example, the displayed plurality of influencers 270 is part of a larger ‘set of candidate influencers’ and user controls are only displayed for influencers that have a relatively ‘large’ contribution or importance (i.e. relative to ‘rejected’ influencers which have a lesser contribution or importance and for which no control and/or influencer identifier is displayed).
[0254] In one example, the set of ‘candidate influencers’ may include influencers A, B, C, and D. In this example, three sliders for receiving user-proposed influencer values are displayed.
[0255] In this example, for a first future time period (for example, this month), the relative influencer contribution for influencer A=0.4, the relative influencer contribution for influencer B=0.1, the relative influencer contribution for influencer C=0.15, the relative influencer contribution for influencer D=0.35. For a second future time period (for example, this quarter), the relative influencer contribution for influencer A=0.4, the relative influencer contribution for influencer B=0.13, the relative influencer contribution for influencer C=0.12, the relative influencer contribution for influencer D=0.35. Thus, in this example, in the event that a time period of action 196 of ‘this month’ is selected, then sliders (or any other GUI or user control) for only influencers A, C and D will be displayed, and no GUI control for B will be displayed. In the event that a time period of action 196 of ‘this quarter’ is selected, then sliders (or any other GUI or user control) for only influencers A, B and D will be displayed, and no GUI control for C will be displayed.
[0256] Another feature of FIGS. 7A-7B is that each influencer identifier and/or GUI control 170 for accepting a user-proposed deviating influencer value is associated with (i) a scale (in FIG. 7A , the scale for contacts 150A is 1-6, the scale for shipping delay 150E is 0-6, etc); and (ii) default values (in FIG. 7A , the default value of contacts 150A is 1.8, the default value of shipping delay 150E is 0.6, etc).
[0257] In some embodiments, the default value and/or slider scale are selected in accordance with predicted influencer values 162 for any future time period (for example, specified by 196 or 120A or 198)—this is the case for FIG. 7A where the default values are for the future time period of 120A and are identical to the predicted values 162 displayed in FIG. 5A for the ‘next day/time frame.’
[0258] Alternatively or additionally, the default values and/or slider scale are selected in accordance with calculated deviating influencer values 160 (for example, as displayed in step S225 of FIG. 2—also see FIGS. 5A-5C ) for any future time period (for example, specified by 196 or 120A or 198)—this is the case for FIG. 7B where the default values are the calculated deviating influencer values 160 displayed in FIG. 5A .
[0259] In some embodiments, the ‘what-if’ interface is configured according to how the user navigates into the ‘what-if interface’—in one example, in the event that the user navigates to the ‘what-if’ interface from the ‘root-cause’ interface (see, for example, FIG. 4A and FIG. 7A which is labeled at the top as ‘coming from root cause GUI’ and ‘Path A’ of FIG. 8 ), then the default values and/or slider ranges may be determined according to predicted values for the influencers, as discussed above.
[0260] In the event that the user navigates to the ‘what-if’ interface from the ‘optimization’ interface (see, for example, FIG. 5A and FIG. 7B which is labeled at the top as ‘coming from optimization GUI’ and ‘Path B’ of FIG. 8 ), then the default values and/or slider ranges may be determined according to deviating values for the influencers, as discussed above.
[0261] The reader is referred to steps S611, S229 and S233 of FIG. 7C , steps S611, S615, S229 and S233 of FIG. 7D , and steps S613, S625, S229 and S233 of FIG. 7E .
A Discussion of Steps in FIGS. 9A-9D
[0262] FIGS. 9A-9D are flowcharts of computational techniques for generating data related to KPI predictions.
[0263] Reference is made to FIG. 9A . In step S711, historical KPI-related data (i.e. data describing historical values of KPIs and influencers) is stored in memory 1160 of one or more computers. In some embodiments, the KPI-related data is provided using a data feed (for example, via network interface 1150 of computer 1200). Alternatively, data may be loaded to memory 1160 in a ‘batch process,’ or using one or more user input device(s) 1190, or using non-volatile memory 1210 (for example, a USB flash drive or DVD or any other non-volatile memory) or in any other manner.
[0264] In step S715, using one or more microprocessor(s) of the one or more computer(s) 1200, predicted future behaviors of influencers and KPIs are calculated—for example, using one or more prediction model (e.g. including but not limited to regression models, Markov chains, time series models, state space models, Bayesian models, neural networks, or any other appropriate model). In step S715, predicted values (for example, which are displayed as element 130 in FIG. 3 ) of KPIs for a plurality of future time frames (for example, next 7 days, this month, this quarter, or any other future time frames) are computed for multiple KPIs, Towards this end, it may be necessary to also compute predicted values of influencers.
[0265] It is noted that the specific non-limiting examples of how to implement step S715 provided are not intended as comprehensive, and the skilled artisan will appreciate that other implementations are possible.
[0266] Reference is made to FIG. 9B . In step S751, data describing predicted future behavior of one or more specific KPIs (referred to as ‘root-cause’ KPIs because the purpose of the technique of FIG. 9B is to effect root cause analysis) and ‘candidate influencers’ (i.e. influencers for which it is desired to know the predicted degree of influence upon future KPI behavior) is provided into memory 1160 of the one or more computer(s). In one non-limiting example, the data provided in step S751 describing predicted future behavior of one or more specific KPIs and/or future behavior of one or more influencers is provided according to the results of step S715.
[0267] In one non-limiting example, one or more KPIs are associated with a preferred range (or preferred values). According to this non-limiting example, when it is predicted that one or more KPIs will be ‘out of range’ (or differ from a ‘preferred value’) (for example, according to results of step S715), this ‘triggers’ the carrying out of steps S751 and S755 for (i) the KPI that is predicted (for example, in step S715) to be out of range; and (ii) the future time frame in which the KPI is predicted (for example, in step S715) Thus, in some embodiments, steps S751 and/or steps S755 may be carried out in a manner that is contingent upon the detecting of an out-of-range future KPI value (e.g., according to step S715).
[0268] In some non-limiting examples, step S219 of FIG. 2 is carried out for a particular KPI and/or future time frame that is selected from a plurality of KPIs and/or future time frames respectively according to a relationship between a predicted KPI value and its preferred range. Thus, steps S219 may be carried out in a manner that is contingent upon the detecting of an out-of-range future KPI value (e.g., according to step S715).
