Claims:WE CLAIM
1. A method for contextual recommendations in sales engagement context in response to electronic communications 122 received in sales or CRM platform 120 comprising
processing of an inbound electronic communication 122 by a categorizer 131 for sales engagement context and contextual recommendation;
presenting the contextual engagement recommendations 123 by an engagement recommendation module 141 to the users of sales or CRM platform 120.
2. The method for contextual recommendations in sales engagement context in response to electronic communications 122 received in sales or CRM platform 120 as claimed in claim 1, wherein the categorizer 131 comprises a parser 204, sales engagement context category module 206, vector generator 212, context detector 216, contact points extraction module 208, and an update module 224.
3. The method for contextual recommendations in sales engagement context in response to electronic communications 122 received in sales or CRM platform 120 as claimed in claim 1, wherein the parser 204 parses the electronic communication 122.
4. The method for contextual recommendations in sales engagement context in response to electronic communications 122 received in sales or CRM platform 120 as claimed in claim 1, wherein the sales engagement context category module 206 identifies a list of features that are associated with a particular type of sales engagement context.
5. The method for contextual recommendations in sales engagement context in response to electronic communications 122 received in sales or CRM platform 120 as claimed in claim 1, wherein the contact points extraction module 208 detects the appearance of words related to prerequisite engagement details.
6. The method for contextual recommendations in sales engagement context in response to electronic communications 122 received in sales or CRM platform 120 as claimed in claim 1, wherein the vector generator 212 generates a feature vector, which is indicative of each feature of sales engagement context category module 206 that is present in email message 122.
7. The method for contextual recommendations in sales engagement context in response to electronic communications 122 received in sales or CRM platform 120 as claimed in claim 1, wherein the context detector 216 is operative to receive the feature vector generated by sales engagement context category module 206 and the contact points availability indicator 220 generated by the contact points extractor module 208.
8. The method for contextual recommendations in sales engagement context in response to electronic communications 122 received in sales or CRM platform 120 as claimed in claim 1, wherein the update module 224 provides update data to categorizer 131.
9. The method for contextual recommendations in sales engagement context in response to electronic communications 122 received in sales or CRM platform 120 as claimed in claim 1, wherein the engagement recommendation module 141 receives an output from the categorizer 131, engagement recommendations 123 comprising engagement context of the inbound electronic communication 122 and suitable engagement actions to be performed in specific to the engagement context, and in response to the output received, present engagement recommendations 123 to the users of sales or CRM platform 120.
10. A system 400 comprising a network 101, an alphanumeric input device 412, a memory, a processing device coupled to the memory for processing of an inbound electronic communication 122 by a categorizer 131 for sales engagement context and contextual recommendation; and an display device 410 for presenting the contextual engagement recommendations 123 by an engagement recommendation module 141 to the users of sales or CRM platform 120.
, Description:SYSTEM AND METHOD FOR CONTEXTUAL RECOMMENDATIONS IN SALES ENGAGEMENT CONTEXT IN RESPONSE TO ELECTRONIC COMMUNICATIONS
TECHNICAL FIELD
The present disclosure relates to a system 400 and method for detecting sales engagement contexts of inbound electronic communications 122 received in a sales or Customer Relationship Management (CRM) platform 120. More specifically, the disclosure relates to contextual recommendations responsive to electronic communications122 received in a sales or CRM platform 120.
BACKGROUND
In a sales or CRM environment, more difficulties are faced by sales persons in manually analyzing the engagement contexts of inbound email messages received, including deciding the immediate engagement actions to be performed in response and identifying the availability of pre-requisite information (For example contact details, suggested timeline) required to perform the responsive engagement actions.
Manual discovering, extracting, and analyzing data patterns in communication text data and understanding the engagement contexts are not only found to be tedious and time consuming, but most of the time results in less accurate or inconsistent results. Also manual analyzing overlooks or miss to analyze some important parts of the communication text data, thus leading to non- identification of availability of pre-requisite information required to perform the engagement actions and non-capturing of valuable engagement data needed for future engagements.
In a sales or CRM environment, sales persons often receive emails of two natures, one being sent from the sender with no engagement expectations in response and one being expecting some sort of engagement back in response. These messages are not easily identified as such from the subject line that those are engagement expected or not and a user may spend several hours each day opening and reading all messages to identify the ones that needed some sort of engagement back in response. As the inbound emails that required some sort of engagement back bags more priority than those are not, wasting more time in attending non priority emails (i.e those need no response or engagement back) often results in missing to attend some priority emails (i.e those required immediate engagement back) on time.
