WO2024052887A1

WO2024052887A1 - User engagement analyzer and methods for its use

Info

Publication number: WO2024052887A1
Application number: PCT/IB2023/058961
Authority: WO
Inventors: Ariel Maislos; Itamar FALCON; Michael Ehrlich; Or MALKAI
Original assignee: Coho Ai Ltd.
Priority date: 2022-09-11
Filing date: 2023-09-10
Publication date: 2024-03-14

Abstract

Computerized methods and systems analyze subject behaviors. The methods and systems operate to access, by one or more processors (202), a segmentation model and a prescription model, stored in a program memory (204); and analyze engagement data associated with a subject (120) engaging with a digital product (110). The segmentation model (241) analyzes the engagement data to classify the subject into at least one segment. The prescription model (243) receives the classified segment for the subject and coupled with the engagement data, outputs one or more prescribed actions associated with the subject's engagement with the digital product. The one or more prescribed actions are then taken based on an occurrence of a trigger, detected from monitoring (230) of the engagement data.

Description

USER ENGAGEMENT ANALYZER AND METHODS FOR ITS USE

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is related to and claims priority from commonly owned U.S. Provisional Patent Application, Serial No. 63/405,445, entitled: Coho - A Generic Customer Interaction Analyzer, filed September 11, 2022, the disclosure of which is incorporated by reference in its entirety herein.

TECHNICAL FIELD

The disclosed subject matter is directed to digital products and user interactions associated therewith.

BACKGROUND

Consumers typically interact with digital products, such as web conferencing services, web page building web sites, and games. Initially, the consumer is introduced to the digital product on a free or trial basis, at no or nominal cost. As user engagement increases, the user’s familiarity and comfort level with the digital product increases, such that the user typically wants to use additional features. These additional features are typically available on a pay or subscription basis.

SUMMARY

The present disclosure provides computerized methods and systems to analyze subject behaviors. The methods and systems operate to access, by one or more processors, a segmentation model and a prescription model, stored in a program memory; and analyze engagement data associated with a subject engaging with a digital product. The segmentation model analyzes the engagement data to classify the subject into at least one segment. The prescription model receives the classified segment for the subject and coupled with the engagement data, outputs one or more prescribed actions associated with the subject’s engagement with the digital product. The one or more prescribed actions are then taken based on an occurrence of a trigger, detected from monitoring of the engagement data.

The present disclosure is directed to a computer-implemented method for analyzing subject behaviors. The method comprises: accessing by one or more processors, a segmentation model and a prescription model, stored in a program memory; obtaining engagement data associated with a subject engaging with a digital product; providing, to the segmentation model, a first portion of the obtained engagement data, such that the segmentation model classifies the subject into at least one segment; monitoring a second portion of the engagement data associated with the subject engaging with the digital product for a trigger; and, providing the one or more segment classifications for the subject, and the first portion of the engagement data, to the prescription model, wherein the prescription model outputs one or more prescribed actions associated with the subject’s engagement with the digital product, the one or more prescribed actions taken based on an occurrence of the trigger.

BRIEF DESCRIPTION OF THE DRAWINGS

Non-limiting examples of embodiments are described below with reference to figures attached hereto that are listed following this paragraph. Identical structures, elements or parts that appear in more than one figure are generally labeled with a same numeral in all the figures in which they appear. Dimensions of components and features shown in the figures are chosen for convenience and clarity of presentation and are not necessarily shown to scale.

In the drawings:

FIG. 1 is an example environment in which that disclosed subject matter operates;

FIG, 2 is an example architecture of the Engagement Analysis or Home System of the example environment of FIG. 1 ;

FIGs. 3A-3F are diagrams of example journeys in accordance with the disclosed subject matter;

FIG. 4 is a timeline in accordance with the disclosed subject matter;

FIG. 5 is a diagram of an example segmentation model, in accordance with the disclosed subject matter;

FIG. 6A is a diagram showing an example of training a prediction model, in accordance with the disclosed subject matter;

FIG. 6B is a diagram showing an example prediction model, in accordance with the disclosed subject matter;

FIG. 7 is a diagram showing an example prescription model, in accordance with the disclosed subject matter; and

FIG. 8 is a process or flow diagram showing an example process for analyzing and acting based on user engagement with a digital product or object, in accordance with the disclosed subject matter. DETAILED DESCRIPTION

Before explaining at least one embodiment of the disclosure in detail, it is to be understood that the disclosure is not necessarily limited in its application to the details of construction and the arrangement of the components and/or methods set forth in the following description and/or illustrated in the drawings. The disclosed subject matter is capable of other embodiments or of being practiced or carried out in various ways.

As will be appreciated by one skilled in the art, aspects of the present disclosure may be embodied as a system, method or computer program product. Accordingly, aspects of the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a "circuit," "module" or "system." Furthermore, aspects of the present disclosure may take the form of a computer program product embodied in one or more non-transitory computer readable (storage) medium(s) having computer readable program code embodied thereon.

Throughout this document, textual references are made to trademarks, and domain names. These trademarks and domain names are the property of their respective owners, and are referenced only for explanation purposes herein.

In this document, references to "n” and “n*” refer to the last member of a finite series or set.

In this document, a “subject” includes a user, a team, or an account, with a one or more users (or members) associated with a team or an account. In some cases, the subject, typically the user, may also include an alias. This alias correlates an alternate designation for the user, for example, in the form of an anonymous identifier (ID), which is saved in a cookie or local storage, to the identity of the user, until the user is actually designated as a user by the system (element 102 of FIGs. 1 and 2).

Reference is now made to FIG. 1, which shows an exemplary operating environment, including a network(s) 100, to which is linked a home system 102, i.e., an Engagement Analysis System, which may be a computer system, formed of servers, computers, computer components, and the like, linked to the network, or alternately, servers and the like in the cloud. The system 102 operates a platform, which analyzes and provides recommendations for user engagement with digital products of a given customer 110 of the entity of the system 102, represented by the customer server or customer system 110 (the “customer server”, “customer system” and “customer” are used interchangeably and use the element number 110). With respect to the home system or system 102, the terms “system”, 102, and “platform” are used interchangeably herein. The system 102, for example, may also include other computers, including servers, components, and applications, e.g., client applications, associated with home server 102, as detailed below.

The term “digital product” is representative of, for example, a digital product, digital object, platform, web-service, software program, and the like, available via a network, a website, a website service or product, an application, and an application service or product.

The system 110 represents an entity who is a customer of the entity associated with the home server 102. The system 110 includes computers, servers and the like. The customer system 110, for example, provides a digital product, such as a web conference service, which is used by one or more of the users 120 (represented by their computers Al, A2, B 1 and B2, the computers including for example, desktop computers, laptop computers, tablet computers mobile telephones, and the like) in web-based (networked) communications between the users Al, A2, Bl and B2, as well as with one or more of third party users 130 (via their respective computers 130a, 130b, 130c), who are also users of the web conference service provided by the customer system 110. The web conference service, provided by the customer systeml lO, for example, provides and/or facilitates video and audio communications, chat, messaging, whiteboard, multiple participant including group conferencing, and the like. The customer system 110 links to the network 100, and is also linked to the home system 102, either directly, or over the network 100.