[0269] Alternatively or additionally, the ‘root cause’ KPI may be selected using user input device 1190 (for example, using GUI control 110B of FIG. 4A ) or in any other manner.
[0270] In the non-limiting example of FIG. 4A , the ‘recommend likelihood’ root-cause KPI and the ‘next 7 days’ future time frame may be selected in any manner—for example, using GUI controls 110B or 120A, or ‘automatically’ because the KPI(s) is predicted (for example, in FIG. 9A ) to be out of range for the future time frame (for example, in FIG. 4A the recommend likelihood is predicted to out of range, or below 80, for the next 7 day time frame—therefore, it may be desired to ‘drill-down’ and this decision may be carried out automatically and/or manually.
[0271] Reference is made to FIG. 9C , which is a flow chart of an exemplary routine for determining one or more calculated deviating influencer values (i.e. associated with an influencer action time frame for calculated deviating influencers) which (i) move a KPI from its predicted value to a target value or range for a future KPI moving time frame; and (ii) satisfy one or more constraints.
[0272] In step S771, indications of (i) an influencer action time frame for calculated deviating Influencers; (ii) for each KPI of at least one KPI, a respective target value or range, for which it is desired to move the respective KPI in a KPI-moving future time frame are provided into memory 1160 of one or more computers 1200.
[0273] There is no limitation upon how this data may be provided into the memory 1160 in step S771. In one example, the influencer action time frame may be provided via user input device 1190 (for example, via GUI control 196 or any other GUI control). Alternatively or additionally, a default value may be used or the data may be provided according to results of step S755 or S715 or in any other manner.
[0274] Similarly, a description (for example, an identifier) of one or more target KPI(s) may be provided into memory 1160 in any manner—automatically or manually—for example, using GUI control 240 of FIG. 5A , or in accordance with results of step S715 which may indicated that a KPI or KPIs are predicted to be ‘out of range.’ Thus, in response to the predicted out of range KPI, steps S771-S797 may be carried out ‘automatically’ in order to determine one way of ‘fixing the problem’ of the predicted out-of-range KPI.
[0275] In steps S775-S793, a search for a ‘solution’ or a ‘preferred solution’ is carried out to locate a set of calculated deviating influencer values which are capable (i.e. according to a prediction model—for example, the model used in FIG. 9A ) of moving one or more out-of-range predicted KPI(s) for a future time frame into a target range or to a target value. As will be discussed below, there may be multiple solutions or sets of deviating influencer values that are predicted to be capable of moving the KPI to the target range or target value. The routine of FIG. 9C is useful for predicting a ‘preferred’ or ‘optimal set’ of calculated deviating influencer values—it is appreciated that the term ‘optimal’ is a relative term, and refers to deviating influencer values that satisfy some constraint or are preferred according to some sort of metric, and is not limited to the absolutely optimal set of deviating influencer values.
[0276] In step S775, a candidate set of influencer values CS is selected using some sort of search routine. In steps S779 and S783, it is determined if the set of influencer values is capable of moving the KPI to the target value or target—for the example of FIG. 5A , it would be determined if the candidate set of influencers are capable (i.e. assuming that they are implemented in the next day) of moving the KPI ‘Recommended Likelihood’ from its predicted value of 79.06 to the target range of 80 in the ‘next 7 days’ future time frame.
[0277] The predicted ability to move the KPI into the target range or value is necessary but not sufficient for the candidate set of influencer values CS to be selected as the ‘preferred’ or ‘optimal’ set of values (and, in some embodiments, be displayed numerically and/or according to some other indication—for example, in element 160 of FIG. 5A ).
[0278] Thus, in the example of FIG. 9C , the deviating influencer values, in order to be optimal, must satisfy, or best satisfy, some sort of cost function and/or constraint function. Towards this end, in step S789, one or more constraint and/or performance parameters are determined for the candidate set CS of influencers. In one example, it is desired to pay a minimal amount of overtime hours to shipping agents and/or phone agents. Thus, in this example, in step S789, the overtime dollar cost may be computed in step S789.
[0279] In step S793, it is determined if wear finished. For example, steps S775-S789 may be carried out multiple times.
[0280] Thus, for a multiple candidate sets CS[1], CS[2] . . . CS[N] which all satisfy the requirement (i.e. as determined in steps S783 and S789) the preferred or optimal set of values (i.e. a CS[i] where i is a positive integer less than or equal to N) is selected in step S797 in accordance with: (i) the set of constraint or performance parameters (for example, a cost and/or value(s) of one or more other predicted KPI(s) associated with each CS[i] or any other CS[i]-dependent factor); and (ii) one or more business rules which allow a selection of which CS[i] is ‘preferred’ or ‘optimal.’
[0281] Any business rule or constraint may be enforced in steps S775, S789 and S797 to determine which influencer values are ‘preferred’ or ‘optimal.’
[0282] Reference is now made to FIG. 9D . In step S811, historical KPI-related data is provided into memory of one or more computers (see, for example, step S711 of FIG. 9A ). In step S815, one or more user-proposed deviating influencer values are received into memory 1160—for example, from user input device(s) 1190. Also, in step S815, a description of indication of an influencer action time frames for user-proposed deviating influencers is received into memory 1160—for example, using user input device 1190 (for example, when a user engages GUI control 196 of FIG. 7A or in any other manner) and/or ‘automatically’ or in any other manner.
[0283] In step S819, it is possible to ‘march forward’ in time and predict one or more deviating KPI values for example, assuming that the user-proposed influencer values prevail at the influencer action time frame for user-proposed deviating influencers. The skilled artisan will appreciate that any appropriate technique may be used for predicting the one or more deviating KPI values, including but not limited to regression techniques, Markov chain techniques, time series, etc.
[0284] In some embodiments, the technique of FIG. 9B and/or the technique of FIG. 9C and/or the technique of 9D may be carried out in response to detecting an ‘exception’ and/or sub-optimal and/or ‘out-of-range’ predicted KPI (i.e. for a particular future time frame)—for example, as predicted in FIG. 9A .
[0285] The skilled artisan will appreciate that the techniques of FIGS. 9A-9D are brought as illustrative examples, and are not intended as comprehensive—other techniques may also be used (including techniques not described explicitly herein), in accordance with knowledge of the skilled artisan.
