What You Should Know Now — and What's Next

What You Should Know Now

• How relational databases organize data into tables, rows, and columns
• Querying with SELECT, filtering with WHERE, sorting, and limiting
• Handling NULL correctly, and using built-in text, number, and date functions
• Aggregating with GROUP BY and HAVING
• Combining tables with inner, outer, and self joins
• Subqueries, common table expressions, CASE, and set operations
• Creating tables, choosing keys and constraints, and designing normalized schemas
• Inserting, updating, and deleting data safely, inside transactions
• Window functions and recursive queries for advanced analysis
• Indexes, query plans, isolation levels, and locking
• JSON in SQL, views, stored procedures and triggers, security, migrations, and production operations

What Should You Learn Next?

• One database in depth — PostgreSQL is an excellent choice: its documentation is outstanding, and it supports almost every feature covered here. Learn its specific tools, extensions, and administration.
• SQL from a programming language — Python (with pandas and SQLAlchemy), R (with dbplyr), or your back-end language's database library and ORM.
• Data engineering — data warehouses (BigQuery, Snowflake, Redshift), tools such as dbt that build analytics models with SQL, and orchestration.
• Analytics — statistics fundamentals and a visualization tool, so your queries turn into decisions.
• Other data models — document databases (MongoDB), key-value stores (Redis), and graph databases, to know when a relational database isn't the best fit (it usually is).

Build Something

Find a dataset about something you care about — your city's open data, a sports league, your own spending — load it into PostgreSQL or SQLite, and answer five questions you're genuinely curious about. Write the queries up with their results; SQL skills are easiest to prove with real analysis.