Posts

About Me

Image
Hi, I'm Shivangi Agrawal My journey into the world of business and analytics started not in a classroom or a boardroom, but at the dinner table of my family home in Nimach, India. I grew up watching my parents run our family business with grit, grace, and an instinctive sense of numbers. I didn’t know it then, but sitting beside them—listening to conversations about inventory, customers, and cash flow—was my first lesson in business strategy. They didn’t call it “data-driven decision-making,” but that’s exactly what it was. Education: Inspired by resilience and resourcefulness, I pursued a Bachelor’s degree in Information Technology and went on to earn my Master’s in Business Analytics at Northeastern University. Along the way, I blended the lessons of my childhood with the tools of modern analytics—learning to forecast demand using Python, build dashboards in Power BI and Excel, machine learning techniques, and translate raw data into strategic business moves. Familiar with DevOps...
Read more

Text Analytics - Sentiment Classification on Amazon Product Reviews

Image
NLP | Supervised Learning | Customer Feedback Analysis Tools & Techniques: Python, Pandas, Sklearn, NLTK, TF-IDF Vectorizer, Logistic Regression, Confusion Matrix, Classification Report Overview: In the world of e-commerce, reviews carry more weight than ads. Customer reviews are one of the most powerful tools that shape e-commerce decision-making. I wanted to understand how natural language processing could help businesses extract insight from that unstructured feedback. So, I built a sentiment classification model using real Amazon product reviews, turning raw text into structured, decision-ready insight, applying end-to-end text analytics and machine learning in Python Approach: -Explored a dataset of Amazon reviews labelled as positive or negative -Preprocessed text (lowercasing, removing stop words/punctuation, stemming) -Converted text to vectors using TF-IDF to capture important terms -Trained a Logistic Regression model for sentiment classification -Split data ...
Read more

Case Study: Weee! — Delivering Culture, and the Need for Accuracy

Image
Business Case | Supply Chain Analysis | CX Strategy As an international student living in the U.S., shopping for Indian groceries often felt like a hunt. I had access to nearby Indian stores, but either their pricing was too steep, or the affordable ones were too far and inconvenient to reach. That’s when I turned to   Weee! , which offered both competitive pricing and doorstep convenience for authentic Indian groceries. With one scroll, I could find my favourite fruits, vegetables, atta, masalas, and snacks — it felt like shopping from home. But as I became a regular customer, I also began noticing small gaps that, if resolved, could turn a good experience into a great one. This case study was my way of applying a strategic lens to a service I both use and understand deeply SWOT Analysis — Grounded in Experience Strengths Niche focus with deep cultural relevance, carries regional brands, not just generic “Asian” items Strong pricing advantage over competitors for Indian and ...
Read more

Pizza Sales Dashboard, SQL + Power BI

Image
  Data Analysis | Dashboarding | Business Insight This project began with a simple business question: How can we use sales data to improve restaurant decision-making? Approach: I explored a Pizza Hut sales dataset to uncover patterns in order volume, sales by category, time-based performance, and customer behavior. I used SQL to clean the data and answer key queries such as top-selling items, peak order times, and sales by size and category. Once the insights were extracted, I designed an interactive Power BI dashboard to visualize those findings — providing restaurant managers a quick view of KPIs and sales trends in real time Tools & Technologies:   SQL, Power BI, DAX, Excel Outcome: Delivered a clean dashboard with filters and dynamic charts Helped simulate business decisions like which pizza size to promote, when to offer discounts, and which categories to upsell Improved data storytelling by combining raw query logic with visual business insight Project Links: www.l...
Read more

Real Estate Price Prediction

Image
  Regression Model | EDA | ML Tuning Real estate prices often vary due to a mix of unpredictable factors — location, square footage, amenities, etc. I wanted to build a model that could make sense of this complexity and deliver a reliable price estimate. Approach: I began with exploratory data analysis on 7,000+ records to uncover hidden patterns and remove outliers, and created visualizations such as scatter plots and histograms to gain insights into the data distribution. From there, I tested multiple machine learning models using K-Fold Cross Validation to ensure performance wasn’t biased by a single split.   Tools & Technologies: Python, Pandas, Matplotlib, Scikit-learn, K-Fold Cross Validation, GridSearchCV, Linear Regression Outcome: The final model, a tuned linear regression, achieved 81% accuracy, and helped me understand how even simple algorithms, when applied thoughtfully, can yield powerful results.
Read more

The Palace of Illusions

Image
  The Palace of Illusions This is the Mahabharata retold from Draupadi’s perspective. It explores her strength, pride, questions, and the complex role she played in one of the greatest epics ever written. Draupadi’s story made me reflect on the weight of choices, especially when your voice challenges tradition. She wasn’t perfect — and that’s what made her real. Through her journey, I learned that courage and clarity of purpose are key to both personal identity and business dynamics. In business, navigating conflict with clarity and staying grounded during emotional chaos is what defines strong leadership.
Read more

The Forest of Enchantments

Image
  The Forest of Enchantments A retelling of the Ramayana through Sita’s perspective, this novel explores love, pain, power, and the silent strength of a woman who endures more than she’s ever allowed to express Sita’s story taught me that resilience doesn’t always look loud, and that inner strength can be soft yet unbreakable. In business and life, I’ve learned the importance of voice, empathy, and boundaries — especially for women navigating expectations. Sita’s journey reminded me that self-worth isn't given by others — it’s chosen, and strong leadership often grows through silent endurance.
Read more

The Kalki Vishnu Avatar

Image
The Kalki Vishnu Avatar A mythological vision of the future where divine intervention restores balance in a chaotic world, representing the eternal cycle of destruction and rebirth. This book showed me how transformation is rarely soft — it comes through disruption. In business, I’ve learned that adaptability, resilience, and strong values are crucial during uncertain times. Like Kalki, real change-makers lead not with ego but with purpose and timing.
Read more

Skill Set

Image
  Technical Skills Python, R SQL (DDL, DML, DCL, TCL, DQL) Excel (Pivot Tables, VLOOKUP, VBA, Macros) Power BI, Tableau JavaScript, HTML, CSS Analytics & Business Intelligence Data Modeling & EDA Forecasting & A/B Testing Classification & Clustering Statistical Analysis Data Storytelling ML Tools: NumPy, Pandas, Scikit-learn, Matplotlib Finance & Operations Financial Modeling & Budgeting P&L Understanding SWOT & Scenario Analysis Supply Chain Optimisation Inventory & Procurement Planning ABC Analysis, 6S Methodology Tools & Platforms Git, GitHub Docker, Azure  Jupyter Notebook, Anaconda SAP ERP, Jira Microsoft Office Suite Methodologies & Soft Skills Agile & Scrum Lean Six Sigma Strategic Thinking Stakeholder Communication Adaptability & Collaboration Negotiation Contact Page
Read more