SQL questions asked during interviews can vary widely across companies, but even more so across positions. You won’t see data scientists asked the same SQL questions as software engineers, and that’s because data scientists have to write different types of queries.

Generally, each SQL interview question can be bucketed into these categories:

• Basic SQL questions
• Reporting and metrics SQL questions
• Analytics SQL questions
• ETL SQL questions
• Database design questions
• Logic-based SQL questions

In this section, we’ll go over what these different kinds of SQL questions are in detail.

Basic SQL Questions

Basic SQL questions are what they sound like. These questions will generally be easy and focus on assessing if you know the basics.

Definition-based SQL questions are grouped into this category because they’re super easy to learn. All you have to do is study a list of definitions of SQL terms and applications. These questions will include understanding the differences between joins, what kinds of aggregations exist, and knowing the basic functions like CASE WHEN or HAVING.

Basic SQL interview questions that involve a user actually writing a query are slightly different. These will involve getting the COUNT of a table, knowing what the HAVING clause does, and figuring out how to utilize LEFT JOIN versus INNER JOIN to give you the values that you need.

Reporting and Metrics SQL Questions

Reporting and metrics SQL questions are probably the most common type of SQL question to show up in interviews.

Reporting SQL interview questions replicate the work that many business and reporting analysts do on a day-to-day basis. This means writing queries that end up in dashboards and key metrics.

For example, a typical B2C software company would likely want to understand how many new users signed up. Better yet, they’d want to understand the daily, weekly, and monthly active users that are on their platform. What about the number of week-over-week new users that signed up as well?

These questions are generally some of the most common problems that analysts and data scientists will run into when pulling metrics for their day-to-day job. The output of what they need to pull is very clearly defined and the queries themselves require complex joins, sub-queries, self-joins, window functions, and more.

Analytics SQL Questions

Analytics SQL interview questions are some of the trickiest interview questions that you will face. This is because they test two concepts:

1. Understanding what metrics we need to answer the question.
2. Writing the correct SQL query that will output these metrics.

Analytics SQL interview questions are designed to test how you would think about solving a problem, and are purposely left more ambiguous than other types of problems. The tough part is that you not only have to think critically about what the SQL output has to look like, you also need to understand EXACTLY the right data points to pull.

For example, an interviewer might ask you to write a SQL query (given a few tables) to understand which AB test variant won. But there might not even be any understanding of what winning actually means.

ETL SQL Questions

ETL stands for “Extract, Transfer, Load” and describes the process for which data flows between different data warehousing systems.

“Extract” does the process of reading data from a database. “Transform” converts data into a format that could be appropriate for reporting, analysis, machine learning, etc., and “Load” writes the transformed data into another database, table, or any other data storage service that can be then used by another data scientist or engineer for reporting.

Many times, ETLs are stacked on top of each other, creating a system of complex data-flows that eventually need to be managed by scheduling systems, such as Airflow or MLflow.

In the interview, ETL concepts are important to know for virtually all roles. The more difficult interview questions, however, will likely be focused and asked in data engineering, business intelligence, and related interviews.

Database Design SQL Questions

Database design SQL questions test your knowledge of data architecture and design. Most importantly, it tests whether you know how to design a database from scratch, given a business idea, application, or any other software that needs to interact with a database.

Many times, when databases need to scale for performance or breadth size of tables, we need to realize how to modify our database to fit the new requirements. Starting from a solid initial design is always important when building out databases.

Logic-Based SQL Questions

Logic-based SQL interview questions are very tricky. They aren’t really based on real life examples so much as putting the trickiness of algorithms and data structure interviews into SQL questions. This is exemplified on sites such as LeetCode, where you’ll see a lot of interview questions that aren’t very practical for real life scenarios.