The users Al 120A1, A2 120A2 are members of an account, i.e., Account A with the customer 110, for example, of an enterprise, the local area network (LAN) 120a, which links to the network 100. Similarly, the users Bl 120B1, B2 120B2 are members of another account, Account B, different from Account A, represented by the enterprise local area network (LAN) 120b, which links to the network 100. Third party users 130 of the customer’s 110 digital product are entities such as an accountant, represented by the computer 130a, a banker, represented by the computer 130b, and a consultant, represented by the computer 130c, link to the network 100. For explanation purposes, the users 120A1 (via computer Al) and 120A2 (via computer A2) typically have web conferences via the digital product of the customer system 110 (web conference facility), with one or more of the Accountant 130a, Banker 130b and/or Consultant 130c.

The network(s) 100 is, for example, a communications network, such as a Local Area Network (LAN), or a Wide Area Network (WAN), including public networks such as the Internet. As shown in FIG. 1, the network 100 may be a single network, such as the Internet, but is typically a combination of networks and/or multiple networks including, for example, cellular or Bluetooth or other networks. "Linked" as used herein includes both wired or wireless links, either direct or indirect, and placing the computers, including, servers, components and the like, in electronic and/or data communications with each other.

FIG. 2 shows an architecture for the system 102. The system 102 may be spread across numerous servers, computerized components and the like, and may be in computer systems and/or servers in the cloud, which are not shown.

The architecture for the system 102, shown, for example, in FIG. 2, includes one or more components, engines, modules and the like, for providing numerous additional server functions and operations, and, for running the processes of the system 102. Those components, engines and modules of the system 102 are shown and described below, but additional components, engines, models, and modules are also permissible as part of the system 102, to perform any additional functions. For example, a “module” includes one or components for storing instructions, (e.g., machine readable instructions) for performing one or more processes, and including or associated with processors, for example, the CPU 202, for executing the instructions. The system 102 may be associated with additional storage, memory, caches and databases, both internal and external thereto. For explanation purposes, the system 102 may have a uniform resource locator (URL) of, for example, www.example.hsystem.com.

The architecture of the system 102 (platform), includes a central processing unit (CPU) 202 formed of one or more processors, electronically connected, i.e., either directly or indirectly, including in electronic and/or data communication with storage/memory 204, and components including databases 206, an interface module 210, a communications module 211, a rules and policies module 212, data processing components, such as a raw data collector 220, a preprocessing module or preprocessor 222, a behavior monitoring and triggering module (behavior monitor and trigger) 230, and an engine 240, representative of a segmentation model 241, a prediction model 242 and a prescription model 243. The aforementioned components, modules, engines, and models are linked to each other, either directly or indirectly, with some linkages noted below, so as to be in direct or indirect communications with each other.

The Central Processing Unit (CPU) 202 is formed of one or more processors, including microprocessors, for performing the system 102 (platform) functions and operations detailed herein. The processors are, for example, conventional processors, such as those used in servers, computers, and other computerized devices, including hardware processors. For example, the processors may include x86 Processors from AMD (Advanced Micro Devices®) and Intel®, Xenon® and Pentium® processors from Intel, as well as any combinations thereof. The processors for example, may also comprise general-purpose computers, which are programmed in software, including trained models, to carry out the functions described herein. The software may be downloaded to the computer in electronic form, over a network, for example, or it may, alternatively or additionally, be provided and/or stored on non-transitory tangible media, such as magnetic, optical, or electronic memory.

The storage/memory 204 is any conventional storage media, program memory or the like. The storage/memory 204 stores machine executable instructions for execution by the CPU 202, to perform the disclosed processes and methods (collectively “processes”). The storage/memory 204 also includes machine executable instructions associated with the operation of the components, including the interface module 210, a communications module 212, data processing components, such as a raw data receiver 220, a preprocessing module or preprocessor 222, a behavior monitoring and triggering module (behavior monitor and trigger) 230, and an engine 240, representative of a segmentation model 241, a prediction model 242 and a prescription model 243. The storage/memory 204 also, for example, stores rules and policies for the system 102 and the home server 102, and may also, for example, store the models 241, 242, 243 (detailed below).

The processors of the CPU 202, and the storage/memory 204, although shown as a single component for representative purposes, may be multiple components. These multiple components may be outside of the system 102, and linked to the network 100.

The databases 206 provide one or more storage media for various databases and data storage to allow the system 102 to perform the processes disclosed herein.

The Interface module or interface 210 facilitates communications between the system 102 and the customer system 110.

The communications module or communicator 211 sends and receives communications, e.g., data communications, between the system 102 and the customer server 110, as well as other components, servers, computers, and the like along the network 100. The rules and policies module 212 stores and received various rules and policies, such as those used to tag or label data, for example, by the preprocessing module 222, including labeling data for the models 241, 242, 243 and/or running the various models 241, 242, 243.

The raw data collector module or raw data collector 220 collects raw data from the customer server 110 and/or from the digital product associated with the customer server 110. The raw data is, for example, generated from interactions between the subject (e.g., users 120), from a computer associated with the subject, and the digital product of the customer system 110. The raw data is, for example, sent or pushed by the customer system 110 to the system 102, and the module 220 over the network 100. Alternately, the raw data may be obtained by being pulled from the customer system 110 by the module 220. Combinations of the aforementioned pushing and pulling may also be used to collect the raw data by the module 220. The raw data includes, for example, 1) Subject (User/Team/ Account) Properties, 2) Events, and/or, 3) Application Objects.

Subjects -Users, Teams and Accounts and Subject Properties

A User (e.g., an individual user also called an end-user) is the digital representation of a person or entity who operates or interacts with the customer’s 110 digital product. Every end-user has a unique identifier and additional properties which can change over time. Other user properties include, for example, user name, email, sign-up date and time, language, country, state, subscription plan, role (e.g., administrator), position/job title (e.g., company president), and the like. A user can be either a “natural person”, sometimes identified with a user ID or a “virtual person”, sometimes identified with an “API key”. A user may also be an automated process which interacts with the digital product on behalf of the subject.

A Team, also known as a user group, is an identifiable collection of end-users. Like end-users, every team has a unique identifier and additional properties. Other team properties include, for example, team name, email, sign-up date and time, language, country, state, subscription plan, contact, and the like. A user can be associated with one or more teams, and each team can contain a hierarchical structure of other teams.

An Account is the digital representation of a paying user or entity, e.g., an enterprise, business, or other collective group of users, for the digital product of the customer 110. Each account has a unique constant identifier and additional properties. Other account properties include, for example, account name, email, sign-up date and time, language, country, state, subscription plan, contact, and the like. An account can contain one or more users, with or without additional team hierarchy. Events

An event is a unit of information about a a user, team or an account, which is tied to a specific point in time. Events may include subject, i.e., user, team, or account, actions in the digital product. Each event has a unique identifier, a name, a timestamp, associated end-user and/or team and/or account ID(s)(Identification) and arbitrary additional information about the event.

An event usually signifies an interaction between a subject and the digital product of the customer 110, a change in mutable properties of an identifier, a change in the product state or status, a change in the business relationship between the subject and the digital product of the customer 110. However, sometimes, an event may capture backend interactions, such as API calls, periodic tasks or triggers that cannot be assigned to a specific user but rather to a specific team/account or machine associated with a subject. Because events are the lowest level of interaction, they can be unstable, they change over time.

