Jnana Deepika Boppana
JB
Business & Data Analyst

Jnana Deepika Boppana

Transforming data insights into impactful business strategies and measurable outcomes.

Dallas, TX (Open to Relocation)
Get in touch
2+Experiences
6+Projects
16+Skills
01.

About

I'm Jnana Deepika Boppana, a Business and Data Analyst with expertise in SQL, Power BI, and predictive modeling. I thrive on solving complex business challenges and delivering actionable insights that drive profitability and efficiency.

02.

Experience

Aug 2025 – May 2026

Technology Administrator (Part-Time) · Chartwells Higher Education

Richardson, TX
  • Minimized IT rollout delays across 5 campus locations from 2–3 weeks to on-schedule delivery by gathering functional requirements upfront and authoring SOPs and system design specifications prior to each deployment.
  • Achieved zero end-user-facing defects across 5 locations by conducting structured UAT validation on every system update, identifying issues before they reached go-live.
  • Reduced daily transaction discrepancy rate by 40% by validating over 2,000 records in SQL daily across Oracle MICROS POS systems and providing process-improvement recommendations adopted by operations within 2 weeks.
Oct 2022 – Dec 2022

Data Analyst Intern · Squpus Private Limited

Hyderabad, India
  • Reduced manual resume-screening time by 50% by developing a supervised classification model in Python and scikit-learn that evaluated candidates against 8 weighted hiring criteria for the firm's recruitment workflow.
  • Facilitated model training and validation by engineering a 100,000-row dataset across 8 weighted features using a parameterized data pipeline in Python and pandas, compensating for an unavailable data source.
  • Empowered company leadership to act on findings by translating model performance metrics into a technical report and a 19-slide presentation for a non-technical audience at the conclusion of a 13-week engagement.
03.

Selected work

K

KPI Reconciliation & Metric Governance

PostgreSQL, PowerBI, MS Excel
  • Reconciled three conflicting definitions of 'active customer' across marketing, finance, and product teams by constructing a governed SQL layer in PostgreSQL, resolving a 170-customer reporting discrepancy and establishing a single trusted source per metric.
  • Corrected an inflated customer count by 15% (328→278) by resolving duplicate identities, non-sale orders, and refund-timing errors using multi-table joins, anti-joins, and CTEs across six source tables.
  • Developed a governed SQL validation suite and metric catalog that surfaced a timezone defect exposing $34K in sales to misattribution, ensuring every reported figure was traceable and audit-ready.
PostgreSQLPowerBIMS Excel
M

Marketing Campaign Profitability & Customer Targeting Analysis

SQL (PostgreSQL), PowerBI
  • Transformed a 2,973-unit campaign loss into an 844-unit profit by ranking customers based on pre-campaign propensity in SQL and modeling net profit across every contact depth, validated out-of-sample against held-out data.
  • Stress-tested the recommendation against higher contact costs and thinner margins, confirming it remained profitable as the break-even contact depth tightened from 46% to 18%.
  • Created a Power BI dashboard on a live PostgreSQL connection and identified a channel-mix double-count that overstated the base by 18%, restoring the Store's true share from 39% to 46%.
SQL (PostgreSQL)PowerBI
C

Causal Price Elasticity Modeling & Demand Forecasting – FleetPride

Python (Pandas, Statsmodels)
  • Produced a clean, leakage-free, analysis-ready FleetPride transaction dataset with zero missing values – including correctly flagging 55% of '$0' competitor-price entries as structurally missing rather than true zero prices – by cleaning and engineering features in Python.
  • Developed a regression model that corrected demand-driven pricing bias and isolated true price sensitivity, achieving a 9.0% average forecast error while explaining 98% of price variation and controlling for customer and product differences.
  • Translated technical econometric analysis into pricing recommendations for FleetPride's sponsor, authoring the first full draft of a 20-page technical report and presenting the solution architecture and key findings sections as part of an 8-person analytics practicum team.
Python (PandasStatsmodels)
H

Healthcare Operations: 30-Day Readmission Risk Prediction

R
  • Selected Random Forest over Logistic Regression and Decision Tree as the strongest readmission-risk model despite Logistic Regression's higher raw accuracy (61.2% vs. 60.7%), because Random Forest delivered a better sensitivity/specificity balance (60.1% balanced accuracy, AUC 0.64) – correctly flagging more true at-risk patients instead of defaulting to the majority class.
  • Identified lab-procedure count, medication burden, length of stay, and prior visits as the strongest readmission predictors via variable importance, translating findings into hospital recommendations to prioritize high-utilization patients for discharge follow-up and direct resources to Internal Medicine and Emergency, the two highest-readmission specialties.
R
C

COVID-19’s Impact on U.S. Airfare – Data Visualization

Tableau
  • Constructed an analysis-ready dataset spanning 5 quarters and 10 states by consolidating 2,127 airport-level BTS fare records with daily COVID case data into 4,580 clean records, resolving 15 missing-field rows, 10 missing fares, and 5 duplicates.
  • Quantified a strong inverse relationship between COVID case surges and airfare pricing – an average correlation of r = -0.85 across 10 states (peaking at r = -0.91 in Texas) – and measured a 31% drop in average domestic airfare from the pre-pandemic baseline ($392, Q4 2019) to the pandemic trough ($269, Q3 2020).
  • Designed a 4-panel interactive Tableau dashboard – a dual-axis cases-vs-fares trend, a 10-state quarterly heat map, and state/quarter filters – plus a guided Tableau Story to walk non-technical stakeholders through the pandemic’s pricing impact.
Tableau
R

Ray-Ban High-Value Customer Analysis

Google Analytics (GA4), Looker Studio
  • Identified iPhone users as Ray-Ban’s highest-value customer segment by analyzing revenue-by-device data in GA4, finding iPhone’s share of top-device revenue grew from 56% to 68% YoY as its revenue nearly doubled (+100%) while competing Android models grew at a much slower rate.
  • Pinpointed 5 major U.S. tech hubs – including New York, San Francisco, and Sunnyvale – that consistently drove approximately 27% of total U.S. revenue in both Oct 2024 and Oct 2025, by building a geography-based revenue analysis in GA4 and Looker Studio.
  • Analyzed high-value-customer revenue by channel as part of a 6-person marketing analytics team, discovering that Direct and Organic Search remained the top two sources while Referral (+74%) and Email (+54%) grew the fastest YoY – findings that directly informed 4 of the team’s 5 final targeting recommendations to Ray-Ban.
Google Analytics (GA4)Looker Studio
04.

Skills & tools

SQL (PostgreSQLMySQL)Power BITableauPythonR ProgrammingGoogle AnalyticsSalesforce CRMMicrosoft ExcelProject ManagementKNIMEGenerative AIRequirements GatheringData WranglingData ValidationKPI Definition & Goverance
05.

Education

The University of Texas at Dallas

M.S. Business Analytics & Artificial Intelligence
May 2026

P.B. Siddhartha College of Arts & Science

B.B.A. Business Analytics
May 2023
06.

Certifications

Salesforce Certified Administrator

Salesforce Certified Administrator

Salesforce · 2025

Google Analytics (GA4)

Google

KNIME L1

KNIME

Databricks Generative AI Fundamentals

Databricks

Cisco Data Analytics Essentials

Cisco

Microsoft Power BI Data Analyst (Coursera)

Coursera
07.

Open source

Jnana Deepika Boppana's GitHub contributions
08.

Get in touch

Have a role or a project in mind? I'd love to hear from you.

Gmail (214) 899-1023
Made with TailorCV — build yours free