Welcome to Your AI Data Analyst
The Zoë chat box is your direct portal to interacting with the most advanced analytics agent available. Think of it as having a conversation with your smartest data analyst - one who never sleeps, reducing time-to-action and questions/answers that live outside of dashboards.
This guide will walk you through everything you need to know to master the chat box and unlock the benefits of self-service analytics.
What You'll Learn
Start with simple metric exploration
Build complex multi-dimensional analyses
Use voice transcription for hands-free queries
Upload external files for enhanced analysis
Select different AI models for varied approaches
Quick Start: Your First Question
Starting Simple
The best way to begin is by understanding what data Zoë can access. Try this first question:
Example:
"What kind of metrics are available?"
When you ask this, Zoë will:
Scan your entire data model
List all existing columns and data
Show custom metrics created by your team
Present everything in an easy-to-understand format
Pro Tip: This is always a great first question in any new analysis session to understand your available data.
Building Complex Analyses
Once you're comfortable with basic questions, you can build sophisticated analyses using natural language. Zoë excels at multi-dimensional queries that would typically require complex SQL.
Example Complex Query:
"Chart total revenue for the past 90 days compared to cost of goods sold. Split out by product type and identify which states have the largest or smallest growth compared to the previous period."
What happens behind the scenes:
Zoë scans your data model for relevant metrics
Identifies or creates needed calculations on-the-fly
Applies time comparisons automatically
Generates visualizations with proper breakdowns
Highlights key insights about growth patterns
This demonstrates why Zoë represents the most effective use of AI in enterprise analytics - complex questions get answered instantly without technical knowledge.
Voice Transcription: Think Out Loud
Sometimes the best insights come from thinking out loud. Use the voice feature to ask questions naturally:
Click the microphone icon in the chat box
Speak your question clearly
Review the transcribed text
Click the up arrow to submit
Voice Question Example:
"Based on the above data, which products would you recommend we improve our supply chain inventory for going forward?"
This feature makes Zoë a true AI agent for data analysis - you're having a conversation, not operating software.
File Uploads: Bring Your Own Data
Need to analyze external data or reference documents? Upload files directly into your chat:
Supported File Types:
CSV files - Sales data, customer lists, inventory sheets
PDF documents - Reports, invoices, documentation
Images - Charts, screenshots, visual data
How to Upload:
Click the attachment icon in the chatbox
Select your file
Ask Zoë to analyze or incorporate the data
Use Case Example: Upload last quarter's vendor report (PDF) and ask Zoë to compare it with your internal procurement data.
Choosing Your AI Model
Different analyses benefit from different AI approaches. Zoë lets you select which language model to use:
Available Models:
Default Model - Set by your admin, optimized for your industry
Alternative Models - Different approaches to analysis depth and style
When to Switch Models:
Detailed Analysis → Choose models optimized for depth
Quick Metrics → Use faster, lighter models
Creative Insights → Select models with broader reasoning
To change models, click the model selector dropdown in the chat interface.
Best Practices for Better Results
Be Specific About:
Timeframes: "last 30 days" vs "recently"
Metrics: "net revenue" vs "revenue"
Comparisons: "vs last quarter" vs "compare over time"
Dimensions: "by product and region" vs "broken down"
Progressive Refinement:
Start broad, then narrow:
"Show me sales performance"
"Break that down by product category"
"Focus on the top 3 categories by growth rate"
"What's driving the growth in category X?"
Navigation Tips
Starting New Analyses:
Click the left panel to see chat history
Click the icon in top left to start fresh
Use Cmd/Ctrl + K for quick new chat
Managing Your Chats:
Pin important conversations
Share insights with team members
Export chat history for documentation
Common Use Cases
For Sales Teams:
"What's our pipeline velocity by sales rep this quarter?"
For Marketing:
"Show me CAC by channel for the last 6 months with trend lines"
For Operations:
"Which warehouses have the highest fulfillment delays?"
For Finance:
"Compare budgeted vs actual spend by department year-to-date"
Troubleshooting
If Zoë doesn't understand your question:
Be more specific about metrics and timeframes
Break complex questions into steps
Check that referenced metrics exist in your data model
If results seem incorrect:
Verify the metric definitions being used
Check the timeframe and filters applied
Ask Zoë to explain her calculation method
Need Help?
In-app support: Click "Help" in left navbar
Contact support: [email protected]
You're Ready!
You now have everything you need to start using Zoë as your AI data analyst. Remember: there's no SQL to learn, no training required - just ask questions and get answers.
Your first challenge: Ask Zoë three questions about your business data today. Start simple, then build complexity. You'll be amazed at what you discover.
Zoë by Zenlytic - The best AI for data analysis and most advanced analytics agent for enterprise teams.
Was this article helpful?
[👍 Yes] [👎 No] [Send Feedback]
Related Articles:
Understanding Your Data Model with Zoë
Building Your First Dashboard
Advanced Query Techniques
Zoë for Retail Analytics
Popular Questions:
How do I share insights with my team?
Can Zoë access real-time data?
What's the difference between AI models?
How do I create custom metrics?