SQL Practice Online: 15 Exercises on a Real Database

The practice database: 3 tables, 20 rows

Every exercise runs against this schema. It is deliberately small enough to verify by eye but has the two ingredients real practice needs: a foreign key to join on (dept_id) and a second join target (projects) for the harder exercises.

departments(dept_id, dept_name)              -- 5 rows
employees(emp_id, name, dept_id, salary, hired)  -- 10 rows
projects(project_id, title, dept_id, budget)     -- 5 rows

-- employees quick view:
--  Engineering: Ada 9500, Grace 10200, Turing 8800
--  Design:      Paula 7600, Massimo 7200
--  Marketing:   Seth 6400, Ogilvy 6900
--  Sales:       Zig 5800, Mary 6100
--  HR:          Patty 5400

Beginner exercises: SELECT, WHERE and ORDER BY

Warm up with single-table queries. Type each one into the playground, press Run, and compare against the expected result before moving on. If a row is missing from your result, the WHERE clause is the first suspect.

SELECT name, salary
FROM employees
WHERE salary > 7000
ORDER BY salary DESC;

Intermediate exercises: aggregates, GROUP BY and HAVING

Aggregates collapse rows into answers. The mental model that makes GROUP BY click: it first splits the table into buckets, then computes the aggregate once per bucket, and HAVING filters the buckets afterward — which is why aggregate conditions go in HAVING, never in WHERE.

SELECT d.dept_name,
       ROUND(AVG(e.salary), 0) AS avg_salary
FROM employees e
JOIN departments d ON d.dept_id = e.dept_id
GROUP BY d.dept_name
ORDER BY avg_salary DESC;

Advanced exercises: multi-table JOINs and subqueries

These mirror interview questions. Exercise 11 is the money query — it joins three tables and aggregates across two of them. If your numbers are off by rows, check whether you are counting employees or projects: the two GROUP BYs answer different questions with the same shape.

SELECT name, salary, dept_id
FROM employees e
WHERE salary > (
  SELECT AVG(salary)
  FROM employees
  WHERE dept_id = e.dept_id
)
ORDER BY salary DESC;

How to run these queries right now

Open the SQL playground and the sample database is already loaded — departments, employees and projects, exactly as described above. Type or paste an exercise query into the editor, press Run, and the result table appears below with row counts. The playground is real SQLite (compiled to WebAssembly), so JOINs, subqueries, HAVING and SQLite-specific functions like ROUND all behave as documented here.

A practical study loop that works: attempt the query before looking at any solution, run it, compare rows to the expected column in the tables above, and only then read the answer. When your result differs, shrink the query — strip the JOIN, then the WHERE — until the intermediate result shows you exactly where the rows went missing.

Questions this guide answers

• sql practice
• sql practice online
• practice sql questions
• sql exercises with solutions
• sql join practice
• sql group by exercises
• sql playground online free
• practice sql online free