The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
End-to-end retail analytics pipeline that turns raw, messy CSV exports into business decisions, built in Python and SQL and runnable with one command.
What it does:
Cleans and validates 6 relational tables, fixes 585 data-quality issues and logs every action
Loads a SQLite warehouse with keys, constraints and window-function SQL
Customer analytics: RFM segmentation, cohort retention, CAC / LTV by channel, product Pareto, market-basket analysis
Churn prediction model validated on future data (ROC-AUC 0.71, no data leakage)
6-month revenue forecast, best of 3 models by backtest (4.2% MAPE)
Auto-generated Excel report (live formulas, native charts) and an interactive Streamlit dashboard
18 automated tests
I've just put a new case study live.
Powermatic Technologies helps growing companies with strategy, design and software. I did the whole project on my own: the logo and identity, the design in Figma, the code and the copy on the site.
The site sits on a four-column grid, and you can see its lines on the screen. Graphite palette with no pure black or white, PP Neue Montreal, and a bit of ASCII art.
The page I'm happiest with is Careers. My first idea was to repeat the home page layout. I dropped it. Now the page opens with the open roles set in 72 px type, with a filter by team. A candidate sees the role, the city and the pay before they scroll.
The menu turns like a drum. You can drag it or use the arrow keys, and it opens on the section you're reading.
It's all hand-written HTML, CSS and JS, no framework. It holds up from a phone to an ultrawide monitor.
The live site is open, so you can click through it yourself.
Nice!
Built a complete SQL Server (T-SQL) data warehouse for a fictional multi-channel, multi-currency retailer (synthetic dataset created for demonstration purposes). Delivered: staging layer loaded from 17 CSV files (~397K rows); one transactional, re-runnable ETL stored procedure that cleans duplicates, mixed date formats and inconsistent labels; a star schema with keys and constraints; reporting views with business rules defined once (net revenue in USD, contribution margin); automated data-quality tests; and 17 analytical queries (cohort retention, RFM segmentation, Pareto/ABC, constant-currency growth, carrier failure alert, inventory health, CAC vs platform claims, budget vs actual). Key insight: marketplace shows 51% gross margin but only 24.5% contribution margin after commission, shipping and refunds.
For the mixed-date cleanup, I'd include a case like 03/04/2026 in the quality tests: it parses under two conventions but can land in different reporting periods. An invalid-date check alone won't catch that. A source-specific date rule, with ambiguous rows held for review and...