Also in a sales or CRM environment, sales persons often receive emails with request for scheduled calls or in-person visits with no explicit contact points mentioned to reach back in the email message. On such cases salesperson has to spend effort to check the availability of contact points in the existing contact database available on the sales or CRM platform 120 before scheduling any such engagement response. Incase if the sales person misses to cross check, then it may lead to either cancellation or rescheduling of the engagement planned resulting in poor customer experience and loss of opportunity.
The process becomes more complex, tedious and high degree of productivity loss for sales persons, when there is a huge volume of inbound email messages received at any particular time to process, leading to failures on ensuring on-time, efficient and consistent level of engagements with customers in a sales environment. This in turn leads to loss of prospective customers, as the sales productivity is not concentrated on the core sales activities and wasted on the above redundant & time consuming analyzing activities.
US20150120374A1 discloses automation of customer relationship management (CRM) tasks responsive to electronic communications 122 of manual operations like linking the message to a record, archiving in storage the message in association with a contact record of the CRM application.
In US20160226811A1 a system and method for priority email management is disclosed. The technology relates to techniques which enable an automatic management of emails based on data from both the email database and the customer information database.
US20160063505A1 discloses a Natural language processing (NLP) of follow up to a scheduled event in a customer relationship management (CRM) system. The invention provides for natural language processing (NLP) follow up to a meeting in a calendaring and scheduling (C&S) component of a customer relationship management (CRM) system.
US20190140995A1 discloses methods, systems, and devices for analyzing communication messages (e.g., emails) and selecting corresponding actions by sending the messages to a backend server for analysis using natural language processing (NLP) to classify the message with one or more binary classifications and extract metadata from each message.
Thus, in available sales or CRM platforms 120, automation of manual tasks that are performed in consequence of the receipt of an email message are largely limited to creating new contacts in application or linking incoming email message to non-email records in the application and triggering some static responses. And even though some classification or categorization oriented automations are available, they are implemented to perform very basic or general level categorization (i.e. spam, junk, automatic out of office response, pricing related) or prioritization and do not include effective methods to identify patterns of a communication indicative of sales engagement context expressed in single or combination of direct, slangy or indirect phrases. Hence they do not offer any effective solution to identify deep contextual levels that can comprehend the email content to sense whether the email needs any engagement back or not and, sense the engagement context relevant to specific environment such as sales engagements.
In the light of the aforementioned discussion, there exists a need for a system and method that would overcome or ameliorate the above-mentioned limitations. What is needed is a system and method to leverage additional, deep contextual analysis of the email content with sales engagement background to comprehend the sender’s exact engagement intentions or expectations to determine suitable engagement actions in response.
SUMMARY
The following is a simplified summary of the disclosure in order to provide a basic understanding of some aspects of the disclosure. This summary is not an extensive overview of the disclosure. It is intended to neither identify key or critical elements of the disclosure, nor delineate any scope of the particular implementations of the disclosure or any scope of the claims. Its sole purpose is to present some concepts of the disclosure in a simplified form as a prelude to the more detailed description that is presented later.
Embodiments of the invention effectively provides a novel system and method to detect sales engagement contexts of inbound electronic communications 122 received in a sales or CRM platform 120 and recommend contextual engagement actions to the users of a sales or CRM platform 120.
Briefly described, a method employing aspects of the invention detects an inbound electronic communication, for eg. email message, 122, received from customers on a sales or CRM platform 120. The method includes analyzing one or more features of the electronic communication 122 in the deeper context of sales engagement. The one or more features are indicative of sales engagement intention or context. The method also includes determining the engagement context as a function of analyzed features. The method further includes generating recommendations for contextual engagement actions to be performed in specific to the determined engagement context. The method also includes making the engagement recommendations 123 to the users of the sales or CRM platform 120.
In additional embodiments, one or more processing devices for performing the operations of the above described embodiments are disclosed. In additional embodiments, a system 400 is disclosed, the system 400 comprising a memory; and a processing device, coupled to the memory, for performing operations comprising a method according to any one of the above described implementations.
BRIEF DESCRIPTION OF DRAWINGS
The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles, in which like reference numerals generally refer to the same parts throughout the drawings.
