Skip to main content

Getting More From Your Questions

Asking a question in Teela is just the beginning. Once you have your first result, there are several powerful ways to dig deeper, explore your data, and uncover insights you might not have thought to look for. This page walks you through all of them.

Follow-Up Questions​

One of the most natural ways to explore your data in Teela is through follow-up questions. After Teela answers your first question, you can keep the conversation going because Teela remembers the context.

Here's an example of a follow-up chain:

  1. "Show me all orders from last month": Teela returns a list of orders
  2. "Now filter that to California": Teela narrows the results to California orders only
  3. "What's the total revenue from those?": Teela adds up the revenue for that filtered set
  4. "Break it down by product category": Teela groups the revenue by product

Each question builds on the one before it, so you don't have to repeat yourself. It's like having a conversation with someone who remembers what you just talked about.

Tips for Follow-Ups​

  • You can refer to "those," "that," or "the results" and Teela knows what you mean
  • Add filters, change groupings, or ask for summaries of previous results
  • If you want to start fresh, just ask an unrelated question or start a new chat

Drill-Down and Explore​

When Teela shows you aggregated results (like totals, counts, or averages), you can click on any value to see the individual records behind it. This is called drilling down or exploring.

For example, if Teela shows you "Orders by Region" and the West region has 847 orders, you can click on that number to see all 847 individual orders. It's a fast way to go from a high-level summary to the specific details.

Show Details on Single Numbers​

When Teela gives you a single-number answer (say, "Total revenue last month was $1.2M"), you can click Show Details to see the records that make up that total. This is great for sanity-checking or understanding what's included in the number.

Sometimes your data uses IDs or codes that aren't very human-readable on their own. Teela is smart about enriching these with more useful information. For example, if your results include a customer_id, Teela may also show you the customer's name alongside it.

When you see these enriched columns, you can explore the related records. Click on a customer name, for instance, and see their full order history or account details.

Investigations: Ask "Why" Questions​

This is where things get really interesting. Instead of just asking what the data says, you can ask why something happened.

Try questions like:

  • "Why did revenue drop in October?"
  • "What's causing the increase in support tickets?"
  • "Why is the West region outperforming the East?"

When you ask a "why" question, Teela launches an investigation. This means it automatically runs multiple queries, looks at the data from different angles, and then puts together a narrative that explains what it found.

An investigation might reveal, for example, that October revenue dropped because a major customer churned, a seasonal trend kicked in, and a pricing change took effect. You get all of that in one answer, with the supporting data to back it up.

Exporting Investigations​

When Teela completes an investigation, you can export the full findings in two formats:

  • Excel workbook: Includes a summary sheet plus supporting data tabs for each query Teela ran during the investigation. Great for sharing with your team or doing further analysis.
  • PDF report: A polished, readable report that presents the narrative and key findings. Perfect for sharing with stakeholders who just want the highlights.

Data Reasoning: "This Looks Wrong"​

Sometimes you'll look at a result and feel like something is off. Maybe the number seems too high, too low, or just unexpected. Instead of shrugging and moving on, tell Teela about it.

You can say things like:

  • "That doesn't look right"
  • "This seems too high"
  • "I expected this to be higher. Can you check?"

When you do this, Teela kicks into data reasoning mode. It will investigate the result, check for common issues, and explain what might be going on. Maybe a recent data load brought in duplicate records, or maybe a filter is excluding something you expected to be included.

This is one of the most powerful features in Teela. Treat it like a second pair of eyes on your data.

"Help Me Get an Answer"​

Sometimes your question just doesn't produce the result you were hoping for. When that happens, look for the "Help Me Get an Answer" button. It generates alternative ways to phrase your question, giving you several options to try.

This is especially helpful when:

  • You're not sure of the exact terminology your data uses
  • Your first attempt returned an error or unexpected result
  • You want to explore different angles on the same topic

Just click the button, review the suggestions, and pick the one that best matches what you're looking for.

Putting It All Together​

The most effective Teela users combine these techniques naturally. A typical session might look like:

  1. Start with a question: "What were our top 10 customers by revenue last quarter?"
  2. Drill down on one customer to see their order details
  3. Follow up with: "How does that compare to the previous quarter?"
  4. Notice something unexpected and say: "That seems like a big jump. Why?"
  5. Export the investigation as a PDF to share with your manager

The more you explore, the more insights you'll uncover.

Keep Learning​