Examples of events may include button clicks (activations), page scrolling, and pages scrolled, pages visited, typed (user-entered) text, subscription(s) created, digital product pricing changed, training video provided, and Application Programming Interface (API) consumed.

Application Objects

An application object is data about the use of an application, which is also, for example, the data product. For example, when the application/data product is a web conferencing product, such as that of the customer 110, the application object may be, for example, one or more of: metadata as to the web conference, number of participants in the web conference at one or more given times over the duration of the web conference, start and/or stop times of the web conference, duration of the web conference, and video/no video of each of the participants in the web conference.

The preprocessing module or preprocessor 222 obtains the raw data and normalizes the raw data for use by the system 102, typically as input for the segmentation model 241, as well as input for the prediction model 242 and the prescription model 243. The normalized data is known as “engagement data”, indicative of a subject’s engagement with the digital product of the customer server 110. The obtained raw data (for normalization into engagement data for a subject) includes, for example, data for Subject (User/Team/Account) Properties, Events, and/or Application objects, as detailed above. This preprocessing of the raw data, results in the normalized engagement data, which includes, for example, one or more of: l)subject properties (the subject being users, teams, or accounts), 2) events, and/or 3) application objects, all as detailed above for the “Raw Data”. The normalized engagement data also includes, for example, data of one or more of: 1) Features, 2) Behaviors (formed from one or more Features), and/or 3) Journeys (formed from one or more features and/or behaviors), this data known as Behavioral Data.

The engagement data serves as input for the segmentation model 241, the prediction model 242 and the prescription model 243. The engagement data, from the preprocessor 222, is monitored by the real time behavior monitor 230 (which monitors, for example, the events portion or events data of the engagement data) for one or more triggers (as detailed below).

The preprocessed data in addition to the engagement data, may, for example, include one or more segments for the instant subject as previously determined by the segmentation model 241, if such a determination for the subject has been made. These one or more segments (along with the engagement data) may also be input for the segmentation model 241, the prediction model 242, and/or the prescription model 243, if such segments were previously output by the segmentation model 241.

Features

A feature is an event within the context of the customer’ s 110 digital product. One or more features serve as building blocks, which makes it possible to further define more complex concepts (such as behaviors and journeys, secondary data as detailed below) without having to use events and abstract query languages to define complicated logic. A feature has an identifier that typically has a meaningful name taken from the customers’ 110 digital product, a timestamp denoting when the feature was invoked/detected/identified, a unique identifier, attribution such as user/team/account that used the feature, optional predetermined properties with generic behavioral meaning, and a set of arbitrary additional properties.

For example, when the customer 110 is a web conference provider and the digital product is a web conferencing service, platform or the like, as shown, for example, in FIG. 1, features may include Schedule a meeting (e.g., web conference), Invite a user to participate in a web conference, and/or send a message to a future or present participant in a web conference.

For example, events in the raw data, known as “raw events”, as determined by the preprocessing module 222, can be mapped to zero or more product features, which are also features. A product feature is defined, for example, as follows: 1. Conditions on the raw event properties (not on the user and account properties) with logical AND between them.

2. Logical OR between the conditional statements of each raw event.

Therefore, the high-level schema of product feature definition is as follows:

“CONDITION on raw event X and CONDITION on raw event X AND ....”

OR

“CONDITION on raw event Y and CONDITION on raw event Y AND ....”

OR

Once a raw event is mapped to a product feature, all the raw event properties are copied to the product feature.

Product features of the digital product, which can be categorized into types, for example, for the web conferencing digital product of the customer system 110, may include, one or more of:

1. Custom/Generic

2. Feature in use

3. Sign up

4. Log in

5. Log out

6. Account created

7. Account added user

8. Account deleted user

9. Page view

10. Page leave

11. Share broadcast

12. Share to users

13. Invite

14. Collaborate

15. Trial started

16. Trial ended 17. Upgrade plan

18. Downgrade plan

19. Payment

Behaviors

A Behavior is a sequence of features, which uses either imperative or declarative language. A behavior may be defined using natural language, a graphical user interface, or machine learning (ML) models. It is the relationship of the features which describe a “behavior”. A behavior can be either deterministic, or stochastic in nature. The behavior, once determined, is subsequently translated into code. The translated code then becomes the input for the segmentation model 241. which, when given a stream of features, can identify and output each and every specific instance of that behavior.

Examples of behaviors which can be defined by the preprocessing module 222, during preprocessing operations, include, for example, Habit patterns, where a subject used a certain feature of the digital product of the customer 110, at least once a day over the past 30 days, and “champion” behavior”, where a subject invited at least 5 new users to use the digital product of the customer 110 or introduced five new users to the customer 110.

Behaviors may also include usage patterns, which are an abstraction for a particular behavior or usage that a subject, i.e., a user, team, or account, can do in the digital product. It consists of logic based on one or more product features, for example, for capturing an onboarding flow, a usage pattern would include the following features:

1. Sign up

2. Connect data source

3. View dashboard

Additionally, a usage pattern can indicate which subject are active, for example, subjects who use one of the following features:

1. Visit application

2. Drill down on graph

3. Share widget

There are different ways to determine usage patterns, including logic based on features of a product. In addition, machine learning algorithms can help uncover usage patterns. Journeys

A Journey includes a progression over time, based on an aggregation of features and/or behaviours. It is composed of a series of steps which a subject, i.e., a user, a team, or an account, follows when interacting with a digital product. Each step in the journey can be defined as a behavior within the digital product or interactions outside the digital product (e.g., sales calls, support tickets). A step in the journey can be defined as ingress only (once an entity enters the step it cannot exit it) or as ingress/egress (a subject can exit the step after entry). Examples of steps include, for example, Ingress only: User-created a support ticket three times, and Ingress/egress: User demonstrated behavior Y in the last seven days.

System subjects, i.e., users, teams, accounts, can define multiple journeys and assign any user, team or account, to any journey. The assignment can be explicit (assign user X to journey Y) or implicit (assign all users with property X=Y to journey Z).

A subject can be assigned to multiple journeys simultaneously or contemporaneously. All the journeys in the system define a state machine for a subject.

For example, in the lifecycle of a subject, i.e., a user, team, or account, the subject goes through many journeys, for example: 1) an onboarding journey, 2) a subscription plan journey - from a free user to a paying user, 3) a maturity in a specific use case journey, and 4) a product maturity journey.

Journeys are modeled as graphs with nodes and edges. Edges mark transitions between nodes in the graph while nodes consist of the following types:

1. Step - A step in the journey can be defined by a logic that applies to all data model 214, 242, 243 entities, features, patterns, context, users, and other steps as well.

2. Sub-journey - A sub-journey is a step that captures all of the subjects who have completed another journey, allowing the subject to create complex hierarchical journeys.

3. Classifier - Based on the classification, a classifier node divides a subject (user/team/account) into different segments by the segmentation model 241, which are then split into different branches. Using a classifier, a user may be classified into different personas, and each persona has a different adoption profile in the product that reflects its maturity journey.

4. Eligibility node - Usually, the eligibility node is the first node that defines which subjects are eligible to enter the journey, for example, active subjects. FIG. 3A shows a diagram of an example of a typical maturity journey model. From a pool of active users at block 302, personas are classified at block 304, for example, for Managers (blocks 306) and developers (blocks 308). For both managers and developers, increased and/or progressive engagement with the digital product results in the manager/developer going from a beginner manager 306a/developer 308a, to an advanced manager 306b/developer 308b, to a champion manager 306c/developer 308c.