FIG 1: Illustration of a System Architecture 100 of the present disclosure
FIG 2: A block diagram of an exemplary embodiment of a categorizer 131
FIG 3: An exemplary diagram illustrating a process according to one embodiment of the invention
FIG.4: a block diagram illustrating one example of a suitable computing system 400
DETAILED DESCRIPTION
Exemplary embodiments are described with reference to the accompanying drawings. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the spirit and scope of the disclosed embodiments. It is intended that the following detailed description be considered as exemplary only, with the true scope and spirit being indicated by the following claims. Additional illustrative embodiments are listed below.
FIG. 1 illustrates an example of system architecture 100, in accordance with one embodiment of the present disclosure.
The system architecture 100 (also referred to as "system" herein) includes a sales or CRM platform 120, one or more server machines 130 through 140, a data store 102, and client devices 110A-110Z connected to a network 101.
In embodiments, network 101 may include a public network (e.g., the Internet), a private network (e.g., a local area network (LAN) or wide area network (WAN)), a wired network (e.g., Ethernet network), a wireless network (e.g., an 802.11 network or a Wi-Fi network), a cellular network (e.g., a Long Term Evolution (LTE) network), routers, hubs, switches, server computers, and/or a combination hereof.
In embodiments, data store 102 is a persistent storage that is capable of storing electronic communications 122, for example an email message 122, which will be used throughout the specification to illustrate the embodiments, as well as data structures to tag, organize, and index the above data items. Data store 102 may be hosted by one or more storage devices, such as main memory, magnetic or optical storage based disks, tapes or hard drives, NAS, SAN, and so forth. In some embodiments, data store 102 may be a network-attached file server, while in other embodiments data store 102 may be some other type of persistent storage such as an object-oriented database, a relational database, and so forth, that may be hosted by sales or CRM platform 120 or one or more different machines coupled to the server sales or CRM platform 120 via the network 101.
The client devices 110A-110Z may each include computing devices such as personal computers (PCs), laptops, mobile phones, smart phones, tablet computers, netbook computers, network-connected televisions, etc. In some embodiments, client devices 110A through 110Z may also be referred to as "user devices."
In embodiments, each client device includes a client interface 111. In one embodiment, the client interface 111 may be applications that allow users to view or access content, such as email messages 122, web pages, documents, etc. For example, the client interface 111 may be a web browser that can access, retrieve, present, and/or navigate content (e.g., web pages such as Hyper Text Markup Language (HTML) pages, email messages 122, etc.) served by a web server. The client interface 111 may render, display, and/or present the content (e.g., a web page) to a user. In another example, the client interface 111 may be a standalone application (e.g., a mobile application or app) that allows users to view content (e.g., email messages 122, web pages, documents, etc.). According to aspects of the disclosure, the client interface 111 may be a sales or CRM application for users to record, edit, and/or upload data specific to customers and sales transactions on sales or CRM platform 120. As such, the client interface 111 may be provided to the client devices 110A-110Z by the server machine 130-140 or sales or CRM platform 120.
In one embodiment, the sales or CRM platform 120 or server machines 130-140 may be one or more computing devices (such as a rackmount server, a router computer, a server computer, a personal computer, a mainframe computer, a laptop computer, a tablet computer, a desktop computer, etc.), data stores (e.g., hard disks, memories, databases), networks, software components, and/or hardware components that may be used to provide a user with access to content specific to customer, sales transactions, email engagements and engagement recommendations. For example, the sales or CRM platform 120 may allow a sales user to view, add, upload, search for, or comment on customer data 121, electronic communications 122, engagement recommendations 123 and sales data 124. The sales or CRM platform 120 may also include a website (e.g., a web page) or application back-end software that may be used to provide a user with access to the data related to customers, sales transactions, email engagements and engagement recommendations.
In embodiments of the disclosure, a "user" may be represented as a single individual or a sales team. For example, the user could be a sales person or executive who make use of the sales or CRM platform 120 for his day to day sales execution activities.
In some embodiments, sales or CRM platform 120 makes engagement recommendations 123, to a user via client interface 111. The client interface 111 configured and has a designated screen space to present the engagement recommendations 123. Engagement recommendations 123 may be presented as a set of indicators (e.g., interface component, electronic message, recommendation feed, etc.) that includes the determined engagement context of the newly received inbound email message 122 and the recommended engagement actions to be performed in specific to the engagement context.
In one embodiment, the engagement recommendations 123 may be a recommendation for one or more inbound email messages 122 currently being received on the sales or CRM platform 120.
In one embodiment, the server machine 130 includes a categorizer 131. The categorizer 131 detects sales engagement contexts of inbound email messages 122. In general, categorizer 131 may include a sales engagement context category module 206, which may identify combinations of features that are statistically significant in an email message 122 relating to a type of sales engagement context.
