Name: Siddhant Chauhan

Roll Number: 22135125

Branch: Mechanical
Title: Internship at Segment Analytics Pvt. Ltd.
Role: Data Analyst Intern
Duration: May 25 – July 15, 2025 (Remote)
Company Location: Noida, Uttar Pradesh
1. Introduction about the Company
Segment Analytics Private Limited is a dynamic and rapidly evolving
data analytics firm based in Noida, Uttar Pradesh. The company is
committed to transforming raw business data into valuable insights
that assist organizations in making informed decisions. It specializes
in business intelligence solutions, focusing on data analysis,
reporting, and visualization services across various sectors such as
retail, finance, human resources, and logistics.
Segment Analytics uses a combination of cutting-edge tools and
technologies, including Python, SQL, Microsoft Excel, and Power BI,
to analyse structured and semi-structured data. The company’s
collaborative and agile environment fosters innovation and
encourages interns to take initiative and contribute meaningfully to
ongoing projects.
What sets Segment Analytics apart is its emphasis on real-world
impact, client satisfaction, and scalability. The team believes in
making data not just accessible but also actionable, helping decision-
makers optimize operations and maximize ROI.

2. Projects Undertaken During the Internship

During my internship at Segment Analytics Pvt. Ltd., I was exposed to
multiple real-world data analytics projects. These projects were
integral to the organization's ongoing operations and client
deliverables. Below is a detailed breakdown of the major projects I
contributed to:
a. Retail Sales Performance Dashboard
Objective:
To analyse sales data from various regional stores and product lines,
with the aim to provide management a clear view of sales trends,
revenue distribution, and underperforming segments.
Tasks & Responsibilities:
 Cleaned and transformed raw Excel files containing ~30,000
transaction records.
 Used Python’s Pandas, NumPy, and Matplotlib libraries for data
manipulation and visualization.
 Generated monthly revenue trends, product-wise sales
breakdown, and regional comparisons.
 Suggested optimization strategies for stock management based
on slow-moving product lines.
Outcome:
Delivered a concise executive summary with visuals and interactive
graphs to the business intelligence team. My insights helped
highlight a 12% drop in sales in specific regions, enabling targeted
interventions.

b. Customer Churn Prediction Model

Objective:
To identify key behavioural patterns among customers who
discontinued their subscription and develop a predictive model to
estimate future churn risks.
Tasks & Responsibilities:
 Queried customer data from SQL databases to extract
behaviour metrics (e.g., login frequency, support tickets,
transaction history).
 Cleaned and labelled churned vs. active customers.
 Built a logistic regression model using scikit-learn in Python.
  Validated the model with accuracy, precision, and ROC curve
metrics.
Outcome:
The model achieved an 83% prediction accuracy. My report also
included actionable insights, such as the critical role of reduced
product usage in the two months before churn, allowing the
company to implement retention strategies.

c. HR Analytics Dashboard in Power BI

Objective:
To build an interactive dashboard for the HR department to track
employee attrition, hiring trends, and department-wise performance
metrics.
Tasks & Responsibilities:
 Combined multiple Excel and CSV files using Power Query
Editor.
 Designed dashboards using Power BI with slicers for
department, location, and experience levels.
 Integrated KPIs such as average tenure, attrition rate, and new
hires per quarter.
Outcome:
Presented the dashboard to the HR team in a virtual demo. The tool
is now used regularly for strategic workforce planning and internal
reporting.
3. Major Challenges Accomplished During the
Internship
The internship came with its own set of challenges, each of which
provided me with new learning opportunities and helped me grow as
a data analyst.
a. Real-World Data Complexity
Academic datasets are often clean and pre-processed, but real-world
data is messy. I faced challenges such as:
 Missing or incorrect entries
 Non-uniform date formats and categories
 Duplicate records
I learned to apply data cleaning techniques and write Python scripts
that automated much of the process, ensuring efficiency and
accuracy.

b. SQL Optimization for Large Databases

One of the projects required pulling large datasets with complex
joins. My initial queries were slow and inefficient. I overcame this by:
 Learning indexing and filtering techniques
 Using subqueries and Common Table Expressions (CTEs)
 Removing unnecessary aggregations and reducing
computational load
This improved query speed by nearly 40% and enhanced overall
dashboard responsiveness.
c. Remote Collaboration and Time Management
Working remotely required me to develop soft skills beyond
technical expertise. I had to:
 Maintain timely and clear communication with my project
mentor and team via Slack and Zoom.
 Deliver updates regularly and follow up on feedback.
 Manage time independently while balancing academic
commitments.
This taught me the importance of self-discipline and proactive
communication in a professional setting.

4. Conclusion
My internship with Segment Analytics Pvt. Ltd. was a transformative
experience. It served as a bridge between theoretical learning and
practical application. I not only improved my proficiency in Python,
SQL, Excel, and Power BI, but also learned to think like a data
analyst—questioning patterns, validating assumptions, and
delivering insights that solve real business problems.
The hands-on exposure to industry-grade datasets, along with
mentorship from seasoned professionals, gave me a realistic
understanding of analytics workflows and stakeholder expectations.
Additionally, managing the internship remotely taught me essential
skills in time management, adaptability, and digital collaboration.
This experience has reinforced my interest in the field of data
analytics and product decision-making. I now feel more confident in
pursuing future roles that involve solving business challenges using
data, and I aim to build on this foundation in upcoming internships or
full-time roles.