Check on Your Agent Using Analytics
Learning Objectives
After completing this unit, you’ll be able to:
- Identify the types of data captured about your agent.
- Explain how Data Cloud relates to agent analytics.
Before You Start
Before you start this module, consider completing this recommended content.
You’ve created your topics, added your actions, and put them to work in a new agent. You even tested it to make sure it acts as expected and activated it. Congrats! But now what? How do you know if your agent is really doing what you intended? And what do your users think of the experience? This is where analytics and monitoring matter.
Monitor and Gather Data
While Salesforce has several methods for gathering and monitoring analytics, agents use tools designed specifically for them. We review the specific types of analytics and tools later. For now, though, let’s discuss agent analytics more generally.
Data Cloud Required for Agent Analytics
Agent analytics are built on Data Cloud reports and objects. In order to use the analytics’ features, Data Cloud must be enabled in your org.
When your agent is active, information is collected about its functions: latency; usage, such as weekly active users; and performance, such as user feedback. You can also find out what your users say to the agent, including what they ask for and if the agent’s response is correct. Regardless of agent type, raw data is gathered using data model objects and stored in Data Cloud.
Use Data Cloud Reports and Dashboards
Your agent’s data is filtered into provided reports, which are then displayed in a dashboard. You access Agentforce Analytics in the Dashboard tab, within the Agentforce (Default) folder. Every dashboard section, or widget, has a link to the underlying report. Select the link to go to the report itself for more detailed information.
What if you want to see other information? Because these are Data Cloud reports, you can modify them as you see fit. However, note that if you directly modify the report, you could break the connection with the data model objects behind the analytics and the original report won’t work. As a best practice, save the report as a new version and then modify that copy.
Creating a presentation about your agent’s analytics? Expand the dashboard widget with the chart or data you want to include, and then download the image. Your slide deck creation just got simpler.
Return to Coral Cloud
Alex Wu is the VP of Business Intelligence at Coral Cloud Resorts and helped drive the decision to add agents to his company. But he knows that agents can’t just be turned on and left on their own. They need to be monitored to make sure they work as intended. To make sure they’re at their peak performance, he plans to use analytics to check up on his agent. We follow Alex to see how he uses the specific types of analytics to monitor his agent and combine that information with other tools to improve agent performance.
Now that you have a general overview of how analytics for your agent work, let’s dive into the types available to you.