Additional Discussion
[0286] In some embodiments, the user interface may allow drill-down over one or more dimensions—for, example process, area/location, category etc. Thus, in FIGS. 3-5 and 7, the examples to the example process of ‘Process1’ and to the example location of ‘Ottawa.’ Moreover, the dimensions may have multiple levels—For example, top level process, 1st level sub-processes, 2nd level sub-processes, etc. In these cases, the software may perform necessary computation and allows the user to drill down accordingly.
[0287] In some embodiments, text analytics techniques (linguistic, lexical, pattern recognition, extraction, tagging/structuring, visualization, and predictive techniques, etc.) may be employed for analyzing text data, for example, analyzing customer feedback that come in textual form. In some embodiments, this may be useful for generating the ‘historical KPI-related data.’
[0288] There is no limitation and when and how certain computations may be carried out. In one example, some or all computation of results may be done in the background and stored in database (i.e. results may be pre-computed). In that case, results (for example, displayed in FIGS. 3-5 and 7) are just fetched from the database and displayed when the user interacts with the GUI. Alternatively or additionally, there is no need to ‘pre-store’; computation results.
A Discussion on FIG. 10
[0289] Embodiments of the systems and methods of the present invention may be better explained with reference to FIG. 10 which depicts one embodiment of a topology which may be used to implement embodiments of the systems and methods of the present invention. Topology 2000 comprises performance metric analysis system 2220 (including associated data store 2222) comprising one or more computer devices 2230. These computing devices 2230 may, for example, by organized as a cluster which may be a loosely or a tightly coupled cluster and include one or more load balancers (not shown). Performance metric analysis system 2220 (e.g. one or more of computing devices 2230) may be coupled through network 2270 to computing devices 2210 (e.g. computer systems, personal data assistants, kiosks, dedicated terminals, etc), one or more locations of an entity 2240 and one or more third party data sources 2250 operable to provide, for example, market data, benchmarking data, etc. Network 2270 may be for example, the Internet, a wide area network (WAN), a local area network (LAN) or any other type of conventional or non-electronic communication link such as mail, courier services or the like.
[0290] Generally speaking then, entity 2240 may be a business, non-profit, or other type of entity which implements a process. This process may, for example, be a business process which relates to the functionality or performance of the entity. As discussed above, for example, such business processes may comprise the implementation of customer service through a contact or call center, the implementation of transaction services, the management of supply or demand chains or other inventory management, the implementation of field services, the implementation and management of sales pipelines, etc.
[0291] No matter the type of processes implemented by the entity 2240 however, it may be useful to measure or otherwise analyze (including predicting, simulating, optimizing, etc.) the performance of such a process utilizing a performance metric, such as a KPI as discussed above. Accordingly, entity 2240 may desire to utilize and monitor these performance metrics related to these processes for a variety of reasons, including improving the performance of such processes, reducing the cost of implementing such processes, controlling the quality of such processes, preempting issues which may occur in the future with respect to these processes, substantially optimizing solutions to future problems and predicatively determine the effect of certain solutions to anticipated future problems, etc.
[0292] To that end, performance metric analysis system 2220 may gather data from entity 2240 or a third party data source 2250 to analyze such data to perform analysis on such data and may present an interface such that one or more users at computing devices 2210 may determine what analytics are utilized, the data used for these analytics, view, or affect the results, of such analytics, etc. Embodiments of such interfaces have been discussed previously herein.
[0293] More specifically, in one embodiment, performance metric analysis system 2220 may implement a set of analytics comprising at least predictive analytics, root-cause analytics, optimization and what-if simulation. Colloquially speaking, predictive analytics allows users (for example, associated with entity 2240) to identify and quantify problems (including opportunities) related to one or more performance metrics, root-cause analysis allows users to identify, quantify and rank influencers of performance metrics which may cause any upcoming problems, optimization may determine substantially optimum solution to preempt (or benefit from) any determined upcoming problems and what-if simulation allows a user to determine the effect of prescribed solutions on performance metrics.
[0294] To implement such analytics, performance metric analysis system 2220 may gather data directly or indirectly related to one or more performance metrics from entity 2240. Entity 2240 may have multiple locations 2240a, 2240n where each entity location 2240a, 2240n may comprise multiple servers or other types of computing devices 2242 which are involved in the implementation of the process by the entity 2240 or the storage of data involved with the implementation of the process. In fact, in some instances entity locations 2240a, 2240n may have computing devices which run according to different operating systems or protocols, or which implement different types of applications or databases in conjunction with the process.
[0295] Thus, each entity location 2240a, 2240n may have an associated ETL collector 2160 which is responsible for collecting appropriate data regarding the process or one or more associated performance metrics from various computing devices 2242 utilized in the implementation of the process or used to store data involved with the implementation of the process. ETL collector 2260 can send the data gathered at the corresponding entity location 2240 to the performance metric analysis system 2220 in response to a request from the performance metric analysis system 2220.
[0296] Thus, performance metric analysis system 2220 may, based upon one or more schedules, send out requests to each ETL collectors 2260 at each of the entity locations 2240a, 2240n and receive, in response, a set of data corresponding to that performance metric and that entity location 2240a, 2240n collected over a certain time period. This data may be stored in data store 2222. Analytics may then be performed by the performance metric analysis system 2220 using the data gathered from the entity locations 2240a, 2240n. The analytics performed by performance metric analysis system 2220 may be determined, at least in part, by a users interaction with an interface presented by performance metric analysis system 2220 and the results of the analytic performed may similarly be displayed to a user through the provided interface.
[0297] It should be noted here that not all of the various entities depicted in topology 2000 are necessary, or even desired, in embodiments of the present invention, and that certain of the functionality described with respect to the entities depicted in topology 2000 may be combined into a single entity or eliminated altogether. Topology 2000 is therefore exemplary only and should in no way be taken as imposing any limitations on embodiments of the present invention.