FIG. 3B shows a diagram of an example of a contextual in-application journey. From a pool of active users at block 312, a first user invites a second user to view a dashboard, e.g., Dashboard X, at block 314. In the invite is also context, with the dashboard X of a different context. The system 102 tracks the journey per invite ID and dashboard ID, for example, as user B opens the invite, at block 316, and views the content of Dashboard X upon opening the invite, at block 318. User A can invite multiple users to view multiple dashboards.

Subjects can define journeys and steps based on all the entities in the data model (241, 242, 243), but because the data model is a generalization layer, the system 102 can find different personas, define journeys, and track users’ progress on different journeys using machine learning algorithms and unsupervised learning.

Using the journeys, the system 102 predicts how a subject will proceed in the journey in a generic way. With the generic journey definitions, the system 102 can predict, via the predictive model 242, for example, the likelihood of purchase, churn, intention to adopt a feature, lifetime values (LTV), and identify expansion opportunities.

By using the generalization layer of the system 102, the system 102 can track journeys per subject and aggregate them for teams/accounts.

FIGs. 3C-3F show additional examples of Journeys.

FIG. 3C shows an example of a journey of a user, as to become a Product Qualified Lead (PQL) via mmultiple pathways. Using the customer system 110 as the web conference service 110, the subject as an individual user, User A 120A1, at block 322, the user 120A1 has signed up for the digital product as a “free” user. Should the user 120A1 have an integrated single sign on (SSO), block 324, invite five other users to use the web conference service, at block 326, a configured production environment, at bock 328, or the user’s account exceeds a storage limit, at block 330, the journey continues, as the user 120A1 has further engagement with the digital product. The journey advances still further when the user 120A1 is now a paid account of the web digital product, at block 332, or a PQL is sent to a hub spot, at block 334. FIG. 3D shows a multistep journey that captures the steps in a configuration of a feature. The process begins at block 340 with an integrated login box for the user 120A1. The process moves to block 342a, where a social login is configured, and then to block 342b, where a single sign on SSO for the user 120A1 is created. From block 342a and 342b, the process moves to block 344 where the configuration is deployed to production, indicating a progression of increasing interaction between the user 120A1 and the digital product of the customer 110.

FIG. 3E shows an example of a high-level subject maturity journey in multiple use cases. For example, a subject, here a user 120A1 is signed up for the digital product of the customer 110, at block 350. Moving in one direction, one journey the user 120A1 is initially a beginner meeting host, at block 352 and then they advance to being an advanced meeting host, at block 353. Returning to block 350, another journey may begin as the user 120A1 is a beginner webinar host, at block 354 and advances to being an advanced meeting participant, at block 355.

Also, for example, as shown by the diagram of FIG. 3F, a journey may be based on a subject’s interaction with a digital product of the customer 110. The subject, for example, a user, starts by an initial interaction with the digital product as an evaluator, at block 362, and then is a beginning user, with increased interaction at block 364. The user moves to an advanced user, at block 366, and ultimately, an expert or champion user, at block 368, with increased engagement with the digital product.

Timeline

The preprocessing module 222 may also be used to create timelines, for example, the timeline 400 shown in FIG. 4. For example, for one or more of the entries, such as end-users, teams, and accounts, a timeline 400 is made based on the above-detailed Events, Features, Application Objects, Journeys and Actions (defined with reference to the Monitoring Module or Monitor 230). The timeline 400 is typically a hierarchical sequence of meaningful occurrences, on which we can show different analytics, trigger actions in real time in other systems, and use it as an input for different statistical models to generate subject segmentations, predictions, inferences, and the like.

The Behavior monitoring and trigger module or behavior monitor and trigger 230 monitors the preprocessed data, i.e., the engagement data, and, for example, the events (events data or events data portion) of the engagement data, to determine whether a threshold event or other event programmed to cause a trigger is sufficient to result in, or otherwise activate, a trigger. The trigger (active or activated trigger) is then sent, for example, in real time, to the engine 240, as data for one or more of the models 241, 242, 243. Additionally, the module 230 analyzes timelines 400 created by the preprocessing module 222, to define and trigger Actions. For example, the Actions allow end-users 120A1 to activate workflows on different systems to support their go-to-market operations.

For example, a trigger may occur: 1) whenever a subject reaches a step in a journey which indicates an intent to upgrade to a higher pricing tier, which results in an Action - send a notification message to a sales representative and create an “Opportunity” object in the CRM; 2) whenever a user demonstrates a behavior which indicates they are ready for an advanced feature - toggle the right feature flag, send them an in-app message and an email with the feature’s documentation; or 3) whenever an account shows a usage anomaly (e.g., usage is lower than 30 days baseline) - alert the customer success representative.

The engine 240, for example, supports models, including, for example, the segmentation model 241, the prediction model 242 and the prescription model 243. The models 241, 242, 243 may be, for example, based on decision trees, neural networks, clustering models, and the like, and for example, are stored in, and/or downloaded into, program memory, such as the storage/memory 204. The models 241, 242, 243, for example, typically perform their operations in real-time, and in succession, contemporaneous in time.

The segmentation model 241 places each detected subject, from the preprocessed data, into a segment or category, and which includes, for example, a segment or category with a probability of the subject being in the segment or category. A segment includes, for example, a subject, i.e., a user, team or account, for example, with common behavior characteristics. One or more users and/or teams or accounts can be grouped by this model 241 based on their segments. A user, team and/or account may be classified into one or more segments.

For example, segments may include personas and populations, and a probability that the user is of a particular persona, and the team or account has a probability of being in a particular population. Accordingly, personas are typically segments of individual users, while populations are typically groups of subjects, the subjects including one or more of users, teams, and accounts. A persona is a profile of a product's typical customer, for example, an administrative assistant, R&D Manager, team leader, product manager, designer, or developer. As a result of the persona awareness, the system 102 can model the interactions between different personas within the product. It is common for products to have multiple customer personas that use the same product, and each persona uses different features, and a high-level look at all the personas and their interactions can provide insights into the customer's propensity for taking action.

For example, user 120A1 schedules many web conferences using the digital product of the customer 110, but rarely attends or participates in the scheduled conferences, and if participating, the participation is short (timewise) and does not extend over the entire conference. This user may have a 90 percent probability for the persona segment “administrative assistant”, based on user behavior, as detected by the system 102.

A population is a group of subjects (users/teams/accounts) that share common characteristics such as, country, operating system, application version, and the like.

For example, user 120A1 and user 120A2 are from Account A, and also qualify as a team, as they are from the same enterprise. The system 102 has detected a large amount of web conference activity with user 130b, a banker, with graphics of financial charts being displayed in these conferences. Accordingly, Account A may have an 80 percent probability that it is the Account for a population segment of a financial department of a company. Account A may also have a 50 percent probability that it is in the population segment Automobile Importers Financial Departments.

FIG. 5 shows the segmentation model 241, with example inputs and outputs. The model 241 is, for example untrained, and is, for example, a decision tree model, neural network model, or the like. Inputs include, for a subject, engagement data including, for example, subject properties, such as user properties 502a, team properties 502b, and account properties 502c, and/or subject, i.e., user, team, or account, features, behaviors and/or journeys 504. Segments for the subject may also be inputs, provided they were previously generated by the segmentation model 241.

