Genesys Voicebots (CE41) for Genesys Engage cloud

From Genesys Documentation
Jump to: navigation, search
This topic is part of the manual Genesys Engage cloud Use Cases for version Current of Genesys Use Cases.
Read this topic for other versions:
Use voicebots to automate customer conversations and seamlessly hand over to an agent if needed.

Use Case Overview

Story and Business Context

Successful companies strive to provide good experiences for their customers, but doing so becomes challenging and expensive given the multitude of communications channels (voice and digital) that must be offered to interact with customers. It is increasingly expensive to serve customers with live agents and companies need to optimize efficiencies in cases when requests can be fully or partially automated and serve customers faster and more economically. Customer Service costs are sustainable for self-service channels like in-app, web self-service, and virtual assistant but they begin to increase with web chat and exponentially skyrocket with voice. The challenge to customer service leaders is to provide outstanding experiences for customers by utilizing automation and artificial intelligence blended with human interaction where required.


Blending bot technology with voice, self-service and digital applications enables companies to optimize service without incurring costs associated with live customer interactions. Designing customer experiences with voice bots delivers the following:

  • Ability to manage the customer entire journey, automated wherever possible, human touch when needed, for example, chatbots, voicebots, and Altocloud
  • Create personalized bot experiences by sharing context across voice and digital channels and with agents
  • Voicebots in general provide significantly higher benefits than a traditional IVR due to more advancements in speech recognition and the introduction of Conversational AI which promote a more natural interaction providing a high-quality customer experience
  • Quickly and easily deploy omnichannel bots with Natural Language Understanding (NLU) and pre-built microapps that leverage industry best practice
  • Move to cloud at your own speed and at low risk with cloud, hybrid, and premise options
  • Inclusion of voicebots powered by Natural Language Understanding from Genesys Dialog Engine, Google DialogFlow and Amazon Lex to provide natural and rich conversational experiences for customers who call Customer Service. Spoken phrases from customers are transcribed to text by Google Cloud Speech to Text which is then interpreted by the chosen NLU application to identify the customer’s intent, for example, book a flight; extract key data for the query such as origin and destination city, date, day, time; and generate an appropriate response and/or further questions to continue the interaction.

Use Case Benefits*

Use Case Benefits Explanation
Improved First Contact Resolution Present a customer experience that is tailored to the individual based on who they are, why they might be interacting, and the status of the contact center.
Improved Net Promoter Score Reduce time to serve, which in turn improves Net Promoter Score (NPS). Improve self-service containment rates, handle time, and overall customer experience by enabling routing to the right self-service application by correctly identifying intent.
Reduced IT Operational Costs Reuse of existing assets and the option to use less expensive speech alternatives
Reduced Volume of Interactions Increase self-service interactions results in reduced agent-assisted interactions for repetitive or common requests.
*You can sort all use cases according to their stated benefits here: Sort by benefits

Summary

Genesys supports a “design once, deploy anywhere” concept for AI-driven bots to enable organizations to provide a seamless customer experience across voice and digital channels. This use case focuses on deploying a bot on the voice channel and can automatically be extended to digital channels by adding the Genesys Chatbots with Blended AI Use Case. The voicebot supports and orchestrates the following capabilities:

  • Personalization – to tailor the experience based on context from the current interaction or from previous interactions
  • Natural Language Understanding (NLU) – to derive intents and entities
  • Intent classification utilizes NLU to classify the customer's intent with high accuracy, and connect the customer to the correct self-service process or agent. Also, NLU derives entities which are pieces of relevant information from customer utterances such as: account numbers, amounts, and other parameters that are returned to complete a process corresponding to the intent thus enabling a conversational experience.
  • Identification & Verification (ID&V) – to identify and verify the customer if necessary
  • Directed Dialog – to automate relevant and structured business processes or provide information
  • Involve another NLU/AI platform – if it specializes in a particular topic
  • Hand off to a voice agent – to connect the customer to a live person with the full context of the interaction
  • Offer and schedule a callback - when outside of business hours or long wait time then the voice bot offers an immediate or scheduled callback from the voice channel
  • Offer a voicebot survey depending on business context


