Data Analyst  ·  Visual Storyteller  ·  I/O Psychology

Turning raw data into stories that move people

I bring a rare lens to analytics — one shaped by research psychology, a trained artist's eye for composition, and the conviction that every dataset deserves to be understood, not just processed.

Scroll

How I Work

From raw inputs to actionable insight — every project follows a deliberate arc.

01

Collect & Understand

With a foundation in I/O Psychology and social research methodology, I approach data collection with both rigor and curiosity. I understand survey design, behavioral measurement, and psychometric validity — and know how to source, audit, and prepare data that reflects the phenomena it claims to measure.

SQLSurvey DesignResearch MethodsData AuditingExcel
02

Transform & Analyze

I unify disparate sources, clean inconsistencies, and build analytical workflows that surface real patterns. From spatial analysis and predictive modeling to statistical testing and segmentation, I apply the right technique to the right question — always looking for the human signal, not just the statistical one.

PythonRSQLPredictive ModelingSpatial AnalysisStatistical Testing
03

Visualize & Communicate

I build dashboards and visual reports that guide decisions. With deep proficiency in Power BI and experience in Tableau, I design interactive tools that respect cognitive load. My background in digital art gives me an edge: I understand color theory, visual hierarchy, and how to make someone feel what the data says.

Power BITableauData StorytellingInfographicsDashboard Design
Data → Insight → Impact

Projects

A growing collection of analytical and visual work. Each project reflects a commitment to both rigor and craft.

Views
239K
Cart
10K
Buy
5.7K
2.4% overall conversion

Funnel & Predictive Analytics

E-Commerce Funnel & Conversion Analysis

Analyzed 525K+ transactions and user session behavior to map where shoppers drop off — then built a logistic regression model to predict which sessions convert. Caught a sequential funnel flaw in the data, corrected it, and identified add-to-cart as the strongest conversion signal.

95% browse→cart drop-off2.4% conversion rate0.76 AUC model
PythonPandasScikit-learnSQLPower BI
danceability → popularity →
r = 0.04
all |r| < 0.06

Exploratory Data Analysis

Do Audio Features Drive Song Popularity?

Explored whether energy, danceability, loudness, and valence predict track popularity across 114K Spotify songs. Combined SQL aggregation with correlation analysis to uncover a surprising answer: audio features alone explain almost none of the variance in commercial success.

114K tracks analyzedAll |r| < 0.06SQL + Python hybrid
PythonSQL (SQLite)PandasSeabornMatplotlib
97.9%
Approval
0.30%
Fraud
56.7%
Representment

Business Intelligence Dashboard

Fraud Monitoring & Chargeback Analytics

Built a 3-page Power BI dashboard tracking $18.6K in transaction losses across 8 merchant categories. Designed star schema data model with DAX measures for fraud rate, alert rate, and chargeback representment analytics.

97.9% approval rate0.30% fraud rate56.7% representment
Power BIDAXData ModelingStar Schema
Coming soon

Dashboard Design

Utilities Operations Dashboard

An operations analytics dashboard tracking work order completion, crew performance, cost variance, and SLA compliance across regions for a utilities firm.

Power BIDAXData Modeling
In development

Education

B.S. I/O Psychology & Statistics — UIUC

The analyst who paints.
The artist who quantifies.

I'm Medha — a recent graduate from the University of Illinois at Urbana-Champaign with a dual focus in Industrial-Organizational Psychology and Statistics. I'm based in Atlanta, and I'm building a career at the intersection of human behavior, data, and design.

Before I ever opened a dataset, I was opening paint tubes. I've spent years working in watercolors, acrylics, and digital media — training my eye to see composition, contrast, and the quiet structure beneath beautiful things. That artistic discipline didn't disappear when I turned to analytics; it became my edge. I don't just analyze data — I compose it.

What drives me is the belief that data, at its best, is an act of empathy. Behind every metric is a person, a team, or an organization trying to make a better decision. I studied I/O Psychology because I care about how people work. I studied Statistics because I believe good intentions deserve good evidence.

I'm looking for a role where I can bring both halves of my brain to the table — where analytical rigor and creative clarity aren't competing priorities, but the same one.

PythonRSQLPower BITableauExcelResearch DesignData StorytellingDigital ArtI/O Psychology

Have a project in mind?
I'd love to hear about it.

Whether you're looking for an analyst who thinks like a designer, or you just want to talk about the intersection of art and data — my inbox is open.

medhakaranam@gmail.com