The segmentation model 241 outputs include, for example, the probability of the given user belonging to Persona 1 512a through the probability of the given user belonging to Persona n 512n. The segmentation model 241 outputs may also include, for example, the probability of the given team/account belonging to population 1 514a through the probability of the given team/account belonging to population n 514n.

The prediction model 242 provides predictions for one or more specific users based on their segments, or one or more segments of users, for the propensity for an event occurrence, for example, of the user/segment(s) of users to further engage or disengage from a digital product, or provide potential engagement with other digital products. For example, predictions output by the prediction model 242 include propensity to purchase/churn the digital product, propensity to upgrade or downgrade the digital product, and/or the propensity of adopt a feature of the digital product. The prediction model 242, for example, may be a decision tree model, a neural network model, or the like.

The prediction model 242 is a trained model, trained as shown in FIG. 6A. The training, for example, represented by block 610, is, for example, by supervised learning with labeled data from a data set. The labeled training data, which is input to the model 242, for training 610 includes, for example, engagement data, for example, subject properties, i.e., user properties 602a, team properties 602b or account properties 602c, user/team/account features, behaviors and/or journeys 604. The labeled training data which serves as input for the training of the prediction model 242, may also include, for example, segments of users/teams/accounts 606, and/or predicted event labels 608. For example, if a predicted event is a 75 percent chance that the user will purchase chat features for the web conferencing digital product (e.g., not available on the “free” version of the digital product) of the customer 110 with the result for that user being “yes”, the “yes” for the event “probability to purchase the chat feature” is the event label. The prediction model 242 is trained, the training processes represented by block 610.

FIG. 6B shows an example operation of the prediction model 242 for a given subject, i.e., a user, team, or account. Inputs include, for example, engagement data, for example, subject properties, i.e., user properties 632a, team properties 632b, and account properties 632c, and/or user/team/account features, behaviors and/or journeys 634. In addition to the engagement data, the inputs include, for example, one or more segments of the subject, i.e., user, teams, or account 636. The prediction model 242 outputs probabilities of occurrences of the given event at a certain time, for example, at a first time 642a, a second time 642b, through an n¹¹¹ time 642n.

The prescription model 243 applies one or more of: the segment classifications for the one or more uses, or one or more segments of users, the obtained engagement data, the propensity of the user/one or more segments of users to further engage or disengage from a digital product, or provide potential engagement with other digital products, and/or triggers, and prescribes one or more actions for the segmented subjects. For example, a prescribed action associated with the subject engagement with the digital product, which is provided by the prescriptive model 243, includes, for example, allowing the subject continued engagement with the digital product, upgrading of offering the subject an enhanced variant of the digital product, and upgrading or offering the subject an enhanced variant of a different digital product. The model 243 is, for example, a decision tree model, neural network model, or the like.

FIG. 7 shows an example of a prescription model 243. This model is partially trained, in that the segments 704 have been determined previously for the given subject (user/team/account). Inputs include, for example, engagement data, such as subject properties, i.e., user properties 702a, team properties 702b, and account properties 702c, and/or user/team/account features, behaviors, and/or journeys 704. In addition to the engagement data, the input includes one or more segments of users/teams/accounts 706, and optionally (for Option 1 as detailed below and shown in FIG. 8), one or more predictions for an event 708 (the output of the prediction model 242), should data output from the prediction model 242 be used. The prescription model 243 outputs one or more Prescribed Actions, from a first action (represented by Action 1 712a), through a possible n¹¹¹ action (represented by Action n 712n).

FIG. 8 is a diagram of an example data flow or process, formed of multiple processes. Each of the processes is performed by the one or more processors of the CPU 202. One or more of the processes may be performed in real time.

At block 802, the system 102 receives raw data from the customer system 110. The raw data, for example, is generated from subjects (e.g., computers associated with subjects) interacting with the data product of the web-based customer 110. For example, the raw data is captured by the module 220. The raw data collected includes, for example, data of subject (i.e., user/team/account) properties, events and/or application objects, which have been described above.

The process moves to block 804, where the raw data, for example, the received data by the data collection module 220 is preprocessed, by the preprocessing module 222. The preprocessed events (event data) are sent to the real time behavior module 230, which monitors the event data and activates triggers, at block 806 as detailed below.

The preprocessing of block 804 is performed on the received raw data including, for example, data of the subject (e.g., user/team/account) properties, events and/or application objects, and may also include the segment of the subject, as updated by the segmentation model 241, at block 808. The preprocessing results in the raw data processed into engagement data. The engagement data, as detailed above, includes for example, one or more of: 1) subject properties, 2) events, 3) subject features, behaviors and/or journeys (behavioral data), and/or 4) application objects, which is output from the preprocessing module 222 and becomes input for the segmentation model 241, as well as the prediction model 242 and the prescription model, as detailed below. The preprocessing module 222 may also tag event data predicted event(s) output from the prediction model 242, in order to train the prediction model 242, as shown in FIG. 6B, and detailed above.

At block 806 the module 230 monitors user the events (events) from the preprocessed raw data, as passed through the preprocessor 222, to determine whether a trigger should be activated, based on the output of the prescriptive model 243 (upon its being applied at block 812), as detailed below.

The preprocessed data from block 804, for example, 1) the engagement data, such as the respective subject properties of the user, team or account, and/or corresponding user/team/account features, behaviors and/or journeys, and/or 2) any one or more segments associated with the subject, are inputted into the segmentation model 241 (as shown for example, in FIG. 5). The segmentation model 241 is applied to this data at block 808. The segmentation model 241, creates one or more segments for the user/team/account, for example, by categorizing the requisite user into a segment, in accordance with a persona, and the requisite team/account, in accordance with a population, with a probability that the requisite subject is a member of the particular persona segment or population segment.

The segmentation model 241 output is also transmitted back to the preprocessing module 222, as block 808. This output is an updated one or more segments for the requisite subject.

From block 808, with the output from the segmentation model 241, the process may move to block 810 and then to block 812, as a first option (Option 1), or may move directly to block 812 as a second option (Option 2).

As per Option 1, the process moves to block 810, where the output of the segmentation model 241, i.e., the one or more segments and associated probabilities for the subject being in the one or more segments, serves as further input, in addition the engagement data, for example, the data of the subject (user/team/account) properties and/or subject (user/team/account) features, behaviors and/or journeys, from the preprocessing (block 804) (as shown for example, in FIG. 6B), for the prediction model 242. The prediction model 242, as discussed above, provides and outputs predictions for events for the respective subject based on the one or more segments for the subject, for the propensity of the subject to further engage or disengage from a digital product (i.e., the digital product of the customer system 110), or provide potential engagement with other digital products. For example, the event predictions may be the probability of the event occurrence at various times, shown, for example, by blocks 642a-642n in FIG. 6B. Continuing with Option 1, the process moves from block 810 to block 812, where the prescription model 243 is applied to given data. This given data, input into the prescription model 243 includes, for example, one or more of some or all of the preprocessed data, for example, the engagement data such as the data of the subject (user/team/account) properties, and/or the subject (user/team/account) features, behaviors and/or journeys; the output of the segmentation model 241, i.e., the one or more segments and associated probabilities for the subject being in the one or more segments; and/or the output of the prediction model 242, e.g., the one or more probabilities for an event occurrence within a specified time.