One embodiment of the invention advantageously utilizes pattern recognition to create one or more sales engagement context category module 206. For example, a particular sales engagement context category module 206 may be created to identify combinations of statistically significant features of an email message 122 relating to engagement type such as call back, email response or activity schedule (organizing in-person visit, online demonstration or video conferencing), a combination thereof, and so on expected or intended by the sender of the email message 122.
Pattern recognition may create a particular sales engagement context category module 206 by using text classification or other techniques to recognize combinations of statistically significant features (e.g., statistically significant keywords, key phrases, and/ or contextual information). Communications relating to a type of sales engagement context often include some features that are commonly shared among such communications. For example some of the features identified will look like “Our team need another session”, “Looking for more inputs”, “Live demonstration will be preferred”, “Need an in-person demo”, “Get us a revised quote on this”, “Lets connect tomorrow”, “Kindly schedule a demo”, “Get back on this”, “Lets discuss over phone”, “Revert with more details”, “No response yet from your side”, “Evening 4 will be opt for the discussion”.
A sales engagement context category module 206 trained by pattern recognition may be able to identify combinations of statistically significant features that may not be identified by simple keyword matching techniques. In particular, the statistical techniques used by pattern recognition may generalize features based on training samples such that the sales engagement context category module 206 may be able to recognize variations of a given feature. For example, the trained sales engagement context category module 206 with its predefined features mapped to a particular type of sales engagement context, may be able to recognize a slangy or indirect phrase such as “Prod demo might help to proceed further” as relating to a sales engagement context. In contrast, the simple keyword matching techniques employed by known systems and methods may not be able to effectively identify such slang or other phrase variations. Nonetheless, it is contemplated by the present invention that keyword matching may be utilized contemporaneously with pattern recognition to identify more accurately an inbound email message 122 relating to a sales engagement context.
Also if only statistic of the identified features have high significance then it will provide reliable results. And these statically significant patterns may not be identified by simple keyword matching techniques. Also on employing generic keyword matching for identifying significant features, simple lookout of industry specific jargons alone will not yield reliable results, identification of industry specific jargons with significance to the context has to be carried out to achieve a reliable result. For example, if a specific keyword such as “Call” is used to identify callback related sales engagement, then it will identify both “Please call me”, “Yesterday call was useful” as features related to callback, in which ““Yesterday call was useful” has no significance related to the context. Only features with industry specific keywords and that also signifies a need or expectation of call in either explicit or indirect sense such as “Let’s have a call”, “Arrange a call”, “Schedule a call”, “we could discuss on a call”, “a call would be better” has to be identified. This could be achieved only if the module is trained to recognize the specific context such as need or expectation of call as stated above in addition to the specific keywords. Also further to the identification of context linked features, each of the identified features have to be assigned a weightage representing their significance. Without such significance weightage, it will be difficult to relate an email message 122 to a particular type of sales engagement context, when more than one identified features shows relevance to different sales engagement context. Example factors which could help determine the significance are frequency of appearance of identified features in the sample data, explicit or indirect intent expressed in the identified features.
Accordingly, the present invention advantageously utilizes a number of factors to determine a word or phrase indicates or suggests that an inbound email messages 122 relates to a particular type of sales engagement context. Considering any aggregating information available in the email message 122 helps in deciding whether the identified word or phrase truly indicates or suggests that the email message 122 relates to a sales engagement context. If such aggregating information is missed out, then either there will be a failure or inaccuracy on the context identification. Also missing out such aggregation information, you will be able to perform some basic classification but not any deep context classification such as identifying a sales engagement context.
The categorizer 131 may make use of multiple NLP (natural language processing) steps such as but not limited to sentence segmentation, word tokenization, identifying stop words, named entity recognition etc., to process the unstructured email content to detect the sales engagement context.
The categorizer 131 may include a contact points extraction module 208, which may detect the appearance of words related to prerequisite engagement details like specific timeline to schedule, contact number to reach back, location details for in-person meet.
One embodiment of the invention advantageously utilizes text analysis to create contact points extraction module 208. A wide variety of text analysis techniques may be utilized to train the contact points extraction module 208 to detect the appearance of specific words or patterns in the email message 122 related to prerequisite engagement details. As one particular example, text analysis techniques such as keyword extraction may be utilized to automatically extract the most important words and expressions in the email message 122. Keyword extraction may internally make use of regular expression for identifying expected pattern by a given text which can include character, symbol and number. For example given a sentence, regular expression can extract a date such as “Jul29” “4thApril” or phone numbers such as “111-234-1234”.
