System and method for interaction routing by applying predictive analytics and machine learning to web and mobile application context
View Patent ↗A system and method that pre-configure a telecommunication path between two users across a network. A server communicates an Application Programming Interface (API) to a website of an entity which, when accessed by a first user at a remote terminal, loads the API into the user terminal, captures data representative of the interaction of the first user with the website in real time, and communicates the data across the network to the server. The server compares the data with stored attributes of second users to identify a matching second user. The data is analyzed and used to predict a successful outcome between the two users. The server selects a network address of the matched second user, and, upon initiation of a telecommunication with the entity by the first user, routes the telecommunication to the network address and communicates the data to a second terminal of the matched second user.
1. A computer-implemented method of selectively configuring a telecommunication path between users and human agents across a network, the method comprising:
via the network, communicating an Application Programming Interface (“API”) to a web site of an entity;
loading the API when the website is accessed by terminals of respective ones of the users;
with the API, capturing user behavior data for each of the users in real time, the user behavior data describing behavior of a given one of the users while interacting with the website;
communicating the captured user behavior data across the network to remote data processing means via the API;
for each of the users, using the corresponding user behavior data to make an interaction prediction related to when a telecommunication with a given one of the users is predicted to occur;
identifying, based on the interaction prediction, a subgroup of the users for which the telecommunications are predicted to occur sooner than the telecommunications for a remainder of the users not in the subgroup;
triggering, for each of the users in the subgroup, a matching operation comparing the corresponding user behavior data with attributes of a plurality of the human agents at the remote data processing means and matching one of the plurality of the human agents, wherein the matching is based on a pattern that predicts a successful outcome likelihood between the user and the corresponding matched human agent;
comparing the predicted successful outcome likelihoods for the users in the subgroup and, therefrom, identifying at least a first user of the users in the subgroup for prioritization over a second user of the users in the subgroup based on the first user having a higher predicted successful outcome likelihood than the second user;
based on the prioritization, causing an invitation for a telecommunication with the entity to be communicated to the terminal of the first user instead of the second user, the invitation including displaying an activation element related to initiating the telecommunication;
receiving input from the terminal of the first user that activates the activation element and, in response thereto, initiating the telecommunication with the first user by:
at the remote data processing means, selecting a network address that corresponds to a first human agent of the plurality of the human agents matched to the first user;
routing the telecommunication to the selected network address; and
communicating the user behavior data of the first user to a terminal operated by the first human agent for display thereon during the telecommunication;
wherein the captured user behavior data comprises one or more clickstreams and one or more search strings related to the interaction of the user with the website.
2. The method according to claim 1 , wherein the first human agent comprises a human agent in a contact center associated with the web site of the entity.
3. The method according to claim 1 , further comprising achieving, between the first human agent and the first user, a successful outcome related to the successful outcome likelihood, wherein the successful outcome comprises the first human agent acting for the website being browsed by the first user.
4. The method according to claim 1 , wherein the captured user behavior data further comprises at least one selected from website data, website metadata, webpage page tags, website cookies, website page headers, a website navigation record representative of historical or sequential user page and link selections, and user social media data.
5. The method according to claim 1 , wherein the captured user behavior data comprises personae and activity user data.
6. The method according to claim 1 , wherein the pattern used to predict a successful outcome likelihood is based on a machine learning step comprising a clustering technique to cluster specific patterns of captured user behavior data and store over time for continuous comparison with newly captured user behavior data.
7. The method according to claim 1 , wherein the step of routing further comprises selecting a telecommunication protocol and a network path between the first user and the first human agent.
8. The method according to claim 1 , wherein the step of routing further comprises selecting a telecommunication protocol and a network path between the first user and the first human agent and the telecommunication protocol is selected for character-based messaging, audio-only or audio-video telecommunication.
