Challenge - AI and SQL

Now it's time to put theory into practice by leveraging ChatGPT to accelerate your SQL query development. As a Data Analyst, you should approach this with the mindset of uncovering actionable insights from your organization's data. The key is asking the right questions that drive business decisions.

Consider these essential analytical scenarios: First, revenue distribution analysis—in 2021, which geographic regions generated the highest revenue? How does performance vary across states, and what factors might explain these variations? Second, growth trajectory assessment—analyze year-over-year revenue changes by comparing 2020 to 2021 data. Are you seeing consistent growth, seasonal patterns, or concerning declines? Your challenge is to construct these queries using ChatGPT as your AI coding partner, transforming business questions into precise SQL statements without writing the code from scratch.

Beyond these foundational examples, consider exploring customer segmentation analysis, product performance metrics, seasonal trends, or operational efficiency indicators. The goal is developing your ability to translate business requirements into data queries while leveraging AI to handle the technical implementation details.

To maximize your success with ChatGPT, follow these critical best practices that will save you significant time and frustration. Rather than spending valuable time explaining your database schema in text, capture and upload a screenshot of your database structure. This visual context allows ChatGPT to understand your table relationships, column names, and data types instantly, producing queries that execute correctly on the first attempt.

Additionally, always specify your SQL dialect upfront. Whether you're working with PostgreSQL, SQL Server, MySQL, or another variant, each has unique syntax quirks and function libraries. This specification prevents ChatGPT from defaulting to generic SQL that may not work in your specific environment.

Quality assurance remains your responsibility, regardless of AI assistance. After receiving generated code, execute comprehensive testing: Run the query and verify it executes without errors. Examine the results critically—do the numbers align with your business knowledge? Review the code logic to ensure it's solving the intended problem. This verification process builds your SQL comprehension while catching potential AI misinterpretations.

Here's a crucial workflow tip: maintain separate ChatGPT conversations for distinct analytical tasks. Extended conversations can cause the AI to conflate different requirements or assume you're iterating on previous queries when you've moved to entirely new problems. Fresh conversations ensure clean context and more accurate results.

Let me walk you through efficient methods for capturing database schemas across different platforms. In DBeaver, double-click your target tables and navigate to the diagram view to access your entity relationship diagram. Clean up the visualization by removing irrelevant tables or views—simply select unwanted elements and right-click to delete them. Once you've isolated the relevant schema components, right-click the diagram and save it as a PNG image for upload to ChatGPT.

For Mac users seeking a streamlined approach, leverage the built-in screenshot functionality. After opening your database diagram in DBeaver, use Command+Shift+4 to activate the selection tool. Drag to select your desired area, then hold Control before releasing to copy directly to your clipboard rather than saving a file. Navigate to ChatGPT and paste with Command+V for immediate upload.

SQL Server Management Studio users can create focused database diagrams by right-clicking on Database Diagrams and selecting "New Database Diagram." Choose only the tables relevant to your current analysis to avoid overwhelming ChatGPT with unnecessary schema information. While you may not have save permissions for the diagram, you can still copy it to your clipboard by right-clicking and selecting copy. If direct pasting into ChatGPT fails, use Microsoft Paint as an intermediary—paste the diagram, then copy and paste it into ChatGPT.

Now it's time to tackle this challenge and discover how AI can transform your SQL development workflow. In our next session, we'll review effective approaches and discuss advanced techniques for maximizing ChatGPT's analytical capabilities.