Based on its analysis of the features of email message 122, categorizer 131 may determine sales engagement context of inbound email message 122 and generates recommendations for engagement actions to be performed in specific to the determined context. An exemplary operation of categorizer 131 in accordance with one embodiment of the invention is discussed hereinafter in connection with Figures 2 and 3.
Server machine 140 includes an engagement recommendation module 141 that provides data (a new inbound email message 122 received in a sales or CRM platform 120 from one of its existing or new customers) as input to the categorizer 131 and obtain output that includes the determined sales engagement context of the received new inbound email message 122 and suitable engagement actions to be performed in specific to the engagement context. The engagement recommendation module 141 further causes the sales or CRM platform 120 to make determined sales engagement context and suitable engagement actions as engagement recommendations 123 to a user via client interface 111. Some operations of engagement recommendation module 141 are described in detail below with respect to Figure 3.
It should be noted that in some other embodiments, the functions of server machines 130, 140 or sales or CRM platform 120 may be provided by a fewer number of machines. For example, in some embodiments server machines 130 and 140 may be integrated into a single machine. In addition, in some embodiments one or more server machines 130 and 140 may be integrated into the sales or CRM platform 120.
In general, the functions described in one embodiment as being performed by the sales or CRM platform 120, server machine 130, or server machine 140 can also be performed on the client devices 110A through 110Z in other embodiments, if appropriate. In addition, the functionality attributed to a particular component can be performed by different or multiple components operating together. The sales or CRM platform 120, server machine 130, or server machine 140 can also be accessed as a service provided to other systems or devices through appropriate application programming interfaces, and thus is not limited to use in websites.
Referring now to FIG. 2, an exemplary embodiment of a categorizer 131 adapted to detect sales engagement contexts of inbound email message 122 according to one embodiment of the invention is shown. FIG. 2 shows categorizer 131 receives an inbound email message 122 as input, as indicated by an arrow 203.
In one exemplary embodiment of the invention, categorizer 131 may include a parser 204. The parser 204 parses the email message 122. For example, parser 204 may break up a content of email message 122 into a group or sequence of constituent tokens. Each token may represent a word, phrase, or any other semantic or syntactic representation of a language.
FIG. 2 shows that the tokens generated by parser 204 are sent to a sales engagement context category module 206, as indicated by an arrow 210. In general, the sales engagement context category module 206 may identify combinations of features that are statistically significant in a communication relating to a particular type of sales engagement context. For example, one or more sales engagement context category module 206 may identify features associated with one or more of the following types of sales engagement context: call back, email response or activity schedule (organizing in-person visit, online demonstration or video conferencing), a combination thereof, and so on.
In one exemplary embodiment of the invention, sales engagement context category module 206 may be a text file identifying a list of features that are associated with a particular type of sales engagement context. For example, sales engagement context category module 206 may be a text file with the following features: “Please callback”, “Lets connect tomorrow”, “Kindly schedule a demo”, “Get back on this”, “Lets discuss over phone”, “Revert with more details”, “No response yet from your side”, “I am generally available in the first-half of the day”, “Evening 4 will be opt for the discussion”. etc. The described features may be associated with one or more types of sales engagement context such as call back, email response or activity schedule (organizing in-person visit, online demonstration or video conferencing). Each feature in sales engagement context category module 206 may also be assigned a weight. The weight of a feature indicates how much consideration is given to the feature when deciding whether email message 122 relates to a particular type of a sales engagement context. In other words, if a given feature frequently appears in an email message 122 relating to the particular type of sales engagement context, then the feature is assigned a greater weight.
A wide variety of training techniques may be utilized to create sales engagement context category module 206. As one particular example, natural language processing based pattern recognition may be utilized to create sales engagement context category module 206 present. Pattern recognition may internally make use of text classification or other techniques to identify features related to sales engagement contexts. The identified features may then be classified as features that are statistically significant in an email message 122 relating to the particular type of sales engagement context. Text classification may be implemented using a rule based classifier where the classification is trained using rules with semantically relevant textual elements (words or phrases) that relate to a type of sales engagement context or machine learning based classifier such as a decision tree, a support vector machine, a content matching classifier etc where the classification is trained using transcripts of actual communications that relate to a type of sales engagement context.
