Skip to main content

​
What does this node do?

The Data Analyzer node lets you analyze tabular data using natural language. Ask questions about your CSV data and get AI-powered insights without writing SQL or code. Common uses:
  • Summarize large datasets
  • Find patterns and trends
  • Answer specific questions about data
  • Generate insights and reports

​
Quick setup

1

Add the Data Analyzer node

Find it in AI Nodes → Data Analyzer
2

Provide CSV data

Connect a data source or paste CSV content
3

Ask your question

Write a natural language query about the data

​
Configuration

​
Required fields

​
csv_data
string
required
The CSV content to analyze.Sources:
  • Direct paste
  • From Google Sheets: {{GoogleSheets_0.csv}}
  • From file upload
​
query
string
required
Your question about the data in natural language.Examples:
  • “What is the total revenue by region?”
  • “Which products have sales above $1000?”
  • “What’s the average score for each category?”
  • “Find the top 10 customers by order value”

​
AI settings

Same as LLM node - model, temperature, etc.

​
Output

{
  "answer": "The total revenue by region is: North: $125,000, South: $98,000, East: $112,000, West: $89,000",
  "data": {
    "North": 125000,
    "South": 98000,
    "East": 112000,
    "West": 89000
  },
  "query": "What is the total revenue by region?",
  "rows_analyzed": 1500
}

​
Examples

​
Sales analysis

CSV Data: 
date,product,region,revenue
2024-01-01,Widget A,North,1500
2024-01-01,Widget B,South,2300
...
Query: “What was the total revenue last month, broken down by product?” Output: 
Total revenue for January 2024:
- Widget A: $45,000
- Widget B: $62,000
- Widget C: $28,000

Total: $135,000

​
Finding patterns

Query: “Are there any unusual patterns in the data?” Output: 
Notable patterns detected:
1. Sales spike every Friday (25% above average)
2. Widget B outperforms in South region (40% higher)
3. December sales 3x higher than other months

​
Data validation

Query: “Are there any missing values or data quality issues?” Output: 
Data quality findings:
- 12 rows have missing 'region' values
- 3 revenue values are negative (possible errors)
- 5 duplicate entries found

​
Best practices

​
Be specific in queries

❌ "Tell me about the data"
✅ "What is the average order value by customer segment?"

​
Reference column names

❌ "What's the total?"
✅ "What's the total of the 'revenue' column?"

​
Ask follow-up questions

Chain multiple Data Analyzer nodes for deep analysis:

LLM

For text analysis

Google Sheets

Get CSV from Sheets