OLHA · ANALYST
Data & Revenue Analyst

I find where the
business makes — and
loses — money.

SQL, Python and Tableau are the tools. The actual job is turning a wall of numbers into a decision someone can act on — where growth is coming from, where users drop off, which segment is worth chasing.

✉ Olha.G.Shaptala@gmail.com LinkedIn ↗ GitHub ↗ Tableau Public ↗ Download CV ↓
SCROLL

Revenue & Product Data Analyst with 10+ years of professional experience spanning Sales, Ad Operations, Marketing Operations and Analytics — from managing enterprise client accounts across 6 countries to automating revenue reporting and shipping dashboards people actually use. I combine SQL (BigQuery), Python and Tableau/BI with applied machine learning to solve business problems and support product and revenue decisions, and I'm currently deepening my data science skills at Turing College.

01 — Core Skills

What I bring to the table

Technical depth, backed by the habits that make it usable.

SQL (BigQuery)Complex queries, joins, aggregations
PythonPandas, NumPy, scikit-learn
TableauDashboard development for business KPIs
Funnel & Retention AnalysisGA4, GTM, BigQuery
Revenue & Performance AnalysisReporting automation, KPI tracking
A/B Testing & Marketing AnalyticsExperiment design, significance testing
Machine LearningClustering, forecasting (ARIMA, KMeans, Prophet)
Looker StudioDashboard development
Soft skills
/ 01

Analytical thinking

Working with complex, messy datasets — finding patterns and turning them into decisions, not just charts.

/ 02

Attention to detail

Cleaning messy data, verifying accuracy, catching the thing that would've made the dashboard wrong.

/ 03

Cross-functional collaboration

Worked across marketers, developers and finance — translating between "what the data shows" and "what the team needs."

/ 04

Adaptability

Managing multiple projects and tight deadlines, comfortable picking up new tools when the question demands it.

02 — Approach

What I actually do

Every project starts from a business question, not a dataset.

/ 01

Revenue decomposition

Breaking growth into users × frequency × AOV to find which lever actually moved the number — and which channel or cohort drove it.

/ 02

Retention & cohorts

Tracking cohorts over time to find exactly where users drop off, and whether it's a product problem or an acquisition-quality problem.

/ 03

Segmentation

RFM and clustering to split customers into groups that actually behave differently — so marketing spend stops treating everyone the same.

/ 04

Forecasting

Time series and ML models for demand and sales prediction, built to be explainable to non-technical stakeholders, not just accurate.

03 — Case Studies

Selected work

Full code and notebooks on GitHub.

Revenue Analytics

E-commerce Growth Decomposition

PythonSQLPandas
Problem
Revenue was growing, but nobody could say whether it was coming from more customers, more frequent purchases, or bigger baskets.
Approach
Decomposed revenue into Users × Frequency × AOV across time periods and customer cohorts to isolate the real growth driver.
Result
A repeatable framework that turns "revenue is up" into "here's exactly why, and here's the lever to pull next."
GitHub →
Segmentation + Forecasting

Customer Segmentation & Revenue Forecasting

PythonK-MeansProphetTableau
Problem
A UK online retailer had transaction data but no way to tell which customers mattered most, or where sales were headed next quarter.
Approach
Built RFM-based customer segments ("Champions", "At-Risk", etc.) with K-Means clustering, then forecast future sales with the Prophet model — packaged into a Tableau dashboard and a stakeholder-ready presentation.
Result
A segmentation + forecasting pair that tells the business who to retain now and how much revenue to expect next.
Multi-Stakeholder BI

AI & Workforce Impact Dashboards

PythonStreamlitPlotly
Problem
One dataset on AI adoption in the workforce — but a data analyst, a marketer, and an HR lead all need a different answer from it.
Approach
Built one interactive Streamlit app with 5 linked dashboards, each answering a different stakeholder's core question — from "can we trust the data?" to "where is workforce risk highest?"
Result
A single source of truth that serves five different roles without five different reports.
GitHub →
Automation

CRM Data Cleaning Automation

PythonStreamlitPandas
Problem
CRM exports routinely came in with missing contact fields and duplicate records, and cleaning them by hand ate hours before any analysis could start.
Approach
Built a Streamlit tool that ingests a raw CRM CSV, drops incomplete key records, deduplicates, and outputs a cleaning report plus a ready-to-use file — no manual spreadsheet work.
Result
Same automation instinct behind the 60% faster reporting cycle at Motorsport, turned into a reusable self-serve tool.
GitHub →
Segmentation + LLM

AI Marketing Agent

PythonGPT / LLMRFM
Problem
Customer segments existed on paper, but nobody had time to translate them into actual marketing actions per group.
Approach
Built RFM segmentation, then connected segments to an LLM pipeline that drafts targeted marketing actions automatically per group.
Result
A working bridge between "who are our customers" and "what should we actually send them" — without a human doing it manually every time.
Experimentation

A/B Testing: Product & Retention

PythonStatistics
Problem
Product and marketing teams needed to know if a change actually moved retention and conversion — not just whether numbers looked different.
Approach
Designed and evaluated A/B tests with proper statistical significance testing, including the classic Cookie Cats retention case.
Result
Clear, defensible answers on what to ship — separating real effects from noise.
GitHub →
04 — Commercial Experience

Where I've worked

Data, marketing and revenue operations roles.

Present

Data Analyst
Independent Project · Part-time

GA4GTMBigQueryTableau
  • Set up web analytics tracking from scratch using GA4, GTM and BigQuery
  • Built Tableau dashboards to monitor traffic, conversions and marketing KPIs
  • Analyzed user behavior and website performance to surface actionable patterns
  • Automated recurring performance reporting
2024 – 2025

Junior Data Analyst
Women Go Tech · Remote

TableauBigQuerySQL
  • Built automated Tableau dashboards to monitor marketing and business KPIs
  • Queried and transformed datasets in BigQuery to analyze user conversion funnels
  • Identified behavioral patterns in website event data and proposed optimization actions
  • Supported marketing campaign performance analysis using GA data
2020 – 2024

Ad Operations Analyst
Motorsport, Dnipro · Media

ExcelPythonNetSuite
  • Automated revenue allocation reporting with Excel & Python — cut period-end closing time by 60%
  • Built financial dashboards and revenue tracking systems in NetSuite
  • Maintained sales planning and budget forecasting data
  • Contributed to a successful financial audit through structured data management
2013 – 2018

Account Manager, Sales
Luxoft

Client RelationsSalesCross-border Accounts
  • Managed client relationships and accounts across 6 countries
  • Acted as the main point of contact between clients and delivery teams on IT services engagements
  • Supported contract renewals and account growth within an enterprise sales cycle
05 — Credentials

Certificates & education

Formal background behind the practical skills.

Certifications

  • Google Analytics Certification, 2025
  • Google Cloud Analytics Certificate, 2025
  • Complete Data Science Bootcamp, 2025
  • Google Digital Marketing & E-Commerce
  • DataCamp: SQL & Python Data Analyst

Education

  • Turing College — Data Science & AI
    09/2025 – 11/2026, in progress
  • Oles Honchar Dnipro National University
  • Master's Degree
  • Cybersecurity & Computer-Integrated Tech

Languages

  • English — Upper-Intermediate (EFSET 68)
  • Ukrainian — Native
  • Russian — Native
Open to opportunities

Let's talk about
your data.