In one embodiment of the invention, sales engagement context category module 206 may examine each token of email message 122 to examine whether the token matches a feature of sales engagement context category module 206 (i.e., a feature that is statistically significant in a communication relating to a type of sales engagement context). If a token of email message 122 matches a feature of sales engagement context category module 206, then sales engagement context category module 206 identifies the feature as included in email message 122.
One or more features of sales engagement context category module 206 identified as included in email message 122 are sent to a vector generator 212 of categorizer 131, as indicated by an arrow 214. Sales engagement context category module 206 may also indicate to the vector generator 212 a frequency that a feature of sales engagement context category module 206 that appears in email message 122 (i.e., the number of tokens of email message 122 that matches the feature of sales engagement context category module 206). The vector generator 212 generates a feature vector, which is indicative of each feature of sales engagement context category module 206 that is present in email message 122 (e.g., as determined by a match between a token of email message 122 and a feature of sales engagement context category module 206). The feature vector may also indicate a frequency that a feature of sales engagement context category module 206 appears in email message 122. The feature vector generated by vector generator 212 may be applied to a context detector 216 of categorizer 131, as indicated by an arrow 218.
FIG. 2 shows that the tokens generated by parser 204 are sent to a contact points extractor module 208, as indicated by an arrow 220. In general, the contact points extractor module 208 may detect the appearance of words related to prerequisite engagement details (specific timeline to schedule, contact number to reach back, location details for in-person meet). Specifically, a distribution of specific words within an electronic communication 122 may also indicate a prerequisite engagement details related to a sales engagement context. For example, appearances of the words “first half of the day” and “coming week” in an email message 122, is relating to specific timeline scheduled mentioned related to a sales engagement context. And on another example, appearances of the numbers in some specific format such as “xxx-xxx-xxxx” relates to contact number mentioned to reach back related to a sales engagement context.
A wide variety of training techniques may be utilized to create contact points extractor module 208. As one particular example, natural language processing based text analysis techniques such as keyword extraction may be utilized to automatically extract the most important words and expressions in the email message 122. Keyword extraction may internally make use of regular expression for identifying expected pattern by a given text which can include character, symbol and number. For example given a sentence, regular expression can extract a date such as “Jul29” “4thApril” or phone numbers such as “111-234-1234”.
In an embodiment of the invention, contact points extractor module 208 may examine each token of electronic communication 122 to detect the appearance of words related to prerequisite engagement details (specific timeline to schedule, contact number to reach back, location details for in-person meet). In response to the examination of tokens for appearance of words related to prerequisite engagement details, contact points extractor module 208 generates an output contact points availability indicator 220, specifying the availability of prerequisite engagement details in the email message 122 to engage back. The output generated by contact points extractor module 208 may be applied to a context detector 216, as indicated by an arrow 220.
Generally, context detector 216 is operative to receive the feature vector generated by sales engagement context category module 206 and the contact points availability indicator 220 generated by the contact points extractor module 208. The context detector 216 may also be configured to cross check with the contact database available with the sales or CRM platform 120 to examine the availability of prerequisite engagement details (contact points of the sender of the received email message 122). Based on the features included in the feature vector and their associated weights (as well as a frequency that a given feature of sales engagement context category module 206 appears in email message 122) and the availability factors of contact points, the context detector 216 applies probability analysis to categorize that email message 122 received related to a particular type of sales engagement context and generates a recommendation of one or more engagement actions specific to the engagement context identified.
In embodiments, output engagement recommendation 123 of categorizer 131 may include one or more sales engagement contexts of inbound email messages 122 and one or more suitable engagement actions to be performed in specific to the determined engagement contexts. The sales engagement contexts and engagement actions are in the form of simple sentences that can be easily understood by the users of a sales or CRM platform 120. For example, the engagement context determined may relate to a call back expected from customer, an email response requested or an activity schedule request to meet or visit in person. For an example in one scenario, if the engagement context of an inbound email is determined to be a request for callback at specific time slot, then the engagement context may be framed as similar to “Callback Request” and the engagement actions as similar to “Customer has requested for call back schedule via email. Please schedule”. In another scenario, if the engagement context of an inbound email is determined to be a request for callback with no mentioning of specific timeslot, then the then the engagement context may be framed as similar to “Callback Request” and the engagement action as similar to “Customer has requested for call back with no schedule mentioned via email. Please check and schedule”.
