Building a Microsoft Virtual Agent for Customer Support

By - February 25, 2020

Building a Microsoft Virtual Agent for Customer Support

Hi, I’m George Casey here at the Technology Experience Center in Denver, and I want to talk to today a little bit about virtual agents. A lot of companies are using digital transformation to find new and better ways to engage with their customers and virtual agents is a great way to do that. Today I’m going to show you the Microsoft Power Virtual Agent capability, and how we can use this in a low code or no code environment for business users to create an interactive experience through the use of a virtual agent or chat bot.

So today I’m going to demonstrate that with RSM bot. So RSM bot. Hi. Tell me a little bit more. I’d like to learn a little more about blockchain. So the chat bot comes back and engages with me and now it says, sure, “RSM has a blockchain consulting practice with industry leading solutions like ClearThru,” and it offers me the option to learn more from experts. So I would say, yes, I’d like to talk to an expert.

Now it’s going to say, well let me connect you with George Casey and when are you available to talk? And it provides me with some options. So I’m going to select in an hour and it says, “Great, please enter your email and we’ll send you a meeting link in 55 minutes. Talk to you soon.” So I’ll enter and it asks, “Did that answer your question?” I can say yes, and it’s going to give me an option for a quick survey or satisfaction. And I’ll say my experience was excellent.

So let’s take a little bit of a look at the power virtual agent configuration, and show you how this is done. So now I’m in the power virtual agents authoring canvas and this gives me the opportunity to identify things like what are the trigger phrases. And what you’ll notice here is there’s no exact match for the phrase I used in our demonstration, but the system using cognitive services and text analytics can understand and infer what I was looking for. So it drops me down this path.

In this case, because it identified I was looking for blockchain, it let me know that there’s a blockchain solution, it gives me a hyperlink to drill directly to that detail, and prompts me a question that would I like to send additional learning resources. So now we said the next step is a multiple choice option and it gave me those options and you can see I can just edit those text options directly here in the Canvas and then based on those conditions routes to me to one of these three trees, and based on that outcome. In this case, since I said I wanted to expert, it brought up a new topic which we defined as escalate, and the escalate topic allowed us to create an additional flow step which was going to send an email to the user in 55 minutes, like it said, and schedule that appointment on the expert’s outlook calendar. So this is a quick example of how you can very quickly build that message chain or that flow using the Canvas authoring app.

In the analytics pane here we get a dashboard that gives us some key metrics about how customers are engaging with our virtual agent, including the ability to drill exactly on customer satisfaction, where we note that last survey question, how have customers responded to their experience? And you notice of late, this has gone way up and we saw it today, we’ve had one session and it was a perfect score, a 5.0.

So again, the key value here is that we’re taking advantage of this text AI being able to deploy it in this no-code solution that business users can create and have that real time interaction with customers.

Ready to take the next step? Contact us. You may also contact us by telephone: 800.274.3978.

Collaborative leader, data scientist, and problem solver aligning clients with technology and process. Specialties include predictive analytics, marketing automation, CRM, and ERP.

Receive Posts by Email

Subscribe and receive notifications of new posts by email.