Data analysis has always been a "technically demanding" job. But AI tools are breaking down that barrier.

1. Excel + AI: The Lowest Threshold to Start

1.1 Excel Data Cleaning Prompt

I have an Excel spreadsheet with the following data issues:
Inconsistent date formats (some are 2024/1/1, some are Jan 1, 2024)
Some rows have the "元" character in the amount column (e.g., 500元)
Duplicate rows
Empty values
Please give me specific Excel functions/steps to clean these issues.

1.2 Let AI Write Excel Formulas

My data is in column A (product name) and column B (sales amount).
I want to:
Find the top 5 products by sales amount
Calculate each product's percentage of total sales
Mark products with sales below average (in red)
Please give me the corresponding Excel formulas and conditional formatting steps.

1.3 ChatGPT Code Interpreter: Direct Upload and Analyze

Using ChatGPT Plus's "Data Analysis" mode:

Directly upload CSV/Excel files

Describe analysis needs in natural language

AI automatically writes and executes Python code

Generates visual charts

Prompt template after uploading a file:
This is our company's sales data for the past 12 months.
Please help me analyze:
Monthly sales trend (line chart)
Sales share by product category (pie chart)
Which month performed best/worst and possible reasons
Any obvious seasonal patterns

2. Python + AI: Advanced Analysis

2.1 Let AI Write Data Analysis Code

Even if you don't know Python, you can use AI to generate code:

Write a Python data analysis script for me, requirements:
Data source: CSV file with columns: date, product, region, sales amount, quantity
Analysis goals:
Total sales by region (bar chart)
Monthly sales trend per product (multi-line chart)
Identify top 10% and bottom 10% outlier records
Output an Excel report
Technical requirements: use pandas + matplotlib + openpyxl

2.2 AI Explains Errors and Debugging

When code throws an error:

I ran this code and got the following error:
[error message]
The code is:
[code]
Please explain the cause of this error and provide a fix.

3. BI Tools + AI

3.1 AI Features of Major BI Tools

ToolAI FeatureUse Case

Power BI CopilotGenerate reports and interpret data via natural languageEnterprise BI Tableau AISmart chart type suggestionsData visualization Metabase AIQuery databases via natural languageTechnical teams Feishu Multidimensional Table AIChinese data analysisDomestic enterprises

3.2 Using Power BI Copilot

In Power BI's Copilot panel, enter:
"Create a report showing sales performance comparison across regions,
 including month-over-month growth rate, and highlight regions that exceeded targets"
Power BI will automatically:
Select appropriate chart types
Set data fields
Add conditional formatting

4. Data Analysis Automation

4.1 Automated Periodic Reports

python
Use Python + OpenAI API to automatically generate data reports
import pandas as pd
from openai import OpenAI
def generate_weekly_report(data_path: str) -> str:
    df = pd.read_csv(data_path)
    
    # Calculate key metrics
    stats = {
        'total_sales': df['sales'].sum(),
        'avg_sales': df['sales'].mean(),
        'top_product': df.groupby('product')['sales'].sum().idxmax(),
        'growth_rate': (df['sales'].iloc[-7:].sum() / df['sales'].iloc[-14:-7].sum() - 1) * 100
    }
    
    client = OpenAI()
    response = client.chat.completions.create(
        model="gpt-4o-mini",
        messages=[{
            "role": "user",
            "content": f"""
            Based on the following data, write a concise weekly report summary (within 200 words):
            Total sales: {stats['total_sales']:,.0f} yuan
            Average: {stats['avg_sales']:,.0f} yuan
            Best product: {stats['top_product']}
            Weekly growth rate: {stats['growth_rate']:.1f}%
            
            Use formal but readable Chinese, highlighting strengths and issues to watch.
            """
        }]
    )
    return response.choices[0].message.content

---

Further Reading

Python Data Science AI Complete Guide 2026

AI Financial Report Automation Case Study