The art of asking better questions

1. The art of asking better questions

Welcome back! What if your results aren't quite what you need?

2. Why refinement matters

Your first Genie query might not give you exactly what you need, and that's completely normal. Learning to refine prompts—adding filters, specifying timeframes, requesting grouping—turns Genie from a useful tool into a productivity multiplier. Think of it like asking a colleague for help. The more specific you are, the better answer you'll get. This iterative approach is actually how analysts work best.

3. Demo of vague query

Let me show you what I mean. I'll start with a vague question: "Show me my transactions." Watch what happens. Genie returns results, but I'm getting everything—all transactions, all columns, no particular order. This is overwhelming and not actually useful. I have thousands of rows and I'm not sure what I'm looking at.

4. First refinement

Let's refine this. I'll add a time constraint: "Show me transactions from the most recent month." Much better! Now I'm looking at a specific time period. The results are more focused. But I can make this even more useful.

5. Second refinement

Let's add more specificity: "Show me transactions from the most recent month where the amount was over $100, grouped by franchise." Now we're talking! Genie gives me exactly what I need—recent high-value transactions, organized by franchise. I can immediately see which Bakehouse locations are driving the biggest sales.

6. The formula

Notice the pattern here? I started with what I wanted—transactions. Then I added when—the most recent month. Then I specified how—grouped by franchise. And I added a constraint—amount over $100.

7. Sentence builder

This is what we call the Sentence Builder: what plus when plus how, with any relevant constraints. Most beginners are too brief; they might only type "Sales." Use all four parts and you'll get much better results.

8. Another example

Here's another example with our Bakehouse data. I'll refine the question: "Show me revenue by franchise for May 2024, grouped by city, only transactions over $50." Now I have exactly what I need. And here's a pro tip: Genie shows you the SQL it generated. You can recognize table names like `sales_franchises` or city names like Sydney to confirm Genie is looking at the right things. Think of it as a verification step, not a coding step.

9. Iterative approach mindset

Refining prompts is iterative. Start with your basic question, review the results, then add constraints one at a time. You don't need to write the perfect prompt on your first try. Genie remembers the conversation context, so you can build on previous queries naturally.

10. Let's practice!

Now it's time to practice this yourself. In the exercises, you'll work with the Bakehouse dataset and start with broad questions—then refine them with filters, dates, and grouping until you get exactly the insight you need. Let's get started!

Create Your Free Account

By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.