Analyze Business Performance and Improve Decision Making With Zia Dashboard- Online Help | Zoho Desk

Zia Dashboard: Analyze Business Performance and Improve Decision Making

Businesses track a range of data to monitor their overall performance, make informed decisions, and improve their operations. While the type of data and its usefulness varies from business to business and their goals, some of the most common data measured across industries include: 
  1. Customer Data, which primarily includes the detail about their customer demographic, purchase history, and even customer satisfaction and feedback. 
  2. Marketing Matrix from website traffic to social media engagement, campaign outreach, and ROI. 
  3. Sales and Revenue Data, along with total sales, data on revenue by product or service. 
  4. Employee Performance, including individual's  performance, productivity, efficiency of training programs, employee satisfaction, and so on. 

Importance of measuring business data

Measuring this provides your business with important insights, which in turn allows you to make data driven decisions for: 
  1. Improving daily operations
  2. Maintaining continuity of various business operations.
  3. Identifying bottlenecks
  4. Troubleshooting potential flaws in the process
For example, by knowing which marketing technique is fetching better ROI, your business can allocate more funds for them. Knowing which support channel receives more requests can lead to allocating resources accordingly. Insight on what type of service or product is trending helps stocking inventory. 
Understanding data is the key to making decisions for any business. However, the biggest challenge for any business is to tap this vast information, break them down into comprehensible pieces of information, and eventually process the underlying information. 
This is where a dashboard is useful. A dashboard is a visual representation of data. It takes the available data and represents it in pictorial or graphical format for a comprehensive understanding. 
Zia provides various dashboards, that can help businesses understand and analyze their support operations. 
​
Availability
Info Permission Required
Admins can enable Zia for the organization and users with the permission to access Reports and Dashboards can access Zia Dashboard.
Check Feature Availability and Limits

Zia Dashboards

Admins can enable Zia for the organization, or for a particular department where they want to monitor and analyze business performance. Zia displays two dashboards:
  1. Predictions
  2. Field predictions
To configure Zia
  1. Navigate to Set up page > Zia. 
  2. Toggle Zia option on top, to enable Zia.
To access Zia Dashboard
  1. Navigate to the Analytics module.
  2. On the left panel, under Dashboards, click Zia Dashboard. 

Prediction Dashboard

The Prediction dashboard provides a visual representation of the current trends. The dashboard displays the following components: 
  1. Trends Vs Incoming Responses or Outgoing Responses 
  2. Trending Auto Tags
  3. Sentiment Analysis
  4. Sentiment Trend Analysis

 Zia analyzes the  ticket traffic for  the last 30 days and predicts the trend for current day. When there is a surge or dip in the predicted pattern, it marks it as an anomaly and notifies the assigned agent via Zia Notification Center
The line graph displays the time versus the number of responses in the x and y axis respectively. 

  1. The yellow line shows the predicted trend as observed for the last 30 days.
  1. The blue line represents the actual number of incoming tickets and outgoing responses for that particular day.
  1. Significant deviation between the two data is marked with a star to indicate anomaly. 

Zia Notifications 

Zia monitors your incoming and outgoing ticket responses and proactively notifies you of anomalies such as sudden surge or dip in the pattern. When an anomaly is detected, the notification panel at the bottom of the screen will be highlighted with a red button like icon. The notification also displays the 

The notification panel shows anomalies and alerts. 
  1. Anomalies display the surge or dip in the incoming and outgoing responses.
  2. Alerts display the training status of the Answer bot. 

Notes
Note: The Zia notifications will be available for 15 days from the time anomaly is predicted. After this period it will be removed.
To setup Zia notification
  1. Click the Setup > Zia > Intelligence.
  2. On the Intelligence page, do the following:
    1. Select the department for which you want to enable notifications from the top of the page.
    2. Toggle Zia Notifications.
    3. Under Choose the profile/agent you want to send the notifications select Agents, Teams, Roles, or Roles and Subordinates.
  3. Click Save.