A Discussion on FIG. 11
[0298] It may be useful here to delve into more details regarding one example of the architecture of a performance metric analysis system. Moving on to FIG. 11 then, one embodiment of a performance metric analysis system is depicted. Performance metric analysis system 2300 comprises an interface provider computer 2310 which is coupled to one or more other computers, such as computers at a location of an entity or a user computer (not shown), and is configured to receive data from, and send data to, these other computers. This interface provider computer 2310 may be configured to communicate via one or more protocols or standards including HTTP, XML, SOAP, Web Services Definition Language (WSDL), FTP, etc. and may additionally be configured to provide one or more interfaces for interaction with performance metric analysis system 2300, including, for example, web pages, command line interfaces, web services interfaces, database server functionality, application server functionality, web server functionality, etc.
[0299] In particular, in one embodiment, a web interface provider module 2312 may be configured to provide one or more web page interfaces for interacting with performance metric analysis system 2300 such that a user at a user computer may be presented with an interface comprising certain data and the user may utilize this interface to be presented with different or additional information, to refine the information that is presented, to select various parameters, to provide various settings (e.g. values or ranges) or to receive, or provide, other desired information to or from performance metric analysis system 2300. Examples of such interfaces which may be provided by web interface provider module 2312 have been discussed thoroughly herein and will not be elaborated on further.
[0300] Thus, utilizing the interfaces and protocols according to which the interface provider computer 2310 is configured, interface provider computer 2310 is operable to receive data from one or more computers external to performance metric analysis system 2300 (as will be discussed below in more detail) such as data collected from an ETL collector at an entity location 2410, one or more user selected performance metrics 2412, one or more user selected time periods 2414, a user selected location 2416, a value for one or more influencers 2418, one or more business rules 2420 or other information from one or more computers external to performance metric analysis system 2300. Interface provider computer 2310 may be configured to transform this data according to one or more communication protocols, file formats, etc., and communicate the extracted data to one or more of data manager computer 2320, predictive analysis computer 2330, root analysis computer 2340, optimization computer 2350 or what-if analysis computer 2360.
[0301] Conversely, the interface provider computer 2310 may utilize the configured protocols and interfaces to communicate data received from data manager computer 2320, predictive analysis computer 2330, root analysis computer 2340, optimization computer 2350 or what-if analysis computer 2360 to one or more computers external to performance metric analysis system 2300 (as will be discussed below in more detail). Such data may include a request for collected ETL data 2422, and one or more values for one or more performance metrics 2424 and one or more values for one or more influencers 2426. As will be noted from a review of the disclosure above, the values of the one or more performance metrics 2424 or the one or more values for the one or more influencers 2426 may be displayed using one or more web pages provided by web interface provider module 2312, where the web page interfaces may be configured to present this data using a variety of formats or displays, including for example, one or more dashboards comprising tables, bar graphs, sliders, pull down menus, buttons, check boxes, etc. As mentioned previously, these interfaces have been discussed in great detail and will not be elaborated on further here.
[0302] Data manager computer 2320 is configured to take collected data 2428 from ETL collectors at entity location(s) and transform the data for storage in data store 2326. More specifically, data manager computer may be executing transformation manager module 2322 and loading manager module 2324. Loading manager module 2324 may be responsible for pulling data from the various entity location(s) and may issue requests 2430 to be sent to one or more entity locations. Loading manager module 2324 may receive the raw collected data 2428 and stage this data in a storage area (for example, in data store 2326).
[0303] Transformation module 2322 may transform this staged data by applying one or more sets of rules to the staged data before the data is stored for subsequent use in data store 2326. To transform the data, transformation module 2322 may apply one or more client specific modification rules based upon the entity, location, application, database type, etc. from which the collected data was received, one or more data cleaning rules intended to sift or filter the data (for example, to discard out of range data), one or more performance metric analysis, one or more performance metric analysis system rules to conform the data to an internal format utilized for processing or one or more data joining rules to associate desired data. The transformed data can then be stored in data store 2326. By applying these sets of rules to the collected data, data from a wide variety of client types and in a wide variety of protocols, formats and stored in a variety of database types may be easily integrated.
[0304] Predictive analysis computer 2330 may be configured to utilize this transformed data to determine the values for one or more performance metrics for one or more future time periods (referred to also as time periods). More specifically, in one embodiment, predictive analysis module 2332 executing on predictive analysis computer 2330 may receive one or desired performance metrics 2432, one or more time periods 2434 and one or more locations 2436 (other categories, such as product, process, etc. may also be utilized and will not be elaborated on further) and determine the value for the one or more desired performance metrics 2432 for each of the one or more time periods 2434 and the one or more locations 2436 (where the location may correspond to all locations).
[0305] Predictive analysis module 2332 may utilize a set of predictive models 2334, where there is at least one predictive model corresponding to each of the performance metrics 2432, such that the model corresponding to a particular performance metric can be fitted utilizing appropriate data corresponding to a location. Using this fitted model then, a prediction for that performance metric for each of the one or more time periods 2434 may be determined using historical data 2440 associated with the performance metric stored in data store 2326. The determined performance metric prediction values 2438 can then be displayed to a user using an interface provided by interface provider computer 2310.
[0306] Examples of interface(s) which may be displayed to a user in conjunction with predictive analysis computer 2330 providing information to, or obtaining information from, a user are presented in FIG. 3 .
[0307] Based on interaction with such an interface provided by interface provider computer 2310 a user may indicate that he wishes to see a root cause analysis. Root cause analysis computer 2340 may be configured to provide an analysis of why a performance metric has been determined to have a particular value for a particular time period. In one embodiment, root cause analysis module 2342 executing on root cause analysis computer 2340 may determine a ranked list of influencers (for example, explanatory variables) associated with the calculation of a value for a performance metric (for example, output variables), where the list may be ranked by size of contribution (for example, to the value of the performance metric) or according to some other ordering criteria. In addition, in certain embodiments, the ranked list of influencers may exclude certain influencers over which an entity may have little or no control.
[0308] Specifically, a user may indicate through an interface provided by interface provider computer 2310 that the user wishes to see a root cause analysis corresponding to a particular performance metric 2444 for a particular time frame 2446 and a particular location 2448 (other categories, such as product, process, etc. may also be utilized and will not be elaborated on further). Root cause analysis module 2342 may access a set of root cause models 2344, where the set of root cause models 2344 may comprise one or more root cause models corresponding to a particular performance metric. Using the root cause model corresponding to the desired performance metric 2444, the historical data 2440 stored in data store corresponding to the location 2448 and the value for that performance metric 2438 determined for that future time period by predictive analysis computer 2330, root cause analysis module 2342 may determine a set of influencers and associated contributions of these influencers on the value determined for the future time period such that the influencers may effectively be ranked by contribution.