9. The method according to claim 1 , wherein the step of routing further comprises selecting a telecommunication protocol and a network path between the first user and the first human agent and the network path is selected from at least one of VoIP, WebRTC, jabber Guest, PSTN or other network path.
10. The method according to claim 1 , wherein the step of routing further comprises selecting a telecommunication protocol and a network path between the first user and the first human agent and the step of selecting further comprises selecting as a function of the telecommunication functionalities of the terminal of the first user and capabilities of a connection to the network.
11. The method according to claim 1 , wherein the website is a mobile application site or portal associated with the entity capable of communicating with an app downloaded onto a terminal.
12. A system for selectively configuring a telecommunication path between users and human agents across a network, the system comprising:
a processor;
a memory in communication with the processor, the memory storing instructions that, when executed by the processor, causes the processor to perform the steps of:
via the network, communicating an Application Programming Interface (API) to a web site of an entity;
loading the API when the website is accessed by terminals of respective ones of the users;
with the API, capturing user behavior data for each of the users in real time, the user behavior data describing behavior of a given one of the users while interacting with the web site;
communicating the captured user behavior data across the network to remote data processing means via the API;
for each of the users, using the corresponding user behavior data to make an interaction prediction related to when a telecommunication with a given one of the users is predicted to occur;
identifying, based on the interaction prediction, a subgroup of the users for which the telecommunications are predicted to occur sooner than the telecommunications for a remainder of the users not in the subgroup;
triggering, for each of the users in the subgroup, a matching operation comparing the corresponding user behavior data with attributes of a plurality of the human agents at the remote data processing means and matching one of the plurality of the human agents, wherein the matching is based on a pattern that predicts a successful outcome likelihood between the user and the corresponding matched human agent;
comparing the predicted successful outcome likelihoods for the users in the subgroup and, therefrom, identifying at least a first user of the users in the subgroup for prioritization over a second user of the users in the subgroup based on the first user having a higher predicted successful outcome likelihood than the second user;
based on the prioritization, causing an invitation for a telecommunication with the entity to be communicated to the terminal of the first user instead of the second user, the invitation including displaying an activation element related to the telecommunication;
receiving input from the terminal of the first user that activates the activation element and, in response thereto, initiating the telecommunication with the first user by:
at the remote data processing means, selecting a network address that corresponds to a first human agent of the plurality of the human agents matched to the first user;
routing the telecommunication to the selected network address; and
communicating the user behavior data of the first user to a terminal operated by the first human agent for display thereon during the telecommunication;
wherein the captured user behavior data comprises one or more clickstreams and one or more search strings related to the interaction of the user with the website.
13. The system according to claim 12 , wherein the captured user behavior data further comprises at least one selected from web site data, web site metadata, webpage page tags, website cookies, website page headers, a website navigation record representative of historical or sequential user page and link selections, and user social media data.
14. The system according to claim 12 , wherein the captured user behavior data comprises personae and activity user data.
15. The system according to claim 12 , wherein the pattern used to predict a successful outcome likelihood is based on a machine learning step comprising a clustering technique to cluster specific patterns of captured user behavior data and store over time for continuous comparison with newly captured user behavior data.
16. The system according to claim 12 , wherein the step of routing further comprises selecting a telecommunication protocol and a network path between the first user and the first human agent.
17. The system according to claim 12 , wherein the step of routing further comprises selecting a telecommunication protocol and a network path between the first user and the first human agent and the telecommunication protocol is selected for character-based messaging, audio-only or audio-video telecommunication.
18. The system according to claim 12 , wherein the step of routing further comprises selecting a telecommunication protocol and a network path between the first user and the first human agent and the network path is selected from at least one of VoIP, WebRTC, jabber Guest, PSTN or other network path.
19. The system according to claim 12 , wherein the step of routing further comprises selecting a telecommunication protocol and a network path between the first user and the first human agent and the step of selecting further comprises selecting as a function of the telecommunication functionalities of the terminal of the first user and capabilities of a connection to the network.