# Your First Chat

Ready to uncover insights? This guide will take you from a simple file upload to complex reasoning in three steps.

## 1. Start with a "Quick Chat"

The fastest way to interact with your data is the standard **Quick Chat** mode. This is perfect for immediate answers without setting up a workspace.

1. **Upload Data:** Drag & drop your spreadsheet (CSV/Excel) directly into the chat window.
2. **Ask Basic Questions:** Get instant feedback on the facts contained in your file.
   - _"Show me total sales by month."_
   - _"What is the average order value?"_

> **Note:** Quick Chat is designed for speed. It handles direct queries well but does not support complex reasoning or long-term memory.

* * *

## 2. Go Deeper: Create a Project

When you need to understand _"Why"_ something happened (Attribution Analysis) or need the AI to perform complex multi-step tasks, you must move to a **Project**.

**How to upgrade:**

- **From Chat:** If you are already in a Quick Chat, click **"Save as Project"** to migrate your current data and context into a dedicated workspace.
- **From Scratch:** Click **"\+ Create Project"** to start fresh.

Creating a project unlocks the **"Deep Analysis"** capability and the **Smart Semantic** engine.

* * *

## 3. Unlock Deep Analysis

Once you are inside your new Project:

1. **Toggle the Switch:** Locate and enable the **"Deep Analysis"** toggle in the input bar.
2. **Ask Advanced Questions:** Now that ChartGen AI has structured your data semantically, you can ask complex analytical questions:
   - _"Analyze the root cause of the sales drop in Q3."_
   - _"Identify anomalies in user traffic and suggest improvements."_
   - _"Forecast revenue for the next quarter based on current trends."_

### Summary: Which mode should I use?

| Feature | Quick Chat | Project (Deep Analysis) |
| --- | --- | --- |
| **Best for...** | "What" questions & Fact-checking | "Why" & "How" questions |
| **Data Handling** | Reads raw file content | Builds Semantic Models |
| **Complexity** | Simple lookups & Summaries | Root Cause, Forecasting, Attribution |