Zia Auto Tag generates tags for tickets after analyzing the content of the ticket. These tags serve as labels or identifiers, helping agents quickly understand the nature of each query.Auto tagging also helps in grouping similar tickets. See also: Understanding Zia Auto Tags 
A word cloud visualization of the tags that have been trending in the tickets, i.e, the words most recurring in the help desk for the past 24 hours. The tags are represented in different sizes and colors. 
  1. Size of the text indicates the number of occurrences of each tag. 
  2. Colors are used for distinction. 
  3. When you hover over each tag, it will also show the number of tickets associated with that tag, as well as the sentiments associated with tickets. 

The tabular format displays more details. It shows the number of tickets and sentiments associated with each tag.

This insight helps the user to identify recurring problems or issues faced by their customers and allows them to address causes, reduce complaints, and ultimately enhance the overall customer experience. 
For example, if "cash deposit" is a recurring tag that is associated with the highest negative sentiment, it might indicate that customers are facing problems with cash deposit. A quick analysis of the tickets can help understand the cause and rectify the issue before it escalates. 

Sentiment analysis 

Alert
Sentiment analysis and sentiment trend analysis does not use the Insights (Setup > Zia > Generative AI > Insights) to generate these dashboards.
Desk’s native sentiment analysis (accessible via Setup > Zia > Intelligence > Sentiment Analysis) evaluates the sentiment of the most recent incoming thread. The sentiment is classified as Positive, Neutral, or Negative, and is determined using key phrases within the ticket. For example:
  1. "Great service" → Positive
  2. "Poor product" → Negative
  3. "Payment failed" → Negative
These thread level keywords are displayed within the ticket and can also be leveraged in workflows to automate responses or trigger specific actions, making it easier to prioritize and address customer concerns effectively.
Having insight into each customer response gives agents an advantage in handling tickets on a case-by-case basis. Zia’s sentiment analysis evaluates the tone of incoming responses in the past 24 hours, categorizing them as positive, neutral, or negative.

For example, a sudden surge in tickets with negative sentiment after introduction of new pricing plan can indicate dissatisfaction and businesses can prepare a contingency plan by offering a limited time discount on the new plan to retain customers.

Sentiment trend analysis

Sentiment trend analysis displays the sentiment of responses in a bar graph format. This detailed view shows sentiment analysis on an hourly basis for the past 24 hours and on a daily basis. 
Having access to insight of both hourly and day-wise breakdown of sentiment analysis helps the team identify patterns and rework on resource distribution and support strategy accordingly. 

For example, during a promotional sale, the team notices an increase in negative sentiment related to time of delivery. 
The customer service manager can resolve it by identifying the peak hours and increasing the number of field agents to reduce delivery time. They also monitor the sentiments in real time to gauge the effectiveness of their plans.

NotesNote: Sentiment analysis provides an overview for the past 24 hours, while sentiment trend analysis gives hourly data.

Field prediction dashboard

Zia can predict values of certain fields based on its training. It can predict the ticket category, priority, issue type, ticket owner, and more, improving assignment and automations. See also: Zia Field Prediction

For example, if Zia predicts incorrect value for the field "Issue Type" in about 500 tickets out of a total of 800 tickets, it indicates that Zia needs to be retrained using the new set of tickets as its analyses needs to be widened with more variety. 
The dashboard shows a comparison of predicted and actual response. 

Incoming vs. Predicted response

By comparing incoming responses with Zia's predicted responses, you can monitor how well the system’s prediction is aligned with the real-time data . 

Incoming vs. Missed predicted responses

This bar graph shows the number of incoming responses versus the number of tickets for which Zia skipped the predictions.
In this case, Zia missed 300 incoming responses that were supposed to have their issue type field updated as Bug, but were instead misclassified as Feature Request. 


All predicted responses vs. Overridden responses

Zia's field prediction capability or efficiency is based on the pattern it understands by training from previously available data, so there may be instances where the predicted field value is not relevant. In such cases, the agent can manually change the field value.
For example, if all high-priority tickets were previously handled by senior support agents, Zia may predict the senior agent as the ticket owner for all high-priority tickets. However, if a high-priority ticket is escalated by a customer and needs to be handled by a manager, the agent or manager can update the ticket owner. In this case, the value predicted by Zia is overridden by the agent. 

To access predictions dashboard
  1. Navigate to the Analytics module.
  2. In the left panel, under Dashboards, click Zia Dashboard. 
  3. Click Field Predictions Dashboard.