Data environment

Practice datasets

Six fictional but realistic relational databases. Query them in the browser, download CSV or Excel files, or load them into Power BI, Tableau or Python. All people and companies are invented.

Retail & e-commerce

Orion Retail

A fictional omnichannel retailer with 8 physical stores and an online store across the Caucasus, Turkey and Europe, 2023–2025. Includes seasonality, campaigns, store openings/closures (for like-for-like analysis), a sales plan, inventory and a deliberately messy customer extract.

10 tables · 123,204 rows

Telecommunications

Orion Telecom

A fictional mobile operator: subscriber base, tariff plans, monthly usage and billing, support tickets, marketing campaigns and a full lead funnel (2024–2025). Built for churn, retention, cohort, lead-conversion and campaign analysis.

6 tables · 156,932 rows

Banking & insurance

Orion Bank & Insurance

A fictional retail bank with an insurance arm: customers, accounts, card/current-account transactions (2025), consumer loans with default outcomes, insurance policies with renewal flags, and claims.

6 tables · 191,755 rows

Human resources

Orion People

Fictional HR data for a ~470-person company: employees including leavers (2016–2025), salaries and ratings, monthly sales-rep performance vs plan for bonus/commission work, and a recruitment pipeline.

4 tables · 7,149 rows

Finance

Orion Finance

Monthly management accounts for the fictional Orion Retail group, 2023–2025: P&L by line item with budget, and a month-end balance sheet that balances (Assets = Liabilities + Equity).

2 tables · 864 rows

E-commerce / product analytics

Orion Shop Experiments

Two months of web sessions for a fictional online shop running an A/B test of a new one-page checkout (variant B). Supports funnel analysis, conversion rates, A/B testing and confidence intervals.

2 tables · 46,383 rows