The prescription model 243 outputs one or more prescribed actions, such as those actions detailed above, or other actions such as allowing the subject continued engagement with the digital product; upgrading or offering the subject to an enhanced variant of the digital product; and upgrading or offering the subject to an enhanced variant of a different digital product. Still other prescribed actions include, for example, ooffering a trial upgrade for the digital product, offering to sell a premium feature or package of features for the digital product, and the like.

Returning to block 808 and taking Option 2, the process moves to block 812, where the prescription model 243 is applied to given data. This given data, input into the prescription model 243 includes, for example, one or more of some or all of the preprocessed data, including: 1) the engagement data such as the data of the subject (user/team/account) data, and/or the subject (user/team/account) features, behaviors and/or journeys; and/or, 2) the output of the segmentation model 241, i.e., the one or more segments and associated probabilities for the subject being in the one or more segments.

From block 812 of both Option 1 and Option 2, the process moves to block 814 where the outputted prescribed actions 712a-712n are analyzed against triggers, to determine whether a trigger should be activated and thus applied to one or more of the prescribed actions 712a-712n, resulting in one or more actions 814 taken by the system 102. Should the monitoring the data indicative of the one or more events, for example, as performed by the module 230, cause one or more triggers to activate, one or more of the prescribed actions (blocks 712a-712n)(e.g., the one or more prescribed actions corresponding to respective activated trigger(s)) will be taken. For example, the trigger activations and actions taken typically occur in real time.

The present disclosed subject matter is directed to a computer-implemented method for analyzing subject behaviors. The method comprises: accessing by one or more processors, a segmentation model and a prescription model, stored in a program memory; obtaining engagement data associated with a subject engaging with a digital product; providing, to the segmentation model, a first portion of the obtained engagement data, such that the segmentation model classifies the subject into at least one segment; monitoring a second portion of the engagement data associated with the subject engaging with the digital product for a trigger; and, providing the one or more segment classifications for the subject, and the first portion of the engagement data, to the prescription model, wherein the prescription model outputs one or more prescribed actions associated with the subject’s engagement with the digital product, the one or more prescribed actions taken based on an occurrence of the trigger.

Optionally, the computer-implemented method is such that it additionally comprises: accessing by the one or more processors a prediction model, stored in the program memory; and, providing the one or more segment classifications for the subject, and the first portion of the engagement data, to the prediction model, wherein the prediction model determines a probability of additional engagement by the subject with the digital product, and provides the probability to the prescription model.

Optionally, the computer-implemented method is such that the one or more prescribed actions includes: allowing the subject continued engagement with the digital product; upgrading or offering the subject to an enhanced variant of the digital product; and upgrading or offering the subject to an enhanced variant of a different digital product.

Optionally, the computer-implemented is such that the segmentation model applies clustering to classify the subject into the one or more segments.

Optionally, the computer-implemented method is such that the digital product includes one or more of: a software program, a web-service, a data object, platform, and/or a web site, accessible along a communications network.

Optionally, the computer-implemented method is such that, wherein the prediction model to determine the probability of additional engagement by the subject with the digital product, and the prescriptive model, to prescribe the further action for the subject for additional engagement with the digital product, operate in real time.

Optionally, the computer-implemented method, is such that the first portion of the engagement data includes one or more of: a) subject properties; and/or, b) subject features, subject behaviors, and/or subject journeys.

Optionally, the computer-implemented method is such that, the second portion of the engagement data includes data indicative of one or more events.

Optionally, the computer-implemented method is such that, the monitoring the data indicative of the one or more events is performed in real-time to activate the trigger in real time.

Optionally, the computer-implemented method is such that the subject includes a single user or a plurality of users.

Optionally, the computer-implemented method is such that, the plurality of users is associated with at least one of a team or an account.

Optionally, the computer-implemented method is such that the subject includes at least one user, team, and/or account.

Optionally, the computer-implemented method is such that the engagement data is obtained from raw data generated from interactions between the subject, from a computer associated with the subject, and the digital product.

Optionally, the computer-implemented method is such that the segmentation model and the prescription model each comprise: a decision tree model or a neural network model.

Optionally, the computer-implemented method is such that, the prediction model comprises: a decision tree model or a neural network model.

The present disclosed subject matter is directed to a system for analyzing subject behaviors. The system comprises: one or more processors; a program memory storing (1) a segmentation model and a prescription model, stored in a program memory, and (2) executable instructions, that when executed by the one or more processors, cause the system to: obtain engagement data associated with a subject engaging with a digital product; provide, to the segmentation model, a first portion of the obtained engagement data, such that the segmentation model classifies the subject into at least one segment; monitor a second portion of the engagement data associated with the subject engaging with the digital product for a trigger; and, provide the one or more segment classifications for the subject, and the first portion of the engagement data, to the prescription model, wherein the prescription model outputs one or more prescribed actions associated with the subject’s engagement with the digital product, the one or more prescribed actions taken based on an occurrence of the trigger.

Optionally, the system is such that the program memory additionally stores a prediction model, and the executable instructions, that when executed by the one or more processors, cause the system to: provide the one or more segment classifications for the subject, and the first portion of the engagement data, to the prediction model, wherein the prediction model determines a probability of additional engagement by the subject with the digital product, and provides the probability to the prescription model.

Optionally, the system is such that the one or more prescribed actions includes: allowing the subject continued engagement with the digital product; upgrading or offering the subject to an enhanced variant of the digital product; and upgrading or offering the subject to an enhanced variant of a different digital product.

Optionally, the system is such that the segmentation model applies clustering to classify the subject into the one or more segments.

Optionally, the system is such that the digital product includes one or more of: a software program, a web-service, a data object, platform, and/or a web site, accessible along a communications network.

Optionally, the system is such that the prediction model to determine the probability of additional engagement by the subject with the digital product, and the prescriptive model, to prescribe the further action for the subject for additional engagement with the digital product, operate in real time.

Optionally, the system is such that the first portion of the engagement data includes one or more of: a) subject properties; and/or, b) subject features, subject behaviors, and/or subject journeys.

Optionally, the system is such that the second portion of the engagement data includes data indicative of one or more events.

Optionally, the system is such that the monitoring the data indicative of the one or more events is performed in real-time to activate the trigger in real time.

Optionally, the system is such that the subject includes a single user or a plurality of users.

Optionally, the system is such that the plurality of users is associated with at least one of a team or an account. Optionally, the system is such that the subject includes at least one user, team, and/or account.

Optionally, the system is such that the engagement data is obtained from raw data generated from interactions between the subject, from a computer associated with the subject, and the digital product.

Optionally, the system is such that the segmentation model and the prescription model each comprise: a decision tree model or a neural network model.

Optionally, the system is such that the prediction model comprises: a decision tree model or a neural network model.

The present disclosed subject matter is directed to a computer usable non-transitory storage medium having a computer program embodied thereon for causing a suitably programmed system to analyze subject behaviors, by performing the following steps when such program is executed on the system. The steps comprise: accessing a segmentation model and a prescription model, stored in a program memory; obtaining engagement data associated with a subject engaging with a digital product; providing, to the segmentation model, a first portion of the obtained engagement data, such that the segmentation model classifies the subject into at least one segment; monitoring a second portion of the engagement data associated with the subject engaging with the digital product for a trigger; and, providing the one or more segment classifications for the subject, and the first portion of the engagement data, to the prescription model, wherein the prescription model outputs one or more prescribed actions associated with the subject’s engagement with the digital product, the one or more prescribed actions taken based on an occurrence of the trigger.

