The Now Platform® Washington DC release is live. Watch now!
on 09-15-2020 05:00 AM - edited on 12-06-2023 04:31 PM by Victor Chen
If you are brand new to Virtual Agent (VA) and Natural Language Understanding (NLU), then this is the place for you! This guide will help you to get up to speed on activating and building your Virtual Agent, and is a great place to come back to as our products evolve over time.
Check out our Get started with Virtual Agent in the Tokyo release below, still applicable for our latest release with some UI refreshes.
Virtual Agent is a key part of an end-to-end conversational experience delivered by the Now Platform. It’s not just a chatbot — it resolves issues and completes business workflows by understanding your users and personalizing each conversation. And it’s available wherever your users need help: on your service portals, ServiceNow mobile apps, Slack, Microsoft Teams, Workplace by Facebook, or any web page. Organizations can use the Virtual Agent to:
Before Getting Started . . .
Before you get started with the setup and configuration of your first Virtual Agent, you must be prepared with the following things:
Entitlement |
To be able to use Virtual Agent and NLU in a production environment, you must have a subscription that entitles you to it. You must have a Pro or Enterprise subscription (e.g. ITSM Pro, HR Pro, CSM Pro.) If you are not sure, you should check with your ServiceNow account executive. Within the ServiceNow instance, the Subscription Management module will not allow you to activate a plugin for which you are not entitled. |
Service Portal or Employee Center | In order to provide the user with an interactive experience, you must have the Chat Interface installed and configured in the Service Portal or Employee Center. For more information on setting up the Service Portal and adding the chat widget, visit the ServiceNow documentation. |
Required Roles | To follow the installation steps, you will need the appropriate security role for the ServiceNow instance. To install the Virtual Agent and NLU plugins, you will need the admin role. To configure Virtual Agent and NLU, you will need the admin or virtual_agent_admin role. |
Release Strategy | As you start working with the Virtual Agent environments, you will want to consider how you will officially deploy the solution to a production instance. It is highly recommended that you first do all of your work in a sub-prod environment to ensure that everything operates as expected. Then you should go through your company's normal release processes to make the configuration changes in production. |
Service Catalog and Knowledge Base | The Virtual Agent provides users with service catalog items and knowledge base articles in conversations regarding questions or submitting a request. It is recommend that customers have a thorough service catalog and knowledge base in place that the Virtual Agent can search through and surface relevant items to users. |
Steps to quick success |
Step 0: Virtual Agent terminology and architectureFirst, get a behind-the-scenes look at the building blocks of Virtual Agent.
Step 1: Activating Virtual AgentDetermine and install the plugins you need in order to start working with Virtual Agent and NLU. Then, install pre-built conversations from ITSM, HR, etc.
Step 2: Publishing Virtual Agent ConversationsIt is time to create conversations that will be supported by your new Virtual Agent. This is how you create a personalized experience with greetings and promoted topics. Then you can publish topics that address and automate your users' issues.
Step 3: Enabling NLU In Your ConversationsNatural Language Understanding, or NLU, can be enabled to improve the overall experience for end-users. NLU allows end-users to engage with the Virtual Agent using natural sentences and not only improves the Virtual Agent's user experience, but it also increases the Virtual Agent's accuracy when presenting topics. The great part is that you don't need to be a data scientist - you can build and train NLU models with low-code tools, and one-click training and testing.
Step 4: How to Extend the OOTB ITSM NLU ModelServiceNow offers several out of the box (OOTB) conversations for ITSM that are designed to get you up and running with little effort. In this article, you’ll learn how to create an NLU Model, import an OOTB Intent, and extend it to meet your needs.
Step 5: Model TuningIn this article, we'll walk through how to identify good vs bad utterances, understanding model accuracy, and tuning techniques.
Bonus: Get (micro) certified!Enroll in Virtual Agent Fundamentals, Implementation, and Natural Language Understanding on Now Learning to learn how to use Virtual Agent to design communications that help your users quickly obtain information, make decisions, and perform everyday work tasks like HR request, or customer service questions. |
More Recommended Practices | |
Achieve desired business outcomes with Virtual Agent using your goldmine of dataWhat do you automate first using the Virtual Agent? You likely have the answer right in front of you. It’s in your data! Continue reading... |
|
Tips for Implementation successTips for a success implementation of your Virtual Agent and NLU. Continue reading... |
|
7 habits for an effective Virtual AgentIn order to maximize the Virtual Agent's ROI, it's critical to remove friction in the end-user experience to drive up the adoption and self-service. Continue reading... |
|
Marrying your IT and HR Virtual AgentsLearn how your organization can manage separate IT and HR Virtual Agents, with respects to separate topics and data, while maintaining a consistent overall user experience. Continue reading... |
|
Implementation guides and Readiness ChecklistImplementation and project management guide for VA and Service Portal or MS Teams Implementation readiness checklist VA Implementation checklist - a deeper dive Virtual Agent and NLU Deployment Guide |
Once you've created a great Virtual Agent experience with your users, it's time to enhance the understanding capabilities with Natural Language Understanding (NLU).
Enable NLU with Virtual Agent |
Migrating Virtual Agent to Natural Language UnderstandingRecommended practices designed to help you migrate your existing Virtual Agent topics to work with NLU. Continue reading... |
Virtual Agent from Keywords to NLUNatural language understanding (NLU) is a form of artificial intelligence that enables your employees and customers to make requests of Virtual Agent the exact same way they would ask another human. It works by first recognizing what the user wants to do and then extracts the supporting details needed to fulfill the request. Enroll to watch a short overview video... If you want to use keywords and NLU together, for instance, with languages not optimized for NLU, see here for an example. Continue reading...
|
How do you improve the Virtual Agent from failed utterances?This article explains where you can find the users' utterances' and utilize them to improve the Virtual Agent experience. As the Virtual Agent admin, you'll want to review these utterances as they will serve as a barometer for the quality of the current NLU model. Continue reading... |
Use NLU to Improve Virtual Agent Success RatesHands on K20 lab session covering how to use Natural Language Understanding (NLU) to improve Virtual Agent success rates. Free to enroll and an instance is provided! Enroll here... |
Virtual Agent and NLU In ActionGet to know VA and NLU a bit better as we walk through the products and see how together they create a great user experience without the need of human intervention. Watch this short overview video... |
Virtual Agent Academy on NLU Model TuningLearn about model tuning and get tips and tricks. Watch it... |
References |
|
Latest release notes
|
Documentation |
|
FAQsFor FAQs regarding Virtual Agent and NLU: |
Incredible write-up and great links / overview!