ChatGPT Prompt for Learning SQL: 12 Practice Drills to Run

A good ChatGPT prompt for learning SQL gives you small practice tables you can run yourself, one concept at a time, and exercises with the exact rows a correct query should return. The 12 prompts below go from first SELECT statements to joins, grouping, window functions, and real report tasks. Run every query in your own database tool, because seeing the result is how SQL sticks.

Safety note: Practice on fictional data. Never paste real customer records, passwords, connection strings, or API keys into a chat.

Set up practice data

1. Practice tables for my database

Write a setup script for [SQLite, PostgreSQL, MySQL, or SQL Server] with 3 small tables about [topic], 5 to 8 rows each, with primary and foreign keys so I can practice joins.

2. Messy data version

Add rows to my practice tables [paste script] that include NULLs, duplicate names, and mixed date formats, so I can practice cleaning. List each problem you added.

3. Level check

Ask me 6 short SQL questions from easy to hard on these tables [paste script]. After my answers, tell me which concept I should learn next.

Learn the core concepts

4. One concept lesson

Teach me [WHERE, JOIN, GROUP BY, subqueries] in under 150 words using my practice tables, then give me 3 exercises with the expected rows and hide the solutions.

5. Join types side by side

Using my tables [paste], show what INNER JOIN and LEFT JOIN return for the same question, row by row, and explain why the row counts differ.

6. WHERE versus HAVING

Give me 4 questions on my tables where I must decide between WHERE and HAVING. After I answer, explain each choice.

7. Window functions intro

I know GROUP BY. Teach me ROW_NUMBER and a running total with OVER on my tables in [dialect], and show how the output differs from GROUP BY.

Get feedback on your queries

8. Grade my query

Here is my query [paste] and what I wanted it to return [goal]. Tell me if it runs in [dialect], what it returns on my tables, and give a hint, not the answer.

9. Explain an error

My query [paste] gives this error in [dialect]: [error]. Explain what the database is telling me in plain words and where in the query to look.

10. Make it readable

Reformat this working query [paste] with clear aliases and line breaks, and add a one-line comment above each clause explaining what it does.

Practice real tasks

11. Job-style report

I am preparing for a [job title] role. Write 5 report requests a manager might send, such as monthly totals by region, using my practice tables, with expected results.

12. Interview drill

Act as an interviewer for a [role] SQL screen. Ask 4 timed questions on my tables, one at a time, and score each answer for correctness and clarity.

Learn with graded exercises in one prompt

The flagship Learning SQL Prompt: Practice Tables and Graded Exercises takes your current level, database, goal, a practice topic, session length, and a query you wrote. It returns a setup script, a short lesson, five graded exercises with expected rows, a review of your query, and the next session plan. Its example teaches joins and GROUP BY on a bakery's orders in SQLite.

Tips for learning SQL with AI

Type the queries yourself instead of copying them. Ask for expected rows before you run a query, then compare. Note which database you use, since SQLite, PostgreSQL, MySQL, and SQL Server differ in details like date functions and grouping rules. Never paste real customer data into a chat.

Related prompts

To speed up slow queries, try the interactive SQL query optimizer and teacher or the Postgres EXPLAIN ANALYZE query reviewer. Many analysts learn both languages, so see the ChatGPT prompt for learning Python next.

Find more in the Coding category, look through Browse, or post your own on Share.