Optionally, the computer usable non-transitory storage medium is such that the steps additionally comprise: accessing a prediction model, stored in the program memory, and, providing the one or more segment classifications for the subject, and the first portion of the engagement data, to the prediction model, wherein the prediction model determines a probability of additional engagement by the subject with the digital product, and provides the probability to the prescription model.

Optionally, the computer usable non-transitory storage medium is such that the digital product includes one or more of: a software program, a web-service, a data object, platform, and/or a web site, accessible along a communications network.

Optionally, the computer usable non-transitory storage medium is such that the prediction model to determine the probability of additional engagement by the subject with the digital product, and the prescriptive model, to prescribe the further action for the subject for additional engagement with the digital product, operate in real time. Optionally, the computer usable non-transitory storage medium is such that the first portion of the engagement data includes one or more of: a) subject properties; and/or, b) subject features, subject behaviors, and/or subject journeys.

Optionally, the computer usable non-transitory storage medium is such that the second portion of the engagement data includes data indicative of one or more events.

Optionally, the computer usable non-transitory storage medium is such that the monitoring the data indicative of the one or more events is performed in real-time to activate the trigger in real time.

Optionally, the computer usable non-transitory storage medium is such that the subject includes a single user or a plurality of users.

Optionally, the computer usable non-transitory storage medium is such that the plurality of users is associated with at least one of a team or an account.

Optionally, the computer usable non-transitory storage medium is such that the subject includes at least one user, team, and/or account.

Optionally, the computer usable non-transitory storage medium is such that the engagement data is obtained from raw data generated from interactions between the subject, from a computer associated with the subject, and the digital product.

Optionally, the computer usable non-transitory storage medium is such that the segmentation model, the prediction model, and the prescription model each comprise: a decision tree model or a neural network model.

The implementation of the method and/or system of embodiments of the disclosure can involve performing or completing selected tasks manually, automatically, or a combination thereof. Moreover, according to actual instrumentation and equipment of embodiments of the method and/or system of the disclosure, several selected tasks could be implemented by hardware, by software or by firmware or by a combination thereof using an operating system or a cloud-based platform.

For example, hardware for performing selected tasks according to embodiments of the disclosure could be implemented as a chip or a circuit. As software, selected tasks according to embodiments of the disclosure could be implemented as a plurality of software instructions being executed by a computer using any suitable operating system. In an exemplary embodiment of the disclosure, one or more tasks according to exemplary embodiments of method and/or system as described herein are performed by a data processor, such as a computing platform for executing a plurality of instructions. Optionally, the data processor includes a volatile memory for storing instructions and/or data and/or a non-volatile storage, for example, non-transitory storage media such as a magnetic hard-disk and/or removable media, for storing instructions and/or data. Optionally, a network connection is provided as well. A display and/or a user input device such as a keyboard or mouse are optionally provided as well.

For example, any combination of one or more non-transitory computer readable (storage) medium(s) may be utilized in accordance with the above-listed embodiments of the present disclosure. The non-transitory computer readable (storage) medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store, a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including, but not limited to, electromagnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

As will be understood with reference to the paragraphs and the referenced drawings, provided above, various embodiments of computer-implemented methods are provided herein, some of which can be performed by various embodiments of apparatuses and systems described herein and some of which can be performed according to instructions stored in non-transitory computer- readable storage media described herein. Still, some embodiments of computer-implemented methods provided herein can be performed by other apparatuses or systems and can be performed according to instructions stored in computer-readable storage media other than that described herein, as will become apparent to those having skill in the art with reference to the embodiments described herein. Any reference to systems and computer-readable storage media with respect to the following computer-implemented methods is provided for explanatory purposes and is not intended to limit any of such systems and any of such non-transitory computer-readable storage media with regard to embodiments of computer-implemented methods described above. Likewise, any reference to the following computer-implemented methods with respect to systems and computer-readable storage media is provided for explanatory purposes and is not intended to limit any of such computer-implemented methods disclosed herein.

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardwarebased systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. The descriptions of the various embodiments of the present disclosure have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments.

As used herein, the singular form "a", "an" and "the" include plural references unless the context clearly dictates otherwise.

The word “exemplary” is used herein to mean “serving as an example, instance or illustration”. Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments and/or to exclude the incorporation of features from other embodiments.

It is appreciated that certain features of the disclosure, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the disclosure, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable sub-combination or as suitable in any other described embodiment of the disclosure. Certain features described in the context of various embodiments are not to be considered essential features of those embodiments, unless the embodiment is inoperative without those elements.

The above-described processes including portions thereof can be performed by software, hardware and combinations thereof. These processes and portions thereof can be performed by computers, computer-type devices, workstations, cloud-based platforms, processors, micro-processors, other electronic searching tools and memory and other non-transitory storage-type devices associated therewith. The processes and portions thereof can also be embodied in programmable non- transitory storage media, for example, compact discs (CDs) or other discs including magnetic, optical, etc., readable by a machine or the like, or other computer usable storage media, including magnetic, optical, or semiconductor storage, or other source of electronic signals.

The processes (methods) and systems, including components thereof, herein have been described with exemplary reference to specific hardware and software. The processes (methods) have been described as exemplary, whereby specific steps and their order can be omitted and/or changed by persons of ordinary skill in the art to reduce these embodiments to practice without undue experimentation. The processes (methods) and systems have been described in a manner sufficient to enable persons of ordinary skill in the art to readily adapt other hardware and software as may be needed to reduce any of the embodiments to practice without undue experimentation and using conventional techniques.

In the description and claims of the present application, each of the verbs, “comprise,” “include” and “have”, and conjugates thereof, are used to indicate that the object or objects of the verb are not necessarily a complete listing of components, elements or parts of the subject or subjects of the verb.

Descriptions of embodiments of the disclosure in the present application are provided by way of example and are not intended to limit the scope of the disclosure. The described embodiments comprise different features, not all of which are required in all embodiments of the disclosure. Some embodiments utilize only some of the features or possible combinations of the features. Variations of embodiments of the disclosure that are described, and embodiments of the disclosure comprising different combinations of features noted in the described embodiments, will occur to persons of the art. The scope of the disclosure is limited only by the claims.

Claims

1. A computer-implemented method for analyzing subject behaviors comprising: accessing by one or more processors, a segmentation model and a prescription model, stored in a program memory; obtaining engagement data associated with a subject engaging with a digital product; providing, to the segmentation model, a first portion of the obtained engagement data, such that the segmentation model classifies the subject into at least one segment; monitoring a second portion of the engagement data associated with the subject engaging with the digital product for a trigger; and providing the one or more segment classifications for the subject, and the first portion of the engagement data, to the prescription model, wherein the prescription model outputs one or more prescribed actions associated with the subject’s engagement with the digital product, the one or more prescribed actions taken based on an occurrence of the trigger.

2. The computer-implemented method of claim 1, additionally comprising: accessing by the one or more processors a prediction model, stored in the program memory; and providing the one or more segment classifications for the subject, and the first portion of the engagement data, to the prediction model, wherein the prediction model determines a probability of additional engagement by the subject with the digital product, and provides the probability to the prescription model.

