Overview
Quail’s Chat interface allows you to explore and analyze your data using natural language. Simply ask questions in plain English, and Quail will automatically generate SQL queries, execute them against your database, and present results in the most appropriate format—whether that’s a table, chart, or summary.Getting Started
Select Connection
Make sure you have at least one database connection set up. The chat will use your currently selected connection, which you can change from the dropdown at the top.
The Chat interface works best when you have a clear understanding of your data structure. If you’re new to your database, explore the Schema Explorer first.
Asking Effective Questions
The quality of your results depends on how you phrase your questions. Here are examples organized by complexity:- Simple Queries
- Time-Based Analysis
- Complex Analysis
- Filtering & Grouping
Start with straightforward questions to explore your data:Counting and Totals:
- “How many customers do we have?”
- “What’s the total revenue this year?”
- “How many orders were placed today?”
- “What’s our average order value?”
- “Show me the most expensive product”
- “What’s the lowest price in our catalog?”
- “Show me sales for the last 30 days”
- “What are today’s orders?”
- “List new customers from this week”
Understanding Chat Responses
Quail provides comprehensive responses that help you understand both the query and the results.Query Explanation
For each question, Quail shows you:Generated SQL
The exact SQL query that was generated from your natural language question.
Plain English Explanation
A clear explanation of what the query does and how it answers your question.
Assumptions
Any assumptions made during query generation (e.g., date ranges, filters, or table joins).
Always review the generated SQL and assumptions to ensure the query matches your intent. You can ask for modifications if needed.
Results Display
Quail automatically chooses the best way to display your results:- Table View
- Chart View
- Summary View
Best for:
- Detailed data exploration
- Viewing individual records
- Exporting data
- Sortable columns
- Pagination for large datasets
- Copy individual cells
- Export to CSV
Refining Your Results
If the initial results aren’t quite what you wanted, you can refine them through conversation:Review Results
Look at the generated query, assumptions, and results to identify what needs adjustment.
Ask for Modifications
Send a follow-up message with your refinement:
- “Can you add profit margin to that?”
- “Show only the top 10 results”
- “Group by month instead of week”
- “Exclude cancelled orders”
Examples of Refinements
Original question:“Show me monthly sales”Follow-up refinements:
- “Make it a line chart instead”
- “Add last year’s data for comparison”
- “Show only the top 5 product categories”
- “Break it down by region”
- “Exclude returns and refunds”
Working with Charts
When Quail generates a chart in response to your question, you have several options:Modifying Charts
Request Changes
Ask Quail to modify the chart:
- “Change this to a bar chart”
- “Make it a stacked area chart”
- “Show percentages instead of absolute values”
- “Change the colors”
Saving Charts
When you create a chart you want to keep:Add Details
Provide:
- Name: Descriptive name for the chart
- Description: Optional details about what the chart shows
- Tags: For organization
Choose Destination
Select where to save:
- Standalone chart: Save to your charts library
- Add to dashboard: Place directly on a dashboard
Exporting Data
Export chat results for use in other applications:Export to CSV
Click Export CSV to download table data as a comma-separated values file.
Copy to Clipboard
Click Copy to copy the current results to your clipboard for pasting into spreadsheets.
Share Results
Click Share to generate a shareable link to the current chat and results.
Download Chart
For chart results, click Download to save as PNG or SVG.
Chat History and Sessions
Managing Conversations
Your chat sessions are automatically saved:- Recent chats: Access from the sidebar
- Search conversations: Find previous questions and results
- Continue sessions: Pick up where you left off
- Delete chats: Remove conversations you no longer need
Starting a New Session
To start fresh:- Click New Chat in the top right
- Previous context is cleared
- You can switch between active sessions from the sidebar
Tips for Better Results
Be Specific
Vague
“Show me sales”
Specific
“Show me daily sales for the past 30 days grouped by product category”
Include Context
Unclear
“What’s the average?”
Clear
“What’s the average order value for premium customers in Q1 2026?”
Use Follow-ups
Less Efficient
Starting a new conversation for each related question
More Efficient
Building on previous questions in the same conversation
Common Patterns
Exploring New Data
- “How many records are in the [table_name] table?”
- “Show me a sample of 10 rows from [table_name]”
- “What columns are in the [table_name] table?”
- “What’s the date range of data in [table_name]?”
Monitoring Metrics
- “What’s [metric] today compared to yesterday?”
- “Show me [metric] trend for the last 7 days”
- “What’s the month-to-date [metric]?”
- “How does this week’s [metric] compare to last week?”
Finding Issues
- “Show me orders with no customer information”
- “Find duplicate entries in [table_name]”
- “List products with no sales in the last 90 days”
- “Show me failed transactions from today”
Limitations and Best Practices
What Works Well
- Analytical queries (aggregations, trends, comparisons)
- Exploring data structure and content
- Creating visualizations from query results
- Iterative refinement of queries
What to Use SQL Editor For
- Complex multi-step queries
- Queries requiring custom functions or procedures
- Fine-tuning query performance
- Queries with very specific, complex joins