📊 Data Analyst • Problem Solver • Storyteller

Turning Raw Data
into Business Decisions

Insight > Instinct

I don't just crunch numbers — I find the "so what?" hidden in your data. From automating reports that save teams 10+ hours a week to building predictive models that cut customer churn by 18%, I translate messy datasets into clear, actionable strategies.

View Projects Download CV
0+ Projects
0+ Years Exp.
0K Rows Analyzed
Sangam Subedi
↗ 24%
Revenue Growth
⬇ 18%
Churn Reduced
⏱ 10h
Saved / Week

Tools & Technologies

My Tech Stack

The tools I use daily to extract insights, build models, and deliver dashboards.

Python

Pandas, NumPy, Scikit-learn, Matplotlib — automating pipelines & building ML models.

SQL

Complex queries, CTEs, window functions, stored procedures — PostgreSQL & MySQL.

Tableau

Interactive dashboards, calculated fields, LOD expressions — published & embedded.

Power BI

DAX measures, Power Query, data modeling — enterprise-grade reports & KPI tracking.

R

ggplot2, dplyr, tidyr, Shiny — statistical analysis & publication-quality visuals.

Excel

Advanced formulas, pivot tables, VBA macros — the Swiss Army knife of data.

Featured Work

Projects Gallery

Each project is designed to deliver measurable business impact — not just pretty charts.

Churn Prediction Model
Python

Customer Churn Prediction

Built a Random Forest model that predicted at-risk customers with 91% accuracy. Reduced monthly churn by 18%, saving an estimated $240K annually.

PythonScikit-learnPandas
GitHub Live Dashboard
Sales Performance Dashboard
Tableau

Sales Performance Dashboard

Designed an interactive Tableau dashboard tracking $12M in quarterly revenue across 8 regions. Identified a $1.2M untapped segment that leadership fast-tracked into Q3 strategy.

TableauSQLExcel
GitHub Live Dashboard
ETL BI Automated Reporting Pipeline
SQL

Automated Reporting Pipeline

Engineered an end-to-end ETL pipeline that replaced 6 manual Excel reports with automated SQL jobs. Saved the analytics team 10+ hours per week and eliminated data entry errors.

SQLPythonAirflow
GitHub Live Dashboard
KPIs HR Analytics Dashboard
Power BI

HR Analytics Dashboard

Built a Power BI dashboard tracking employee attrition, satisfaction, and performance across 1,200 staff. Pinpointed 3 high-risk departments, leading to targeted retention programs.

Power BIDAXSQL
GitHub Live Dashboard
📊 A/B Test Analysis Engine
Python

A/B Test Analysis Engine

Created a reusable Python framework for running and evaluating A/B tests with Bayesian statistics. Accelerated experiment analysis from 3 days to 15 minutes across the product team.

PythonSciPyStreamlit
GitHub Live Dashboard
Supply Chain Optimization
Tableau

Supply Chain Optimization

Visualized logistics data for 50K+ shipments to identify bottlenecks in the delivery pipeline. Recommendations reduced average delivery time by 22% and cut shipping costs by $180K.

TableauPythonSQL
GitHub Live Dashboard

Deep Dive

Case Study

A structured walkthrough of my analytical process — from business question to actionable insight.

Featured Case Study

Reducing E-Commerce Customer Churn by 18%

Jan 2026 4-week project 85,000 records

01 — The Problem

Customer churn spiked 30% quarter-over-quarter

The e-commerce platform saw a sudden rise in customer cancellations. The business team suspected pricing changes, but had no data-backed evidence. Leadership needed to understand why customers were leaving and which customers were most at risk — within 4 weeks.

02 — Data Cleaning

85K records across 4 disparate sources

Merged transaction logs, CRM records, support tickets, and web analytics data. Handled 12% missing values via imputation, removed duplicates, standardized date formats, and engineered 15 new features (e.g., days since last purchase, support ticket frequency, session depth trend).

# Feature engineering example
df['days_since_purchase'] = (pd.Timestamp.now() - df['last_order']).dt.days
df['support_frequency'] = df.groupby('user_id')['ticket_id'].transform('count')
03 — Analysis

Random Forest + SHAP for interpretability

Trained a Random Forest classifier (91% accuracy, 0.88 F1 score) to predict churn probability. Used SHAP values to identify the top 5 drivers: days since last purchase, support ticket frequency, discount dependency, session depth decline, and payment failure rate. Validated results with 5-fold cross-validation.

04 — Insight & Impact

$240K saved annually through targeted retention

Delivered a risk-scored customer segment to the marketing team. High-risk customers (top 20%) received personalized re-engagement campaigns. Results after 3 months:

↓ 18% Churn Rate
$240K Annual Savings
91% Model Accuracy
3.2× ROI on Campaign

Interactive Tableau Dashboard

Replace this with your embedded Tableau/Power BI iframe
Sangam Subedi — Data Analyst

The Human Behind the Data

About Me

I'm a data analyst who believes every dataset has a story — and my job is to find the chapter that changes the business. With 3+ years of experience turning ambiguous questions into clear, data-driven answers, I specialize in customer analytics, operational efficiency, and predictive modeling.

My approach is simple: start with "why does this matter?" before writing a single line of code. This ensures every analysis delivers impact, not just charts.

When I'm not wrangling data, you'll find me playing chess ♟️, experimenting with film photography 📷, or trying to perfect my homemade ramen recipe 🍜. I believe curiosity drives great analysis — whether it's understanding customer behavior or figuring out the perfect broth-to-noodle ratio.

Get In Touch