[0309] The determined ranked list of influencers and associated values 2450 can then be displayed to a user using an interface provided by interface provider computer 2310. Examples of interface(s) which may be displayed to a user in conjunction with root cause analysis computer 2340 providing information to, or obtaining information from, a user are presented in FIGS. 4A-4D .
[0310] Based on interaction with an interface provided by interface provider computer 2310 a user may indicate that he wishes to have an optimization performed. Optimization computer 2350 may be configured to provide a substantially optimal solution for attaining one or more particular value for one or more performance metrics over a time period. In one embodiment, optimization module 2352 executing on optimization computer 2350 may determine such a solution, where the solution comprises a set of influencers and corresponding values such that by attaining the values for the influencers the values for the one or more performance metrics will likely be obtained for that time period. The solution proposed may take into account one or more specified constraints. Additionally, optimization module 2352 may be configured to calculate the effect this proposed solution will have on one or more other performance metrics if implemented.
[0311] Specifically, a user may indicate through an interface provided by interface provider computer 2310 that the user wishes to have an optimization performed for one or more performance metrics 2454 for a particular time period 2456, a desired location 2460 and whether they would like a max/min solution or a desired range solution 2458. Optimization module 2352 may access a set of optimization models 2354, where the set of optimization models 2354 may comprise a set of max/min models and a set of desired range models. Each set of models (max/min and desired range) has a set of models corresponding to each of the performance metrics.
[0312] Thus, based upon the one or more performance metrics 2454, the particular time period 2456, the desired location 2460, historical data 2440, a ranked list of influencers and values 2450 and the desired type of solution (max/min or a desired range) 2458, optimization module 2352 may select an appropriate optimization model. The selected optimization model may then be applied by the optimization module 2352 with respect to the ranked list of influencers for the one or more performance metrics 2450 and the value for the one or more performance metrics for the time period 2438 to determine a substantially optimal solution for attaining a particular value (for example, a maximum or minimum value or a desired range) for the one or more performance metrics over the time period. The determination of his substantially optimal solution may also account for any user imposed constraints, such as, for example, constraints related to the values of other performance metrics, influencers or other measurable criteria. The optimization module 2352 may also determine the effect that this optimal solution may have on the values of one or more performance metrics (which may be the same or different than performance metrics 2454).
[0313] The substantially optimal solution 2464 (for example, a list of influencers and their associated values for attaining desired value for the one or more performance metrics, known as an actionable recommendation) and the effect of this solution of the one or more performance metrics 2466 (for example, performance metrics and associated values) may be displayed to a user using an interface provided by interface provider computer 2310. Examples of interface(s) which may be displayed to a user in conjunction with optimization computer 2350 providing information to, or obtaining information from, a user are presented in FIGS. 5A-5C .
[0314] Instead of, or in addition to, being presented with an optimization, a user may indicate through an interface provided by interface provider computer 2310 that a user wishes to perform a what-if analysis such that the user may determine the effect of certain prescriptive solutions on certain performance metrics. What-if analysis computer 2360 may be configured to provide one or more influencers and associated value range for these of influencers. Additionally, what-if analysis computer 2360 may determine a list of performance metrics which may be most affected by variations in the one or more influencers. A user can thus specify values within the value ranges for one or more of the influencers and the optimization computer will calculate values for the list of performance metrics.
[0315] Specifically, in one embodiment, what-if analysis module 2362 executing on what-if analysis computer 2360 may receive a set of influencers 2468 where the influencers may be provided by a user through an interface or a list of influencers determined, for example, as a solution by optimization computer 2350, a particular time period 2470 and a desired location 2472. What-if analysis module 2362 may access a set of what-if models 2364, where the set of what-if models 2364 may comprise a set of models corresponding to each of the performance metrics. Thus, based upon the set of influencers 2468, the set of models 2364 historical data 2440 and the ranked list of influencers and values 2450 and the particular time period 2470, what if analysis module 2362 may determine value ranges for the set of influencers and default values for the set of influencers 2476, a corresponding set of affected performance metrics and values for the set of performance metrics 2480.
[0316] The set of influencers 2468, value ranges for the set of influencers and default values for the set of influencers 2476, corresponding set of affected performance metrics and values for the set of performance metrics 2480 may be displayed to a user using an interface provided by interface provider computer 2310. By interacting with the interface presented to the user by interface provider computer 2310 a user may specify new values within the ranges for one or more of the set of influencers. What-if analysis module 2362 may receive the update values for the one or more set of influencers 2482 and based on these updated values determine new values for the set of performance metrics 2484. These new values may then be displayed to the user using the interface provided by interface provider computer 2310. Examples of interface(s) which may be displayed to a user in conjunction with what-if analysis computer 2360 providing information to, or obtaining information from, a user are presented in FIGS. 7A-7C .
[0317] After reading the above it should be noted that in addition to performing the specific functionality and transformations discussed, each of interface provider computer 2310, data manger computer 2320, predictive analysis computer 2330, root cause analysis computer 2340, optimization computer 2350 and what-if analysis computer 2360 may be a general purpose computer of the type known in the art having a central processing unit (“CPU”), at least one read-only memory (“ROM”), at least one random access memory (“RAM”), at least one hard drive (“HD”), or one or more input/output (“I/O”) device(s). Furthermore, it will be realized that although interface provider computer 2310, data manger computer 2320, predictive analysis computer 2330, root cause analysis computer 2340, optimization computer 2350 and what-if analysis computer 2360 are depicted as separate computers, the functionality described with respect to these computers may be combined into one or more computers, including one or more general purpose computers.