FIG. 2 further shows an exemplary update module 224 which provides update data to categorizer 131. Particularly, the update module 224 may update categorizer 131 with an updated sales engagement context category module 206 and/or an updated contact points extractor module 208. Updating categorizer 131 might be necessary because the features that are statistically significant in a communication relating to a type of sales engagement context may change over time. In today’s rapidly changing Internet based sales environment, customers constantly find new ways or new languages to communicate with sales persons. Also there are new modern tools or mediums are constantly being introduced for engagement. For example customers often just mention any of the newly introduced communication tool or often use abbreviations as a means to engage (e.g., “lets do a skype this evening”, “lets have a concall”). If categorizer 131 does not have updated information on the way that people communicate related to sales, categorizer 131 might not be able to accurately determine the sales engagement context of the email message 122. By periodically (e.g., once in six months) updating categorizer 131 with an updated sales engagement context category module 206 and/or an updated contact points extractor module 208, categorizer 131 may be able to remain accurate in detecting sales engagement contexts of an email message 122.
FIG. 3 is an exemplary flow diagram illustrating process flow for detecting sales engagement context of an inbound email message 122 and generating recommendations of engagement actions to be performed in specific to the determined engagement context according to one embodiment of the invention. At 302, a categorizer 131 receives an inbound email message 122.
At 304, a content of the email message 122 is broken up into constituent tokens. For example, parser such as parser 204 may be utilized to parse the email message 122 into its constituent tokens. The constituent tokens of the email message 122 are examined to identify whether each feature in a set of predefined features is present in the tokens. A sales engagement context category module 206 may define the set of predefined features. The set of predefined features identifies features that are statistically significant in a communication relating to a type of sales engagement context. Each feature in the set of predefined features may also be associated with a weight indicating how significant the feature is in deciding whether the email message 122 relates to the type of sales engagement context. In response to the examination of the tokens, one embodiment of the invention generates a feature vector associated with the email message 122.
According to an embodiment of the invention, a vector generator such as vector generator 212 may be utilized to generate the feature vector. The feature vector may indicate a presence in the email message 122 of each feature of the set of predefined features. In other words, the feature vector may include the features of the predefined set of features that are included in the email message 122 as well as the weights of the included features. The constituent tokens of the email message 122 are examined to detect the appearance of words related to prerequisite engagement details.
According to the embodiment of the invention, a contact points extractor module 208 may be utilized to detect the appearance of words related to prerequisite engagement details. In response to the examination of the tokens, one embodiment of the invention generates an output contact points availability indicator 220, signifying the availability of contact points in the email message 122 to engage back.
One embodiment of the invention applies the feature vector and the contact points availability indicator 220 to a context detector such as context detector 216. Based on the features included in the feature vector and their associated weights as well as a frequency that a given feature of sales engagement context category module 206 appears in email message 122 and the availability factors of contact points, the context detector 216 applies probability analysis to categorize that email message122 received related to a particular type of sales engagement context.
The engagement recommendation module 141 obtains an output from the categorizer 131 comprising engagement context of the input email message 122 and suitable engagement actions to be performed in specific to the engagement context. Proceeding to 306, in response to the output received, the recommendation module 141 present or make engagement recommendations 123 to the users of sales or CRM platform 120 that comprises of determined engagement context of the new inbound email message 122 and suitable engagement actions to be performed in specific to the engagement context.
FIG. 4 is a block diagram illustrating an exemplary computer system 400, in accordance with an embodiment of the present disclosure. The computer system 400 executes one or more sets of instructions that cause the machine to perform any one or more of the methodologies discussed herein. Set of instructions, instructions, and the like may refer to instructions that, when executed cause computer system 400 to perform one or more operations of categorizer 131 or engagement recommendation module 141. The machine may operate in the capacity of a server130-140 or a client device 110A-110Z in client-server network 101 environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may be a personal computer (PC), a tablet PC, a personal digital assistant (PDA), a mobile telephone, a web application, a server, a network router, switch or bridge, or any machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine. Further, while only a single machine is illustrated, the term "machine" shall also be taken to include any collection of machines that individually or jointly execute the sets of instructions to perform any one or more of the methodologies discussed herein.
The computer system 400 includes a processing device 402, a main memory 404 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM) or Rambus DRAM (RDRAM), etc.), a static memory 406 (e.g., flash memory, static random access memory (SRAM), etc.), and a data storage device 416, which communicate with each other via a bus 408.
