Олексій

Oleksii Abramov
Junior Data Analyst · [відкрити контакти](див. вище в блоці «контактна інформація») · [відкрити контакти](див. вище в блоці «контактна інформація») · github.com/oleksii-abramov
Kharkiv, Ukraine
Summary
Junior Data Analyst with hands-on experience in SQL, Python, Power BI and Excel. Three end-to-end projects: cohort
analysis, A/B testing, dashboard reporting.
Technical Skills
• Power BI, MS Excel, SQL, Python, Statistics

Projects
Customer Cohort & Repeat Purchase Analysis | SQL, Python, Power BI | GitHub
• Analysed repeat purchases of 16,700+ customers (2020 - 2022 cohorts) to measure the 12-month repeat
purchase rate (RPR) and the time to a second order by cohort and customer value group.
• Wrote SQL (PostgreSQL: CTEs, window functions, percentiles) to build cohorts and value groups; reshaped the
data in Python; built a Power BI dashboard.
• Found that the 12-month RPR rose from 5.44% (2020 cohort) to 21.88% (2022 cohort); repeat orders were
spread evenly over the first year, and about 82% of customers did not return within 12 months.

Mobile Game A/B Test & Retention Analysis | Python | GitHub
• Evaluated an A/B test on 90,000+ players of the mobile game Cookie Cats to measure the effect of moving the
first progression gate from level 30 to level 40.
• Ran Chi-Square tests on retention: Day 7 was significantly lower for the level 40 gate (18.11% vs. 18.95%, p =
0.0014).
• Built a 10,000-iteration bootstrap in Python to get a 95% confidence interval for the difference in rounds played, a
non-normal metric; the difference was not significant.
• Recommended keeping the gate at level 30, and flagged a sample ratio mismatch (p ≈ 0.009) and dilution from
players who never reach the gate.

Road Crash Severity Analysis, Queensland (2001–2024) | Excel | GitHub
• Analysed about 320K crashes with casualties (Queensland Department of Transport and Main Roads) to see how
driver age, gender and licence type relate to crash severity.
• Built non-overlapping driver groups with Excel formulas, checked data consistency, and built a dashboard with
PivotCharts and slicers.
• Found that crashes with an unlicensed driver have a 4.0% fatal share vs. 1.9% for regular licences (about 2x),
while the share ending in hospitalisation rose from 30.5% to 49.1% (2001–2024).

Education: Bachelor’s degree in Transport technology, Kharkiv National Automotive and Highway University,
graduated 2026
Languages:
• English - C2 Proficient on EF SET (73/100, reading and listening, certificate)
• Ukrainian, Russian (native)