[0318] Moreover, while web interface provider module 2312, transformation module 2322, loading manager module 2324, predictive analysis module 2332, root cause analysis module 2342, optimization module 2352 and what-if analysis module 2362 have been depicted as separate modules and executing on distinct computer, each of web interface provider module 2312, transformation module 2322, loading manager module 2324, predictive analysis module 2332, root cause analysis module 2342, optimization module 2352 and what-if analysis module 2362 may be combined into one or more modules which encompass the functionality of one or more of web interface provider module 2312, transformation module 2322, loading manager module 2324, predictive analysis module 2332, root cause analysis module 2342, optimization module 2352 and what-if analysis module 2362. Along the same line, as data store 2326 and the data stores which comprise predictive analysis models 2334, root cause analysis models 2344, optimization models 2354 and what-if analysis models 2364 have been depicted as separate, there may be one or more data stores which stores all, or a combination of, the data in data store 2326 and predictive analysis models 2334, root cause analysis models 2344, optimization models 2354 and what-if analysis models 2364.
[0319] In the foregoing specification, the invention has been described with reference to specific embodiments. However, one of ordinary skill in the art appreciates that various modifications and changes can be made without departing from the scope of the invention as set forth in the claims below. For example, though certain embodiments have been described in conjunction with individual purchasers other embodiments will be understood to apply equally well to the provisioning of incentive programs with fleet purchases of vehicles, etc. Accordingly, the specification and figures are to be regarded in an illustrative rather than a restrictive sense, and all such modifications are intended to be included within the scope of invention.
[0320] Benefits, other advantages, and solutions to problems have been described above with regard to specific embodiments. However, the benefits, advantages, solutions to problems, and any component(s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential feature or component of any or all the claims.
Concluding Remarks
[0321] It is further noted that any of the embodiments described above may further include receiving, sending or storing instructions and/or data that implement the operations described above in conjunction with the figures upon a computer readable medium. Generally speaking, a computer readable medium may include storage media or memory media such as magnetic or flash or optical media, e.g. disk or CD-ROM, volatile or non-volatile media such as RAM, ROM, etc. as well as transmission media or signals such as electrical, electromagnetic or digital signals conveyed via a communication medium such as network and/or wireless links.
[0322] Having thus described the foregoing exemplary embodiments it will be apparent to those skilled in the art that various equivalents, alterations, modifications, and improvements thereof are possible without departing from the scope and spirit of the claims as hereafter recited. In particular, different embodiments may include combinations of features other than those described herein. Accordingly, the claims are not limited to the foregoing discussion.
Claims
What is claimed is:
1. A method of displaying KPI-related data on a display device in accordance with information received via a user input device, the display device and the user input device being operatively coupled to one or more digital computers, the method comprising:
a. for each dashboard future time frame of a plurality of dashboard future time frames, for each KPI of a plurality of KPIs, determining a respective predicted KPI value and displaying, on the display device, a respective indication of the respective predicted KPI value;
b. for a selected influencer-ranking future time frame, for a root-cause KPI selected from the plurality of KPIs whose predicted KPI value in the selected influencer-ranking future time frame is predicted to be influenced by behavior of a set of influencers, displaying, on the display device, a predicted future influencer-ranking-hierarchy including identifiers of a first plurality of influencers of the set of influencers, the first plurality of influencers being displayed in accordance with predicted relative influence contributions to the predicted KPI value of the root-cause KPI in the selected influencer-ranking future time frame;
c. for each influencer of a second plurality of influencers of the set of influencers, receiving, via the user input device, a respective user-proposed deviating influencer value, the second plurality of influencers being a subset of the first plurality of influencers;
d. determining one or more deviating future KPI values for one or more of the plurality of KPIs, the one or more deviating future KPI values derived from the user-proposed deviating influencer values based on:
i. a selected influencer action future time frame for the user-proposed deviating influencer values;
ii. a deviating-KPI-viewing time frame for the user-proposed deviating influencer values; and
iii. the received user-proposed deviating influencer values; and
e. displaying, on the display device, for one or more of the plurality of KPIs, descriptions of the one or more deviating future KPI values derived from the user-proposed deviating influencer values.
2. The method of claim 1, wherein the identifiers of the predicted future influencer-ranking-hierarchy are ordered and sized on the display device in accordance with the predicted relative influencer contributions to the predicted future value of the root-cause KPI for a future time frame.
3. The method of claim 1, wherein the first plurality of influencers are selected from the set of influencers in accordance with predicted relative influence strength in the selected influencer-ranking future time frame.
4. The method of claim 1, wherein:
a. the method further comprises:
i. for each given influencer of the second plurality of influencers, displaying on the display device a respective user control that is visually associated with a respective identifier of the given influencer such that the user controls and the influencer identifiers are displayed in accordance with the predicted relative influence contributions to one or more KPIs in a future time frame; and
ii. for the each given influencer, the respective receiving of the respective user-proposed deviating influencer value is carried out in response to a respective one or more user engagements of the respective displayed user controls using the user input device.
5. The method of claim 4, wherein the user controls and the influencer identifiers are sized and ordered on the display device in accordance with the predicted relative influence contributions to the one or more KPIs in the future time frame.
6. The method of claim 4, wherein for a given influencer of the second plurality of influencers:
a. the displaying of the corresponding influencer identifier for the given influencer by the display device; and
b. the displaying of a corresponding user control for the given influencer by the display device; are contingent upon a relationship between:
i. a magnitude of a predicted relative future influence, over the one or more KPIs, by the given influencer over the one or more KPIs; and
ii. magnitudes of predicted relative future influences, over the one or more KPIs, by one or more other influencers, wherein:
a) the one or more other influencers are members of the set of influencers; and
b) the one or more influencers do not belong to the second plurality of influencers.
7. The method of claim 1, wherein:
a. the method further comprises:
i. for each given influencer of the second plurality of influencers, displaying a respective user control that is visually associated with a respective identifier of the given influencer; and
ii. for each given influencer, the respective receiving of the respective user-proposed deviating influencer value is carried out in response to a respective one or more user engagements of the respective displayed user controls using the user input device; and
b. the display device and the user input device are configured such that each user control associated with a respective influencer is associated with a default value that is determined in accordance with a respective predicted future value for the influencer-action time frame for the respective influencer.
8. The method of claim 1, wherein:
a. KPI value descriptions for the one or more deviating future KPI values derived from the user-proposed deviating influencer values are displayed, each deviating future KPI value being mathematically associated with a respective deviation magnitude respectively describing by how much said each deviating future KPI value derived from the user-proposed influencer values deviates from its respective predicted KPI value for a future deviating-KPI-viewing time frame; and
b. the displaying of the descriptions of the one or more deviating future KPI values derived from the user proposed deviating influencer values is carried out in accordance with the deviation magnitudes.
