Hi, I'm Ayush 👋 an entry-level Data Analyst who enjoys turning raw data into clear business recommendations. This page is a summary of my projects, organized by the tool I used.
I work across the full analysis workflow: cleaning and exploring data in Python, answering business questions with SQL, and building dashboards in Power BI, Tableau and Excel. I'm open to data analyst roles in any industry.
Toolkit: Python (pandas, matplotlib, seaborn) · SQL (PostgreSQL, MySQL) · Power BI · Tableau · Excel
| Project Link | Tools | Project Description |
|---|---|---|
| 🛍️ Customer Shopping Behaviour Analysis | Python, PostgreSQL, Power BI | Analyzed the shopping behaviour of 3,900 customers from start to finish: cleaned and prepared the data in Python, loaded it into PostgreSQL to answer key business questions with SQL, built an interactive Power BI dashboard, and turned the findings into 7 business recommendations in a written report and slide deck. |
| Project Link | Area of Analysis | Project Description |
|---|---|---|
| 🚗 Sales & Revenue Analysis | Sales performance, revenue trends | Analyzed 2003–2005 order data for a scale-model vehicle company in MySQL, answering 7 business questions about revenue and sales performance using aggregations, CTEs and window functions. |
| 🎯 Customer Segmentation (RFM) | Customer segmentation, RFM analysis | Segmented 50 customers across 247 transactions (2023–2024) in MySQL: scored Recency, Frequency and Monetary value using CTEs and NTILE, grouped customers into segments, and compared those segments by demographics and product category. |
| Project Link | Area | Project Description | Libraries |
|---|---|---|---|
| 🛒 Black Friday Sales Analysis | Retail customer analysis | Analyzed 537,000+ Black Friday transactions to see how gender, age, marital status, occupation and city tier shape spending. Found that 26–35 year-olds are by far the most active buyers, and unmarried men are the largest customer segment across nearly every cut. | pandas, matplotlib, seaborn |
| 🌍 GDP Analysis | Economic analysis, interactive visualization | Derived year-over-year GDP growth for 256 countries (1960–2016) from World Bank data, automated interactive chart generation for every country, and built a reusable function to compare growth across any set of countries. | pandas, Plotly |
| ❤️ Heart Disease Visual Analysis | Healthcare data visualization | Matched six chart types (distribution, pie, violin, heatmap, joint and pair plots) to the questions each answers best across 918 patient records. Found that ST depression and max heart rate had the strongest links to heart disease. | pandas, NumPy, matplotlib, seaborn |
| 🌾 Sugarcane Production EDA | Agricultural data EDA | Cleaned and explored production data for 102 countries, comparing output, land use and yield by country and continent. Found that Brazil produces about 41% of the world's sugarcane, and that land under cultivation, not yield, drives total output. | pandas, matplotlib, seaborn |
| Project Link | Area of Analysis | Project Description | Excel Features |
|---|---|---|---|
| 🥐 Bakery Sales Analysis | Sales, profitability, forecasting | Analyzed 700 bakery orders over 16 months to evaluate product performance, profit margins and seasonal demand, then forecast Q1 2021 revenue. Found that Chocolate Chip was the only product leading on both revenue and margin, and that two of five customers drove 57% of revenue. | SUMIFS, COUNTIFS, FORECAST, Pivot Tables, Charts |
| 📈 Sales Executive Performance Dashboard | Sales team performance | Built an interactive dashboard tracking 141 sales executives across 8 regions against target, with one region slicer filtering four linked pivot tables and charts at once. Found that the average executive reached only 55% of target. | Pivot Tables, Pivot Charts, Slicers |
| Project Link | Project Description | Dashboard Link |
|---|---|---|
| 🏬 Superstore Sales Analysis Dashboard | Built an interactive dashboard on 9,994 Superstore orders (2014–2017) showing how sales, profit and discounting vary by state, category and time. It combines a sales map, a sales-vs-profit trend, discount and profit distributions, and category breakdowns, with Year, State and Category filters. Found that most sales come from little or no discount (0–20%). | Dashboard |
I'm actively looking for data analyst roles and always happy to talk data. Feel free to reach out on LinkedIn 🤝