The processing device 402 represents one or more general-purpose processing devices such as a microprocessor, central processing unit, or the like. More particularly, the processing device 402 may be a complex instruction set computing (CISC) microprocessor, reduced instruction set computing (RISC) microprocessor, very long instruction word (VLIW) microprocessor, or a processing device implementing other instruction sets or processing devices implementing a combination of instruction sets. The processing device 402 may also be one or more special-purpose processing devices such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), network processor, or the like. The processing device 402 is configured to execute instructions of the system architecture 100 and the categorizer 131 or engagement recommendation module 141 for performing the operations discussed herein.
The computer system 400 may further include a network interface device 422 that provides communication with other machines over a network 101, such as a local area network (LAN), an intranet, an extranet, or the Internet. The computer system 400 also may include a display device 410 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 412 (e.g., a keyboard), a cursor control device 414 (e.g., a mouse), and a signal generation device 420 (e.g., a speaker).
The data storage device 416 may include a non-transitory computer-readable storage medium 424 on which is stored the sets of instructions of the system architecture 100 and of categorizer 131 or engagement recommendation module 141 embodying any one or more of the methodologies or functions described herein. The sets of instructions of the system architecture 100 and of categorizer 131 or of engagement recommendation module 141 may also reside, completely or at least partially, within the main memory 404 and/or within the processing device 402 during execution thereof by the computer system 400, the main memory 404 and the processing device 402 also constituting computer-readable storage media. The sets of instructions may further be transmitted or received over the network 418 via the network interface device 422.
While the example of the computer-readable storage medium 424 is shown as a single medium, the term "computer-readable storage medium" can include a single medium or multiple media (e.g., a centralized or distributed database, and/or associated caches and servers) that store the sets of instructions. The term "computer-readable storage medium" can include any medium that is capable of storing, encoding or carrying a set of instructions for execution by the machine and that cause the machine to perform any one or more of the methodologies of the present disclosure. The term "computer-readable storage medium" can include, but not be limited to, solid-state memories, optical media, and magnetic media.
In the foregoing description, numerous details are set forth. It will be apparent, however, to one of ordinary skill in the art having the benefit of this disclosure, that the present disclosure may be practiced without these specific details. In some instances, well-known structures and devices are shown in block diagram form, rather than in detail, in order to avoid obscuring the present disclosure.
It may be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise, it is appreciated that throughout the description, discussions utilizing terms such as "providing", "receiving", "adjusting", "generating", "obtaining", "determining", “detecting”, or the like, refer to the actions and processes of a computer system 400, or similar electronic computing device, that manipulates and transforms data represented as physical (e.g., electronic) quantities within the computer system 400 memories or registers into other data similarly represented as physical quantities within the computer system 400 memories or registers or other such information storage, transmission or display devices.
The present disclosure also relates to a computer system 400 for performing the operations herein. This system 400 may be specially constructed for the required purposes, or it may include a general purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer readable storage medium, such as, but not limited to, any type of disk including a floppy disk, an optical disk, a compact disc read-only memory (CD-ROM), a magnetic-optical disk, a read-only memory (ROM), a random access memory (RAM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic or optical card, or any type of media suitable for storing electronic instructions.
The words "example" or "exemplar}'" are used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as "example' or "exemplar^'" is not necessarily to be construed as preferred or advantageous over other aspects or designs. Rather, use of the words "example" or "exemplary" is intended to present concepts in a concrete fashion. As used in this application, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or." That is, unless specified otherwise, or clear from context, "X includes A or B" is intended to mean any of the natural inclusive permutations. That is, if X includes A; X includes B; or X includes both A and B, then "X includes A or B" is satisfied under any of the foregoing instances. In addition, the articles "a" and "an" as used in this application and the appended claims may generally be construed to mean "one or more" unless specified otherwise or clear from context to be directed to a singular form. Moreover, use of the term "an embodiment" or "one embodiment" throughout is not intended to mean the same embodiment or embodiment unless described as such.
For simplicity of explanation, methods herein are depicted and described as a series of acts or operations. However, acts in accordance with this disclosure can occur in various orders and/or concurrently, and with other acts not presented and described herein. Furthermore, not all illustrated acts may be required to implement the methods in accordance with the disclosed subject matter. In addition, those skilled in the art will understand and appreciate that the methods could alternatively be represented as a series of interrelated states via a state diagram or events. Additionally, it should be appreciated that the methods disclosed in this specification are capable of being stored on an article of manufacture to facilitate transporting and transferring such methods to computing devices. It is to be understood that the above description is intended to be illustrative, and not restrictive. Other embodiments will be apparent to those of skill in the art upon reading and understanding the above description. The scope of the disclosure may, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.