9. The method of claim 1, wherein:
a. KPI value descriptions for the one or more deviating future KPI values derived from the user-proposed influencer values are displayed, each deviating future KPI value being mathematically associated with a respective deviation magnitude respectively describing by how much said each deviating future KPI value derived from the user-proposed influencer values deviates from its respective predicted KPI value for a future deviating-KPI-viewing time frame; and
b. the display of a given displayed KPI value description derived from the user-proposed influencer values is contingent upon a respective deviation magnitude for the given displayed KPI value.
10. The method of claim 1, wherein:
a. the method further comprises:
i. for a KPI-moving future time frame, one or more target KPIs of the plurality of KPIs, and the selected influencer action future time frame for calculated influencers, displaying on the display device, a set of calculated deviating influencer values which:
a) satisfy one or more provided constraints;
b) deviate from predicted influencer values for the KPI-moving future time frame; and
c) if adopted at the selected influencer action future time frame, are predicted to move a future value of the one or more target KPIs, at the KPI-moving future time frame, from the predicted KPI values to a target value or range.
11. The method of claim 10, wherein:
a. the method further comprises:
i. for a specified deviating KPI-viewing time frame for the calculated deviating influencer values, the specified deviating KPI-viewing time frame for the calculated influencer values ending after the selected influencer action future time frame for the calculated influencers, displaying, on the display device, indications of the one or more deviating future KPI values determined for:
a) the calculated deviating influencers for the selected influencer action future time frame of the calculated influencers;
b) the deviating KPI-viewing time frame.
12. The method of claim 11, wherein:
a. KPI value descriptions for the one or more deviating future KPI values determined for the calculated deviating influencer values are displayed, each deviating future KPI value determined for the calculated influencer values being mathematically associated with a respective deviation magnitude respectively describing by how much said each deviating future KPI value determined for the calculated deviating influencer values deviates from its respective predicted KPI value for a the deviating-KPI-viewing time frame; and
b. the displaying of the descriptions of the one or more deviating future KPI values determined for the calculated influencer values is carried out in accordance with the deviation magnitudes.
13. The method of claim 1, wherein:
a. KPI value descriptions for the one or more deviating future KPI values determined for the calculated influencer values are displayed, each deviating future KPI value determined for the calculated influencer values being mathematically associated with a respective deviation magnitude respectively describing by how much said each deviating future KPI value determined for the calculated influencer values deviates from its respective predicted KPI value for a future deviating-KPI-viewing time frame; and
b. the display of a given displayed KPI value description determined from the calculated influencer values is contingent upon a respective deviation magnitude for the given displayed KPI value.
14. The method of claim 1, wherein the constraint includes at least one of a KPI constraint and a cost constraint.
15. The method of claim 1, wherein:
a. a specification of the root-cause KPI;
b. a specification of the influencer-ranking future time frame;
c. a specification of the influencer action future time frame; and
d. a specification of the deviating-KPI-viewing time frame is received via one or more user input devices.
16. The method of claim 1, wherein at least one user input device is selected from a group consisting of a mouse, a keyboard, a touch-screen, and a microphone.
17. The method of claim 1, wherein at least one influencer is indicative of at least one of:
a. a length of a telephone conversation;
b. an amount of time that a telephone device is in a hold-state; and c. a number of distinct telephone conversations; and
c. whether or not a telephone conversation is an incoming conversation or an outgoing conversation.
18. An apparatus for displaying KPI-related data, the apparatus comprising:
a. one or more computers;
b. a display device operative coupled to one or more of the computers; and
c. a user input device operatively coupled to one or more of the computers; wherein:
i. the one or more computers are operative to compute, from historical KPI-related data, predicted future KPI-related behavior including:
a) predicted behavior of one or more KPIs of a plurality of KPIs;
b) predicted future behavior of one or more influencers of a set of influencers, and
c) predicted future relationships between the one or more influencers of the set of influencers and the one or more KPIs of the plurality of KPIs;
ii. the display device is operative to display, in accordance with the predicted future KPI-related behavior:
a) for each dashboard future time frame of a plurality of dashboard future time frames, for each KPI of the plurality of KPIs, a respective indication of a respective predicted KPI value;
b) for a selected influencer-ranking future time frame, for a root-cause KPI selected from the plurality of KPIs whose predicted KPI value, is predicted, in accordance with the predicted future KPI-related behavior, to be influenced by behavior of the set of influencers, a predicted future influencer-ranking-hierarchy including identifiers of a first plurality of influencers of the set of influencers, the first plurality of influencers being displayed in accordance with predicted relative influence contributions to the predicted KPI value of the root-cause KPI in the selected influencer-ranking future time frame;
c) the user input device is operative, for each influencer of a second plurality of influencers of the set of influencers, to receive a respective user-proposed deviating influencer value, the second plurality of influencers being a subset of the first plurality of influencers;
d) the one or more computers are further operative to compute, from the historical KPI-related data and from the user-proposed deviating influencer values, one or more deviating future KPI values for one or more of the plurality of KPIs;
e) the display device is further operative to display, for one or more of KPIs of the plurality of KPIs, descriptions of one or more of the deviating future KPI values based on:
a selected influencer action future time frame for the user-proposed deviating influencer values;
a deviating-KPI-viewing time frame for the user-proposed deviating influencers values; and
the computed deviating future KPI values.
19. A computer program product, comprising a non-transitory computer usable medium having a computer readable program code embodied therein, said computer readable program code adapted to be executed to implement a method of displaying KPI-related data on a display device in accordance with information received via a user input device, the display device and the user input device being operatively coupled to one or more digital computers, the method comprising:
a. for each dashboard future time frame of a plurality of dashboard future time frames, for each KPI of a plurality of KPIs, determining a respective predicted KPI value and displaying, on the display device, a respective indication of the respective predicted KPI value;
b. for a selected influencer-ranking future time frame, for a root-cause KPI selected from the plurality of KPIs whose predicted KPI value in the selected influencer-ranking future time frame is predicted to be influenced by behavior of a set of influencers, displaying, on the display device, a predicted future influencer-ranking-hierarchy including identifiers of a first plurality of influencers of the set of influencers, the first plurality of influencers being displayed in accordance with predicted relative influence contributions to the predicted KPI value of the root-cause KPI in the selected influencer-ranking future time frame;
c. for each influencer of a second plurality of influencers of the set of influencers, receiving, via the user input device, a respective user-proposed deviating influencer value, the second plurality of influencers being a subset of the first plurality of influencers;
d. determining one or more deviating future KPI values for one or more of the plurality of KPIs, the one or more deviating future KPI values derived from the user-proposed deviating influencer values based on:
i. a selected influencer action future time frame for the user-proposed deviating influencer values;
ii. a deviating-KPI-viewing time frame for the user-proposed deviating influencer values; and
iii. the received user-proposed deviating influencer values; and
e. displaying, on the display device, for one or more of the plurality of KPIs, descriptions of the one or more deviating future KPI values derived from the user-proposed deviating influencer values.
