With the **Virtual Agent Dashboard Calls** and **Virtual Agent Dashboard Chats**
dashboards,


evaluate the performance of your virtual agents to find areas for improvement.

## Use cases

The virtual agent dashboards can help with the following:

- **Performance monitoring**: analyze real-time and historical data on virtual
  agent performance, such as the number of interactions handled, average
  response time, and customer satisfaction ratings. This lets you monitor and
  assess the effectiveness of virtual agents and identify areas for
  improvement.

- **Issue detection and resolution**: identify patterns or trends in end-user
  interactions where virtual agents might struggle to understand requests or
  provide accurate responses. This can help you identify issues and take
  corrective actions, such as updating virtual agent scripts or providing
  additional training.

- **End user insights**: gain valuable insights into end-user behavior,
  preferences, and pain points by analyzing data from virtual agent
  interactions. You can use this information to better understand end-user
  needs, tailor virtual agent responses, and improve the overall end-user
  experience.

- **Resource allocation**: analyze data on virtual agent workload,
  availability, and performance, enabling you to optimize resource allocation.
  For example, you can identify peak hours when additional virtual agents
  might be required to handle increased inquiries or assign virtual agents
  based on their expertise.

- **ROI analysis**: measure the return on investment (ROI) of virtual agents
  by analyzing data on their performance, cost savings, and customer
  satisfaction. This can help you assess the value and effectiveness of
  virtual agents in achieving business goals.

## Use the dashboards

To use the dashboards, follow these steps:

1. In the CCAI Platform portal, click **Dashboard \> Advanced
   Reporting** . If you don't see the **Dashboard** menu, expand the window
   horizontally until the **Dashboard** menu appears. The **Advanced Reporting
   Landing Page** appears.

2. Click **Virtual Agent / Calls** or **Virtual Agent / Chats**. The dashboard
   appears.

3. Click the **Date** field to select a date range, and then do one of the
   following:

   - Select a preset data range. To do this follow these steps:

     1. Click **More** to
        view all of the preset date ranges.

     2. Click the date range that you want.

   - Select a custom range. To do this, follow these steps:

     1. Click **Custom**.

     2. In the **Custom** field, enter a start and end date for your date
        range. You can type your date range or click ![Click the calendar icon to see the
        calendar.](https://docs.cloud.google.com/static/contact-center/ccai-platform/docs/images/icon-calendar.png) **Open calendar** to select start and end dates.

        > [!NOTE]
        > **Note:** The maximum date range that you can specify is 45 days. If you specify a date range of more than 45 days, the dashboard returns an error.

4. Filter your results using the following fields:

   - **Start Time** and **End Time**: Get results from only within this time
     range.

   - **Virtual Agent**: Filter by virtual agent.

   - **Queue**: Filter by queue.

   - **VA Escalation in Queue**: Filter by whether there was a virtual agent
     escalation in the queue.

5. Click ![Click the update icon to update.](https://docs.cloud.google.com/static/contact-center/ccai-platform/docs/images/icon-update.png) **Update**.

## Tiles

These dashboards contain the following metrics tiles for both calls and chats.

- **Total Virtual Agent Interactions**: the sum of interactions in which the
  virtual agent was presented to end-users. For example, if an end-user
  contacts a support center and interacts with one virtual agent and then is
  transferred to another queue where they interact with another virtual agent,
  then that would include two virtual agent interactions.

- **Total Human Agent Initiated Interactions**: the sum of interactions that
  were transferred from the virtual agent to a human agent for further
  assistance

- **Avg Messages per Session**: the average number of messages per session

- **Total Messages Exchanged**: the sum of messages that were exchanged
  between the end user and the virtual agent

- **Resolution Rate %**: the percentage of virtual agent interactions that
  were successfully resolved by the virtual agent

- **Total Consumer Initiated Interactions**: the sum of interactions that were
  started by a consumers

- **Total VA In-Queue Interactions** (calls only): the number of calls where
  wait-time virtual agents were active while end-users were in a queue

- **Total VA In-Queue Time** (calls only): the total time that wait-time
  virtual agents were active while end-users were in a queue

- **Avg VA In-Queue Time** (calls only): the average time that wait-time
  virtual agents were active while end-users were in a queue

- **Virtual Agent Interactions**: provides a graphical representation of the
  volume of virtual agent interactions, so that you can see the usage patterns
  and trends over time. It can be a valuable tool in assessing the
  effectiveness of virtual agents in handling consumer interactions and
  optimizing their deployment in your contact center. The chart is displayed
  in 30 minute intervals, providing insights into the usage of virtual agents
  throughout the day. If you select a date range of more than one day in the
  reporting dashboard, the chart will automatically aggregate the data to
  display the total number of virtual agent interactions by day.

- **Resolved by Virtual Agent**: indicates the number of calls or chats that
  were resolved by the virtual agent, meaning the virtual agent was able to
  successfully handle and resolve the issue without escalation.

- **Planned Transfer**: the occurrences when the virtual agent has answered a
  call or chat from the queue and escalates it to a human agent in the same
  queue. For example, the virtual agent understood what the consumer was
  requesting and sent them to the proper queue/agent/phone number/SIP
  Endpoint.

- **Escalated**: the sum of interactions where the virtual agent was unable to
  handle the issue and it was escalated to a queue where a human agent is
  assigned. This could be due to the complexity of the issue or the virtual
  agent's inability to understand the request.

- **Total Abandons** The number of virtual agent call sessions where the end
  user left before the virtual agent could resolve the call or escalate to a
  live agent.

- **Avg Abandon Time**: the average amount of time that calls or chats waited
  in the queue before being disconnected by the caller without being accepted
  by an agent. This helps you assess the efficiency of your queue management
  and optimize the wait times for better customer experience.

- **Avg Handle Time**: the average call duration in seconds, plus the average
  time spent in wrap-up time. AHT starts when the call is assigned to the
  agent, not when the call is connected to the agent.

- **Avg Sentiment Score**: the average sentiment score across all interactions
  handled by the virtual agent. Sentiment analysis helps you gauge the
  customer's emotional state during the interaction and assess the overall
  satisfaction level.

- **CSAT**: the customer satisfaction scores (CSAT) awarded to the entire call
  or chat interaction. CSAT scores are captured after the interaction ends and
  provide insights into how satisfied customers were with the virtual agent
  service. This metric is displayed as a gauge, allowing you to quickly assess
  the level of customer satisfaction. This only applies if the VA resolved the
  interaction.

- **Fallback Response Rate**: the percentage of virtual agent interactions
  where the virtual agent states that they did not understand the consumer's
  query, and the interaction resulted in a fallback response. It is calculated
  as the number of default fallback responses divided by the total number of
  virtual agent responses, multiplied by 100.

- **Fallback Response Count**: the total number of virtual agent default
  fallback responses. It is the sum of interactions where the virtual agent
  was unable to understand the consumer's query and had to fallback to a
  default response.

- **Fallback Responses**: a visual representation of when virtual agents fail
  to resolve issues, resulting in escalations to human agents. If you select a
  date range of more than one day, the chart shows the total number of
  fallback responses by day. Click a day to get additional data

## Tables

These dashboards contain the following metrics tables. Hold the pointer over a
table heading to see its description.

- **Virtual Agent Performance Metrics Table**: the table displays the metrics
  for each virtual agent

- **Total Interactions per Virtual Agent**: the total interactions for each
  virtual agent

- **Resolution Rate per Virtual Agent**: the resolution rate for each virtual
  agent

- **Top Queues**: the virtual agent in the top queues

- **Virtual Agent Status**: the status of the virtual agent