---
description: How to write effective questions and get the most out of your AI Analyst.
---

# Tips for Asking Questions

This guide will help you get the best results when asking your AI Analyst for insights.

## What Can You Ask?

Your AI Analyst can handle a wide variety of data-related requests:

* **Data exploration**: "What sort of analyses can you do about our shipment data? Give me some ideas."
* **Specific visualisations**: "Create a line chart showing monthly revenue trends for 2025"
* **Lookup information**: "What's the delivery status of order #12345?"
* **Comparative analysis**: "Compare total revenue across all regions for Q1 vs Q2 2025, specifically for our premium products"
* **Comprehensive reports**: "Put together a report on our Q3 carrier performance"

## Crafting Effective Questions

### What Makes a Good Question?

Good questions are specific, include relevant context, and clearly state what you're trying to learn.

#### ✅ Good examples

* "What was our customer retention rate in the Northeast region during Q1 2023 compared to Q1 2022?"
* "Show me the conversion rate for our email campaign last month, broken down by customer segment"
* "Which products had the highest profit margin in our online stores during the holiday season?"

#### ❌ Less effective examples

* "Give me data on customers" — unclear what aspect of customers you're interested in
* "Is our business doing well?" — subjective and lacks specific metrics
* "How did sales go in Benelux last month?" — too broad, lacks specific metrics

### Tips

1. **Specify metrics** — clearly state what you want to measure (revenue, units sold, conversion rate)
2. **Include time frames** — mention the time period you're interested in
3. **Add context** — note any specific segments, categories, or comparisons
4. **State visualisation preferences** — if you want a specific chart type, mention it

## Requesting Charts

Your AI Analyst can generate various chart types:

* **Bar charts** — great for comparing quantities across categories. _Example: regional sales comparison_
* **Pie charts** — useful for showing proportions within a whole. _Example: market share distribution_
* **Line charts** — ideal for visualising trends over time. _Example: monthly revenue growth_
* **Scatter plots** — excellent for identifying relationships between variables. _Example: price vs. rating correlation_

## Follow-up Strategies

Conversations with your AI Analyst are iterative. Try these follow-up approaches:

* **Request different views**: "Can you show this data by week instead of month?"
* **Drill down**: "Show me a breakdown of this spike in June"
* **Clarification**: "How did you calculate this?"
* **Build a report**: "Can you pull all of this together into a report?"

## Troubleshooting

* If you get unexpected results, try rephrasing your question with more specificity
* For complex analyses, break your request into smaller, related questions
* Your AI Analyst works with the data it has access to — some questions may require data your organisation hasn't connected yet

## Using Plan Mode for Complex Questions

For multi-step analyses where you want to stay in control of the approach, enable [Plan Mode](plan-mode.md). The AI Analyst will outline its approach before starting, and you can approve or adjust the plan before it runs.