Patent Citations (19)
| Patent | Date | Inventor | Cited By |
|---|---|---|---|
| US6876988(B2) | 2005-04-01 | Helsper et al. | Applicant |
| US7373633(B2) | 2008-05-01 | Kraiss et al. | Applicant |
| US7979380(B2) | 2011-07-01 | Moyne et al. | Examiner |
| US2004/0068429(A1) | 2004-04-01 | MacDonald | Examiner |
| US2004/0068431(A1) | 2004-04-01 | Smith et al. | Examiner |
| US2004/0148047(A1) | 2004-07-01 | Dismukes et al. | Examiner |
| US2005/0004834(A1) | 2005-01-01 | Smith | Examiner |
| US2006/0010164(A1) | 2006-01-01 | Netz et al. | Applicant |
| US2006/0242033(A1) | 2006-10-01 | Corbett | Applicant |
| US2006/0242035(A1) | 2006-10-01 | Corbett et al. | Applicant |
| US2007/0150324(A1) | 2007-06-01 | Makita et al. | Examiner |
| US2008/0183786(A1) | 2008-07-01 | Shimizu | Applicant |
| US2008/0195440(A1) | 2008-08-01 | Bagchi et al. | Applicant |
| US2008/0201397(A1) | 2008-08-01 | Peng et al. | Applicant |
| US2008/0243912(A1) | 2008-10-01 | Azvine et al. | Examiner |
| US2009/0064025(A1) | 2009-03-01 | Christ et al. | Applicant |
| US2009/0171879(A1) | 2009-07-01 | Bullen et al. | Examiner |
| US2009/0217183(A1) | 2009-08-01 | Moyne et al. | Examiner |
| US2010/0077077(A1) | 2010-03-01 | Devitt | Examiner |
Non-Patent Literature (22)
- Loucks, D.P.“Model Sensitivity and Uncertainty Analysis.” Water Resources Systems Planning and Management, UNESCO 2005, http://ecommons.library.cornell.edu/bitstream/1813/2804/12/09—chapter09.pdf.Examiner
- “Altaworks Selected by Industry Influencers to Demonstrate at DEMO 2002 Conference.” Business/Technology Editors, Business Wire, New York: Jan. 16, 2002, p. 1, Proquest.Examiner
- “Fonterra Canpac Deploys Pavilion's Production Performance Management Solution; Fast Moving Consumer Goods Manufacturer Improves Operational Performance and Reduces Waste.” Business Wire. New York: May 23, 2006, p. 1, Proquest.Examiner
- Frangopol, Dan, M. and Neves, Luis, C.“Probabilitic Performance Prediction of deteriorating Structures Under different Maintenance Strategies: Condition, Safety and Cost.” Apr. 2004.Examiner
- SAS Institute Inc. The SAS Enterprise Intelligence Platform: SAS Business Intelligence, Enabling BI Consolidation and Standardization Without Compromises, White Paper, 2008.Applicant
- SAS Institute Inc., Enhancing Sales and Operations Planning with Forecasting Analytics and Business Intelligence, White Paper, 2008.Applicant
- SAS Institute Inc., Successful Automating Decisions With Business Analytics: How to Identify High-Value Opportunities for Embedding Analytics Into Your Business Processes, White Paper, 2008.Applicant
- SAS Institute Inc., Data Management for Risk, The Importance of Data for Effective Risk Management, White Paper, 2009.Applicant
- Fair Isaac Corp., Predictive Analytics, Product Sheet, 2008.Applicant
- Fair Isaac Corp., Xpress Optimizer, Product Sheet, 2008.Applicant
- SPSS Inc., Clementine® 12.0—Specifications, Product Specifications, 2008.Applicant
- SPSS Inc., Get More Out of Your Analytical Investment, Advertisement, 2008.Applicant
- SAS Institute Inc., “How to Compete on Analytics: The Analyitcal Center of Excellence,” White Paper, 2008, 18 pages.Applicant
- SAS Institute Inc., “Data-Driven Decision Making: Analyzing Your Data to Improve Student Learning,” White Paper, 2008, 19 pages.Applicant
- SAS Institute Inc., “The SAS Enterprise Intelligence Platform: SAS Business Intelligence; Enabling BI Consolidation and Standardization Without Compromises,” White Paper, 2008, 13 pages.Applicant
- SAS Institute Inc., “Enhancing Sales and Operations Planning with Forecasting Analytics and Business Intelligence,” White Paper, 2008, 14 pages.Applicant
- SAS Institute Inc., Ngenera Corporation, and Intel, “Sucessful Automating Decisions with Business Analytics: How to Identify High-Value Opportunities for Embedding Analytics into Your Business Processes,” White Paper, 2008, 22 pages.Applicant
- SAS Institute Inc., “Data Management for Risk: The Importance of Data for Effective Risk Management,” White Paper, 2009, 11 pages.Applicant
- Fair Isaac Corporation, “Predictive Analytics; Custom Insurance Analytics,” Product Sheet, 2008, 2 pages.Applicant
- Fair Isaac Corporation, “Decision Management Tools; Xpress Optimizer,” Product Sheet, 2008, 2 pages.Applicant
- SPSS Inc., Clementine@ 12.0—Spcifications,“Achieve Better Insight and Prediction with Data Mining,” Product Specifications, 2008, 6 pages.Applicant
- SPSS Inc., “Get More Out of Your Analyitica Investment,” Advertisement, 2008, 6 pages.Applicant