Data Analyst

David Zeff

I take business numbers apart to find what’s actually moving them.

Python and SQL for the analysis, Tableau and Plotly for the dashboard, and a deck the room can follow at the end of it. Four analyses below, each taken from raw data to something you can click.

Alon Shvut, Israel · Open to data and BI analyst roles

Email me Download CV
LinkedIn GitHub
David Zeff Profile Picture
rows analysed across four projects
1.4M
of deposits at risk identified
€186M
reporting error found and corrected
27%
live dashboards built and deployed
4

What I do

Computer Science graduate from Sapir College, currently a Financial Analyst at the Gush Etzion Regional Council, reviewing budgets in the tens of millions of shekels. Data analysis and data science certificates from Hebrew University and Bar-Ilan, July 2026.

Analysis

Working out what’s actually driving a number: segmentation, cohort analysis, A/B and hypothesis testing, difference-in-differences.

Python · pandas · NumPy · SQL · PostgreSQL

Dashboards & BI

Interactive dashboards that still get opened after the project ends — deployed, shareable on a link, built to be read at a glance.

Tableau · Plotly · Streamlit · Matplotlib

Reporting & presentation

Findings turned into something a room can follow: KPI design, executive summaries, and decks built for the decision rather than the data.

KPI design · Excel · Google Sheets · Interactive HTML decks

Selected work

  • Two bar charts. Left: churn by geography — Germany 32.4%, France 16.2%, Spain 16.7%, against a 20% average line. Right: churn by number of products held — 28% at one product, 8% at two, 83% at three and 100% at four.

    01 · Retention

    Bank Customer Churn

    Measured churn in euros of deposit instead of customer counts. €186M of a €765M book at risk, concentrated in Germany and in high-balance accounts.

    €186Mdeposits at risk
    20.4%churn rate
    10,000customers

    Python · pandas · Plotly · Segmentation · KPI design

    Case study Live dashboard Code
  • Horizontal bar chart of revenue change by category, January to September 2023 versus the same months of 2022. Art and Crafts grew $1.29M, Toys $286K, Sports and Outdoors $211K, Games $168K, while Electronics fell $311K.

    02 · Margin

    Maven Toys México

    An exact mix-vs-rate bridge put all 3.35 points of margin decline on product mix, with a rate effect of zero. The fix belonged with assortment planning, not pricing.

    829,262transaction lines
    100%of decline was mix
    50stores, 29 cities

    Python · pandas · ABC / Pareto · Mix-vs-rate bridging

    Case study Live dashboard Code
  • Bar chart of revenue by daypart. Morning Peak is the tallest at about $255,000, roughly double Midday and Afternoon at about $125,000 each, with Evening near $105,000 and Early Morning lowest at about $85,000.

    03 · Growth decomposition

    Coffee Shop Sales

    Split revenue into orders × basket size: all the growth was footfall. A grain error in the source data had been understating average order value by 27%.

    +104%order growth
    $5.98true AOV, was $4.69
    149,116transaction lines

    Python · pandas · Plotly · Market-basket analysis

    Case study Live dashboard Code
  • Histogram of tenure at cancellation across all 2,067 cancellations. A single red bar at month 8 reaches about 1,050 cancellations while every other month sits below 60, making the month-8 spike 29 times the neighbouring months.

    04 · Campaign measurement

    Airline Loyalty Program

    Trend-adjusted, a +54% campaign headline was worth +40.7%. And half of all cancellations landed on a single month of membership — a contract cliff costing $2.67M a year.

    +40.7%incremental lift
    51%of churn on month 8
    389,065member-months

    Python · pandas · Difference-in-differences · CLV / CAC · Cohort analysis

    Case study Live dashboard Code

All four use public datasets from the Hebrew University Data Analysis programme. No client data appears anywhere on this site.

Skills

Languages
Python · SQL
Databases
PostgreSQL · MySQL
Analysis & methods
pandas · NumPy · A/B testing · hypothesis testing · regression analysis · cohort analysis
BI & visualisation
Tableau · Plotly · Streamlit · Matplotlib · Seaborn
Data handling
ETL · Excel · Google Sheets · Jupyter
Engineering
TypeScript · React · Node.js · Docker · Git
Spoken
Hebrew — native · English — native

Experience

Financial Analyst · Gush Etzion Regional Council

— Present

  • Review and control budgets in the tens of millions of shekels.
  • Build reports and presentations for council departments most days, covering real-time budget control, welfare and youth services. Some are interactive HTML decks.
  • Analyse welfare case-management data to map caseloads and resource allocation for department reviews.

Teaching Assistant · Perach Program

—

  • Taught JavaScript, React and Node.js.
  • Mentored students building full projects with REST APIs and a database behind them.

Customer Service Representative · Moshe Bernstein

—

  • Looked at customer data to spot recurring problems and improve service.
  • Handled invoicing, billing queries and client issues.

Education & certifications

B.Sc. Computer Science · Sapir Academic College

2021 — 2025

GPA 90

Data Analysis Certificate · Hebrew University of Jerusalem

2026

Completed July 2026. The four projects above came out of this programme.

Data Science Certificate · Bar-Ilan University

2026

Completed July 2026.

Contact

Email is the fastest way to reach me. I’m open to data and BI analyst roles, in Israel or remote.

Email dpzeff@gmail.com
Phone +972 54-523-5567
LinkedIn linkedin.com/in/david-zeff
GitHub github.com/DavidZeff1

CV

Download CV