Be Useful With SQL

Learning SQL is easy. So is mastering it.

Anyone can do both within a few months.

Mastering SQL involves understanding the query language at a basic, intermediate, and advanced level.

It includes learning concepts like:

• Basic: SELECT statements, WHERE clauses for filtering, GROUP BY for aggregation, ORDER BY, basic joins (like inner and left), basic DDL operations (CREATE, DROP, ALTER, and TRUNCATE), and basic DML operations (INSERT, UPDATE, and DELETE).
• Intermediate: Window functions, advanced joins (like right, full outer, self, and cross), set operations (like union and union all), case statements for conditional logic, CTEs and temp tables for modularizing queries, subqueries for inline logic, and data modeling.
• Advanced: Performance tuning and optimization, creating/calling stored procedures, creating user-defined table and scalar functions, creating triggers, TCL operations (like commits and rollbacks), Dynamic SQL, while loop, if/else statements, indexing, and partitioning.

You can master the basic and intermediate levels within a few weeks and the advanced level within a few months. It’s not as difficult as you’d expect.

The real challenge is learning how to be useful with SQL. That’s more complex.

Being useful with SQL does not mean you have memorized all of its features and syntax. Nor does it mean you can format your queries in a pretty, easy-to-read way. And it doesn’t even mean that you can write complex queries to pull data from a DB.

Instead, it means you can effectively use SQL to bridge the gap between your data team and the business team that derives value from the data.

True usefulness means you:

• Fully understand your database tables and objects, including how they relate to or impact one another.
• Understand the data’s source system (3rd party SAAS, API, imported CSV/Excel files, manual inserts, etc).
• Understand how business users interact with the data and what they do with it.
• Can identify enhanced ways to manipulate or deliver data for a better user experience.

Unfortunately, you cannot learn how to do this by analyzing free datasets online. It’s difficult to be useful with SQL without on-the-job experience.

Your analysis in side projects doesn’t mimic the real-world process. In your side projects, you begin with the dataset, find questions to answer, and then use SQL to solve said questions.

The real world is the opposite.

You DO NOT begin with the dataset. You begin with the question to answer. Then, you search for the existing data or collect it yourself if it doesn’t exist. From there, you use SQL to explore, clean, and analyze the data to answer the question.

The difference between knowing SQL and being useful with it lies in the value your queries/scripts provide. Most side projects and learning datasets don’t provide real-world insights. Yup, that means no one actually cares about your Titanic dataset analysis. True SQL usefulness means you can write queries and build DB objects that provide value downstream to users. I can’t think of any way to effectively learn that with free datasets or basic, beginner projects.

On-the-job training is most efficient.

So, while it’s very unlikely you’ll be useful with SQL without prior experience on production-level solutions, there’s good news.

Many entry-level data roles don’t require you to be useful from day one.

You’ll spend the first few weeks or months learning about the business and experiencing how different users operate with your team’s data solutions. If you have a strong manager and a supportive team, you’ll learn how to be useful quickly.

As you witness more and become involved in more projects, your SQL usefulness will naturally grow.

‍