Data Insights
  • 26 May 2025
  • 6 Minutes to read
  • PDF

Data Insights

  • PDF

Article summary

The Data Insights enables users to derive actionable intelligence from operational data across ITSM modules such as Incidents, Service Requests, Changes, Knowledge Base, Assets, and more.

It leverages AI-driven analytics, and context-aware responses enabling faster decision-making and smarter service delivery.

Trend Identification

Applicable Channels: Web, MS Teams

Enterprise IT's copilot identifies trends of historical ticket data by leveraging machine learning algorithms and natural language processing (NLP). Copilot uses the AI system aggregating large volumes of past ticket information, which includes details such as issue types, resolution times, affected systems, and customer feedback. By employing machine learning algorithms, Copilot can sift through this data to detect recurring patterns that might be indicative of broader trends.

Additionally, natural language processing (NLP) capabilities enable Copilot to understand the nuances in ticket descriptions and categorize them accurately. This further enhances trend analysis.

The insights gained from this trend identification process can inform decision-making for support teams, allowing them to prioritize and address systemic issues more effectively. This not only improves the efficiency of service management operations but also enhances overall user satisfaction by reducing downtime and improving response times.

Below sample screenshots showcase the Trend Identification capability of Copilot on MS Teams:

User input: Do you see any trend based in the last week Incidents? 

Copilot response:

Figure: Trend Identification sample

User input: Any open tickets with negative sentiment in its symptom or description?

Copilot response:


Figure: Trend Identification

Root Cause Analysis

Applicable Channels: Web, MS Teams

Enterprise AI's Copilot enhances Root Cause Analysis (RCA) by systematically identifying and resolving the underlying issues of Service Requests or Incidents. Leveraging machine learning algorithms and natural language processing (NLP), Copilot analyzes vast amounts of historical data, logs, and previous incident reports to detect patterns and correlations. It filters through this information to pinpoint commonalities and anomalies associated with the reported problem. Once potential root causes are identified, Copilot evaluates them against known solutions and best practices.

By doing so, it not only narrows down the probable cause but also suggests effective remedies. Throughout this process, Copilot continuously learns from new data, refining its analytical capabilities and improving its accuracy in future RCA tasks. This data-driven approach accelerates the resolution process, minimizes downtime, and enhances the overall efficiency of service management operations.

Below sample screenshots showcase the Root Cause Analysis capability of Copilot:

User input: 'Any priority 1 incidents last week?'

Copilot response:

Figure: RCA Analysis - Web

User input: 'Can you summarize the incident INC00**6 with reported problem and solution?'

Copilot response:

Figure: Copilot RCA Analysis - Web

Recommendation System

Applicable Channels: Web, MS Teams

Enterprise IT's Copilot enhances Service Management by providing resolution recommendations based on the analysis of similar tickets and their resolutions. Utilizing machine learning algorithms, Copilot scans a vast database of previously logged tickets. It identifies patterns and commonalities in the issues reported. By examining the context and outcomes of these past tickets, Copilot suggests the most effective solutions for current problems.

This process not only speeds up the resolution time by presenting relevant information and potential fixes to the support agent but also ensures consistency and accuracy in handling recurring issues. Copilot acts as an intelligent assistant, streamlining the support workflow and improving overall service efficiency.

Below sample screenshots showcase the Recommendation System capability of Copilot:

User input: 'Any tickets related to 'data loss due to system crash' historically? If yes, can you state the resolution provided?'

Copilot response:

Figure: Copilot Recommendation System - Web

User input: 'Any tickets related to 'Application login issue' historically? If yes, can you state the resolution provided?'

Copilot response:

Figure: Copilot recommendation system - MS Teams

Predictive Analysis

Applicable Channels: Web, MS Teams

Enterprise IT's Copilot provides predictive analysis by leveraging advanced algorithms and vast datasets to forecast future trends and outcomes. By analyzing historical data, Copilot identifies patterns and correlations that may not be immediately evident to Analysts. This involves processing large volumes of information in real-time, using machine learning models to predict potential issues, opportunities, or changes in various scenarios.

Below sample screenshots showcase the Predictive Analysis capability of Copilot:

User input: 'What are the average daily usage of incidents in the last 1 week and use this to predict the approximate counts for next week?'

Copilot response:

Figure: Copilot Predictive Analysis

Generate Charts

Applicable Channels: Web, MS Teams

Copilot's users have the ability to create dynamic charts - such as pie charts, bar charts, and line charts - based on response data. These charts are automatically determined by the system, but users also have the option to specify the type of chart they want, such as explicitly requesting a bar chart or a pie chart. Importantly, these dynamically generated charts are tailored specifically to the data at hand and are not part of the standard, pre-built ITSM reports. This functionality allows for more flexible and customized data visualization, providing insights that are directly relevant to the user's specific needs and queries.

Below examples and screenshots showcase how you can generate charts through Copilot:

User input:

1st input - 'Show me top 5 P3 running incidents with its logged date in last 3 week'

Copilot response:

Figure: Copilot response

2nd input: 'Using this data can you generate a chart and group it by status?'

Copilot response:

Figure: Generated charts

Knowledge Management Analytics

Applicable Channels: Web, MS Teams

Enterprise IT's Copilot provides insights into the KB articles in the enterprise support environment. Based on the user query Copilot identifies KB articles addressing the issue and consolidates the output according to the user query.

Let’s consider the following use cases:

Analyze Multiple KB articles resolving same issue

A user reached out to the IT virtual assistant with a query on whether there are multiple Knowledge Base (KB) articles resolving the same issues specific to VPN-related problems. Copilot quickly identified several KBs that cover overlapping topics within the domain of VPN support.

Categorized the findings under three distinct VPN support categories: Setting Up VPN, Troubleshooting VPN Issues, and Configuring VPN.

User Input: “_KB_Analysis: Are there multiple KB articles resolving the same issues within a specific to VPN?"

Copilot Response:

Figure: KB Analysis

Identify overlapping knowledge articles related to a single Issue

A user requested an analysis of Knowledge Articles that have overlapping titles or share common keywords related to email issues. Copilot  generated a list of KB articles from the knowledge repository that matched the query. The analysis revealed multiple designed (but not yet published) articles tied to the "email” domain with shared classifications, overlapping terminology, and common workgroups.

User Input: “_KB_Analysis: Generate a list of articles with overlapping titles or keywords from cases involving email?”

Copilot Response:

Figure: Overlapping articles related to Issue

Identifying Expired Knowledge Base Articles Still in Use

The user queries Copilot to identify expired Knowledge Base (KB) articles that are still being referenced. Copilot analyzes KB article and access logs to display articles that are expired and yet continue to be used by the users.

User Input: “_KB_Analysis: Are there expired KB articles that are still being referenced?”

Copilot Response:

Figure: Identify Expired Knowledge Base Articles

Article not found for specific Category

This use case focuses on optimizing the knowledge base by identifying opportunities for consolidation. The user inquires if there are redundant or overlapping Knowledge Base (KB) articles that can be consolidated for a specific category.

User Input:“_KB_Analysis: How many articles could be marked for consolidation in the software installation category?

Copilot Response:

Figure: Article not found

Note

The Copilot confirms that all existing KBs in the Software Installation category are unique, meaning each article serves a distinct purpose without significant content overlap.


Was this article helpful?

Changing your password will log you out immediately. Use the new password to log back in.
First name must have atleast 2 characters. Numbers and special characters are not allowed.
Last name must have atleast 1 characters. Numbers and special characters are not allowed.
Enter a valid email
Enter a valid password
Your profile has been successfully updated.
ESC

Eddy AI, facilitating knowledge discovery through conversational intelligence