The New York release of ServiceNow has arrived and with it plenty of new features for customers to get their teeth into. In this post I’m going to focus on the latest advances for the Virtual Agent in New York, along with information on a recent implementation carried out by Engage ESM for a UK Bank and the benefits that this has brought to them.
The Virtual Agent is a conversational chat bot that provides user assistance through a messaging interface. This comes with out of the box conversations for ITSM, HR and CSM which is a great starting point for any customers looking to setup and get value from day one.
Natural Language Understanding
The biggest change in the New York release is the Natural Language Understanding (NLU) capabilities.
NLU models are setup to understand statements that a user might make during a conversation and relate those to a task to perform in ServiceNow. This works the same way as Alexa, Google and Siri when you ask, ‘What’s the weather like today?’ or ‘Can you tell me today’s forecast?’. They understand the intention of the user regardless of the statement and respond appropriately.
In ServiceNow we can use the following example to run through how this new functionality works. Our user Joe Bloggs hasn’t heard back from their recently raised ticket regarding an order for an Apple iPad 3. Joe could pose this question in a number of ways:
The Virtual Agent will use the trained model to understand that the intent of Joe is to ‘Check IT Ticket Status’ and trigger the flow of that conversation. But how does this work under the hood?
NLU has the following 3 step approach to understanding a request from a user and ensuring that the right action in ServiceNow is carried out as an output:
Let’s see this working in the following two screenshots through the chat interface:
You can see that despite the difference in statements made through the chat, the Virtual Agent understands that Joe wants to check his ticket status and provides him with the status of his order.
Why does this matter for customers though? This will reduce frustrations from end-users with Virtual Agent as it will understand the user’s intent much more frequently, making conversations more natural and quicker than ever before. As a result, this will reduce users asking to be routed to a live agent and increase self-service.
In addition, ServiceNow has also released an integration to IBM Watson which will provide the intents, entities, and utterances for the Virtual Agent conversations further reducing the chance that the Virtual Agent does not understand the intent of the end-user.
UK Bank Case Study
Engage ESM have recently completed an implementation of Virtual Agent for one of the big four UK banks. The implementation took around 90 days of effort which was a blended team from both Engage ESM and the customer.
Included in the setup were 33 topics including modifications to the out of the box topics for ITSM. As a result of the implementation the bank has seen the following results:
This has been one of the biggest successes of the year for the bank over a relatively short-term engagement bringing real value and freeing up the desk to work on tickets of a higher complexity.
Our lead consultant had the following thoughts on the overall project:
“Virtual Agent has shown to be a powerful tool for both the service desk and colleagues themselves. It frees up a service desks time by migrating many existing ‘repeatable conversations’ to a chatbot, while also reducing the time for colleagues to work their way through these conversations by providing instant responses. Virtual Agent is flexible in that it is not held back by any existing customizations a customer may have and has proven to, with a well-defined plan, allow for a fast implementation which provides benefits from day one.”
Customers will start to see real benefit in implementing Virtual Agent by reducing the number of low complexity tasks that come into the desk and allowing the Virtual Agent to drive self-service on the platform. It’s great to see the improvements coming from ServiceNow with a real focus on the end-user interactions and I look forward to seeing what’s to come in Orlando!
If you would like to find out more information on anything in this post, please contact us at @engage-esm.com.