3. The computer-implemented method of claim 1 or 2, wherein the one or more prescribed actions includes: allowing the subject continued engagement with the digital product; upgrading or offering the subject to an enhanced variant of the digital product; and upgrading or offering the subject to an enhanced variant of a different digital product.

4. The computer- implemented method of claim 1, wherein the segmentation model applies clustering to classify the subject into the one or more segments.

5. The computer-implemented method of claim 1 or 2, wherein the digital product includes one or more of: a software program, a web-service, a data object, platform, and/or a web site, accessible along a communications network.

6. The computer-implemented method of claim 2, wherein the prediction model to determine the probability of additional engagement by the subject with the digital product, and the prescriptive model, to prescribe the further action for the subject for additional engagement with the digital product, operate in real time.

7. The method of claim 1 or 2, wherein the first portion of the engagement data includes one or more of: a) subject properties; and/or b) subject features, subject behaviors, and/or subject journeys.

8. The computer- implemented method of claim 1, wherein the second portion of the engagement data includes data indicative of one or more events.

9. The computer-implemented method of claim 8, wherein the monitoring the data indicative of the one or more events is performed in real-time to activate the trigger in real time.

10. The computer-implemented method of claim 1 or 2, wherein the subject includes a single user or a plurality of users.

11. The computer- implemented method of claim 10, wherein the plurality of users is associated with at least one of a team or an account.

12. The computer-implemented method of claim 1 or 2, wherein the subject includes at least one user, team, and/or account.

13. The computer- implemented method of claim 1 or 2, wherein the engagement data is obtained from raw data generated from interactions between the subject, from a computer associated with the subject, and the digital product.

14. The computer- implemented method of claim 1, wherein the segmentation model and the prescription model each comprise: a decision tree model or a neural network model.

15. The computer-implemented method of claim 2, wherein the prediction model comprises: a decision tree model or a neural network model.

16. A system for analyzing subject behaviors comprising: one or more processors; a program memory storing (1) a segmentation model and a prescription model, stored in a program memory, and (2) executable instructions, that when executed by the one or more processors, cause the system to: obtain engagement data associated with a subject engaging with a digital product; provide, to the segmentation model, a first portion of the obtained engagement data, such that the segmentation model classifies the subject into at least one segment; monitor a second portion of the engagement data associated with the subject engaging with the digital product for a trigger; and provide the one or more segment classifications for the subject, and the first portion of the engagement data, to the prescription model, wherein the prescription model outputs one or more prescribed actions associated with the subject’s engagement with the digital product, the one or more prescribed actions taken based on an occurrence of the trigger.

17. The system of claim 16, wherein the program memory additionally stores a prediction model, and the executable instructions, that when executed by the one or more processors, cause the system to: provide the one or more segment classifications for the subject, and the first portion of the engagement data, to the prediction model, wherein the prediction model determines a probability of additional engagement by the subject with the digital product, and provides the probability to the prescription model.

18. The system of claim 16 or 17, wherein the one or more prescribed actions includes: allowing the subject continued engagement with the digital product; upgrading or offering the subject to an enhanced variant of the digital product; and upgrading or offering the subject to an enhanced variant of a different digital product.

19. The system of claim 16, wherein the segmentation model applies clustering to classify the subject into the one or more segments.

20. The system of claim 16 or 17, wherein the digital product includes one or more of: a software program, a web-service, a data object, platform, and/or a web site, accessible along a communications network.

21. The system of claim 17, wherein the prediction model to determine the probability of additional engagement by the subject with the digital product, and the prescriptive model, to prescribe the further action for the subject for additional engagement with the digital product, operate in real time.

22. The system of claim 16 or 17, wherein the first portion of the engagement data includes one or more of: a) subject properties; and/or b) subject features, subject behaviors, and/or subject journeys.

23. The system of claim 16, wherein the second portion of the engagement data includes data indicative of one or more events.

24. The system of claim 23, wherein the monitoring the data indicative of the one or more events is performed in real-time to activate the trigger in real time.

25. The system of claim 16 or 17, wherein the subject includes a single user or a plurality of users.

26. The system of claim 25, wherein the plurality of users is associated with at least one of a team or an account.

27. The system of claim 16 or 17, wherein the subject includes at least one user, team, and/or account.

28. The system of claim 16 or 17, wherein the engagement data is obtained from raw data generated from interactions between the subject, from a computer associated with the subject, and the digital product.

29. The system of claim 16, wherein the segmentation model and the prescription model each comprise: a decision tree model or a neural network model.

30. The system of claim 17, wherein the prediction model comprises: a decision tree model or a neural network model.

31. A computer usable non-transitory storage medium having a computer program embodied thereon for causing a suitably programmed system to analyze subject behaviors, by performing the following steps when such program is executed on the system, the steps comprising: accessing a segmentation model and a prescription model, stored in a program memory; obtaining engagement data associated with a subject engaging with a digital product; providing, to the segmentation model, a first portion of the obtained engagement data, such that the segmentation model classifies the subject into at least one segment; monitoring a second portion of the engagement data associated with the subject engaging with the digital product for a trigger; and providing the one or more segment classifications for the subject, and the first portion of the engagement data, to the prescription model, wherein the prescription model outputs one or more prescribed actions associated with the subject’s engagement with the digital product, the one or more prescribed actions taken based on an occurrence of the trigger.

32. The computer usable non-transitory storage medium of claim 31, wherein the steps additionally comprise:

Accessing a prediction model, stored in the program memory, and, providing the one or more segment classifications for the subject, and the first portion of the engagement data, to the prediction model, wherein the prediction model determines a probability of additional engagement by the subject with the digital product, and provides the probability to the prescription model.

33. The computer usable non-transitory storage medium of claim 31 or 32, wherein the digital product includes one or more of: a software program, a web-service, a data object, platform, and/or a web site, accessible along a communications network.

34. The computer usable non-transitory storage medium of claim 32, wherein the prediction model to determine the probability of additional engagement by the subject with the digital product, and the prescriptive model, to prescribe the further action for the subject for additional engagement with the digital product, operate in real time.

35. The computer usable non-transitory storage medium of claim 31 or 32, wherein the first portion of the engagement data includes one or more of: a) subject properties; and/or b) subject features, subject behaviors, and/or subject journeys.

36. The computer usable non-transitory storage medium of claim 35, wherein the second portion of the engagement data includes data indicative of one or more events.

37. The computer usable non-transitory storage medium of claim 36, wherein the monitoring the data indicative of the one or more events is performed in real-time to activate the trigger in real time.

38. The computer usable non-transitory storage medium of claim 31 or 32, wherein the subject includes a single user or a plurality of users.

39. The computer usable non-transitory storage medium of claim 38, wherein the plurality of users is associated with at least one of a team or an account.

40. The computer usable non-transitory storage medium of claim 31 or 32, wherein the subject includes at least one user, team, and/or account.

41. The computer usable non-transitory storage medium of claim 31 or 32, wherein the engagement data is obtained from raw data generated from interactions between the subject, from a computer associated with the subject, and the digital product.

42. The computer usable non-transitory storage medium of claim 32, wherein the segmentation model, the prediction model, and the prescription model each comprise: a decision tree model or a neural network model.