Use Case Definition

Business Flow

Business Flow Description

  1. A voice interaction is initiated (reactive or proactive) on the voice channel.
  2. The customer hears a standard welcome message from the voicebot.
  3. Customer information and/or context is retrieved from:
    • Customer profile information in UCS
    • Genesys User Data
    • API call to third-party data source
  4. The customer hears a personalized message or menu or is handed over to a voice agent. Examples include:
    • Custom voice message or update, such as "Your next order is due to be delivered on Thursday before 12"
    • Most likely contact reason, such as "Do you want to find out about the loan application you have in progress?"
    • Customer is provided personalized menu options.
    • Customer is routed over directly to a voice agent if, for example, they have an overdue payment or a callback is scheduled. If the customer is not handed over to a voice agent, then at this point the customer could end their call, confirm the contact reason, or continue.
  5. Once the customer has moved on from the Personalization stage, the voicebot asks an open-ended question like: “How may I help you?” to determine intent (and entities) and capture the customer's response.
    • The customer speaks to the voicebot in natural language. For example, the customer could say “I missed a bill payment and want to check my balance because I think I'm in arrears”
    • The voicebot listens to the customer and converts the raw spoken utterance into text.
  6. The voicebot then sends the customer's response to the configured NLU engine (for example, Dialog Engine). If intent and entities are returned, the conversation moves to the correct point in the interaction flow, which could be within one of the following subflows.
  7. Identification and Verification (ID&V)
    • Automated business process (such as Payment Collection microapp).
    • Handoff to live voice agent or schedule a callback (see the relevant use case for the channel).
    • If intent and entities are not returned, the voice bot plays a prompt such as: "Sorry, I didn’t understand your question. Please ask me another question or say AGENT for live assistance."
    • Fall back to DTMF is an optional capability that can be quoted by Professional Services separately as it is not included in the scope of this use case.
  8. Upon completion of a task, the voicebot asks a follow-up question like: "Is there anything else I can help you with?"
    • If the customer says something like “yes”, they're brought back to Step 5: "How may I help you?”
    • If the customer says something like “no”, the voicebot decides whether to offer them a survey (see the next step). If the customer responds with a more advanced answer, the NLU engine determines intent and entities for further processing.
    • Sources include:
      • Customer profile information in UCS
      • API call to third-party data source
      • Logic defined in Designer
    • Customer information and/or context is retrieved to determine whether to offer a survey.
      • If a survey is to be offered, the voicebot continues to the step
      • If no survey is to be offered, the voicebot shows a goodbye message and ends.
    • The voicebot asks the customer: "Would you like to participate in our survey?
      • If the customer answers "yes", then they continue to the next step and engage in a survey.
      • If the customer answers "no", then they are shown a goodbye message and the voicebot ends.
    • The survey is executed using the Designer survey microapp.
      • Optional: If the survey results meet a certain criteria based on the configured evaluation parameters, a specific action can be taken. For example, if the customer provides a negative response, they can be routed to a live agent.

Business and Distribution Logic

Business Logic

BL1: Agent Handoff: The customer can ask to be connected to an available voice agent. Other context can also be displayed as Case Data. Optionally, this use case could be deployed along with Predictive Routing to identify the best agent.


BL2: Retries: The number of retries for self-service tasks and questions can be configured by a business user. Upon, maximum retries the dialog can be configured to play a message, hand off to a voice agent, or offer a callback if busy or outside of business hours.


BL3: Response Type: The interaction flows can be configured to accept natural language spoken responses and closed spoken responses such as account number, date of birth, and yes/no questions.


BL4: Callback: If outside of business hours, or estimated wait time (EWT) is high, the voicebot can offer an immediate or scheduled callback. If this option is not included, then a message is played back to the caller that a transfer is not possible.


BL5: Survey: The customer can determine whether to address a survey or not. Surveys can be based on:

  1. Customer profile information in UCS
  2. Genesys User Data
  3. API call to third-party data source

Parameters Influencing Voicebot Behavior 'This Use Case is supported across industry verticals. The basic features of Voicebot business logic such as personalization are parametrized. Example parameters include:

  • Personalization
  • Segmentation, offer management, characteristics, and most likely contact reason
  • Intents - the goal of the interaction - for example, a "pay_bill" intent returned by Dialog Engine would indicate that the customer should be presented with an authentication business process followed by a payment business process
  • Entities - other pieces of key information returned by Dialog Engine. Entities can accelerate the conversation by pre-populating answers to subsequent questions
  • Confidence levels

Agent Handoff:

  • Based on user choice, such as "I want to speak to an advisor"
  • Based on default handling, such as retries, timeouts, global commands
  • Based on application logic, such as customer owes money and application decides to transfer

Callback

  • Estimated wait time to determine whether to offer callback

Survey

  • Based on context from UCS, Genesys User Data, or third-party web service


Distribution Logic

When the conversation is handed over to a live agent, the interaction moves to one of these use cases, depending on the channel the customer is using - see the Use Case Interdependencies section for related use cases. The use case interdependencies include use cases not yet available in Genesys Engage cloud.


Use Case Requirements

Customer Interface Requirements

N/A

Agent Desktop Requirements

N/A

Reporting

Real-time Reporting

Key real-time metrics that can be included in a report are:

  • Agent Group capacity for interactions to define whether to offer escalation to customer service
  • Current interactions waiting in the system
  • Total interactions (self-service vs assisted service)
  • Agent Group Status

Historical Reporting

Standard voice reporting tools are used to measure operational performance of the voicebot.

More reporting may be available in the AI/NLU tool.

Assumptions

General Assumptions

  • Use Case supports voice channel. Genesys Chatbots (CE31) for GenesysEngage-onpremises (Cloud) supports digital channels.
  • This Use Case is supported by industry templates that contain examples of voicebot applications combining personalization, natural language understanding, AI, and microapps. Voicebot application requirements including required microapps are confirmed during design. These application templates are created for Financial Services, Telco, and Travel.
  • Handoff to voice agent is on the same channel.
  • Dialog Engine has been "trained" with intents and classifications.
  • Dialog Engine provides English language only support for NLU capabilities.
  • NLU capabilities for other non-English languages can be supported through third-party NLU engines such as wit.ai. Integration to third-party NLU/AI engines is a customization task. Dialogs that do not require NLU can support any language by using Personas.
  • Callback dialog flow is provided by the Designer Callback application.
  • Chat transcript is not passed to callback agent.
  • Survey dialog flow is provided by the Designer Questionnaire Builder microapp.
  • Designer is used configure voicebots, and this User Interface tool is localized to support English (United States)

Customer Assumptions

Interdependencies

All required, alternate, and optional use cases are listed here, as well as any exceptions.

All of the following required: At least one of the following required: Optional Exceptions

    Inbound

      None

        Inbound

          None

          Premise Assumptions

          Cloud Assumptions

          • The voicebot is deployed in a Hybrid model, that is, Genesys Engage infrastructure on premise with Google components in the cloud.


          Related Documentation


          Document Version

          • 1.0.3