Working within the Power Platform – Power Virtual Agents

So, now we’ve focused on 3 of the 4 products that make up the Power Platform – Power BI, Power Automate, and Power Apps. Finally, we turn our focus on the newest member of the Power Platform – Power Virtual Agents.

Power Virtual Agents are chat bots that are created to answer questions from your customers, employees, or visitors to your website. These are low- or no-code agents that can be created without the need of data scientists or developers!

To create a power virtual agent, first you need to decide where the agent will reside.

  1. If your virtual agent will exist within Microsoft Teams, then you’ll be able to create, test, and publish your agent at  https://aka.ms/PVAForTeams (Must have Microsoft Teams for this link to work as it opens Teams)
  2. If your virtual agent will exist on a web page, then you’ll need to create, test, and publish your agent from https://aka.ms/trypva

We will be creating a virtual agent for a webpage, but doing it for Microsoft Teams is basically the same.

After going to the correct page (of those 2 listed above), you’ll need to enter the name of your new chat bot.

Once you’ve got the name you want, click Topics on the left side menu, give the topic a useful name, and finally start typing in 5 – 10 phrases that will be needed to trigger a response from the chat bot.

Once you have all of your trigger phrases entered, be sure to save the topics and then “Go to Authoring Canvas.”

From that point, I edited the workflow answering a few questions and got this lovely diagram for my chat bot.

And with that, we have made our first Virtual Agent (Yes, they can do way more than just simple call and response, but I truly only wanted to show off a quick example of what this can do for you.)

From here, you can publish and try it out with all of your wonderful phrases.

Hopefully, this has made you want to explore what Power Virtual Apps can do for you and your organization. So until the next blog post

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.