JSON exports are ideal for:
API integrations: direct data pipeline connections.
Developer workflows: parse and process programmatically.
Database imports: maintain data structure integrity.
Complex data: preserve nested relationships.
Automated processing: machine-readable format.
AI/LLM analysis: structured data for language models.
How to export as JSON
Prepare your view
Apply filters for specific data
Hide unnecessary columns
Set historical data preference
Export
Click Export button
Select Download as JSON
Click Download
Access your file
Direct download to browser
Or check email for link
What you'll get
Browse AI JSON structure
All JSON exports include metadata wrapping your data.
You'll get a single JSON file of your data
Contains ALL nested data from every list
Complete data relationships preserved
{
"data": [...your extracted data...],
"table": "Table Name",
"schema_version": "1.0",
"export_id": "336b8f3b-c1ee-45d5-b06f-41a04910cc40",
"export_created_at": "2025-12-18T15:31:06.018Z"
}
Single list export
When exporting from a specific list tab you'll get a single JSON file with an array of objects (one per row).
Entire dataset export
If you export from the Main tab you'll get a single JSON file of all your data with the complete data relationships preserved.
💡 Main tab JSON exports include all nested list data in a single file, which can result in large file sizes. Consider exporting individual list tabs instead.
Understanding the data structure
List tab structure
{
"data": [
{
"Position": "1",
"APR": "6.26%",
"Interest Rate": "5.99%",
"Monthly Payment": "$1,647",
"Extract Date": "2025-12-17T16:46:33.783Z",
"Origin URL": "https://example.com/rates"
},
{
"Position": "2",
"APR": "6.855%",
// ... more fields
}
],
"table": "30-Year Refinance Rates",
"schema_version": "1.0",
"export_id": "336b8f3b-c1ee-45d5-b06f-41a04910cc40",
"export_created_at": "2025-12-18T15:31:06.018Z"
}
Main tab structure
{
"data": [
{
"Extract Date": "2025-12-17T19:57:11.707Z",
"Status": "Successful",
"Search Keyword": "laptop",
"Max Products": "200",
"Product list": [
// ALL products nested here
{
"Position": "1",
"Title": "Lenovo IdeaPad...",
"Price": "CAD 399.81",
"Rating": "4.8 out of 5 stars",
// ... all product fields
},
// ... 199 more complete products
]
}
// Plus all other extractions with their full data
],
"table": "Main",
// ... metadata
}What's preserved
All text exactly as scraped, ex: "$99.99", "5 stars", etc.
Field names
Empty values (typically as
null)URLs
Special characters (properly escaped)
Complete data structure and all relationships intact.
💡 Browse AI captures data as it appears on websites, so all values export as strings (e.g., "$99.99" not 99.99). You'll need to parse these if you need numeric values.
Working with JSON exports
Accessing your data
Remember to access the "data" array:
const exportFile = // your JSON file
const extractedData = exportFile.data; // This is your actual data array
For databases
Import directly into:
MongoDB - native JSON support
PostgreSQL - JSON/JSONB columns
MySQL - JSON data type
For analysis tools
Python pandas -
pd.json_normalize()for nested dataR - Import with jsonlite package
Tableau - JSON connector available
Power BI - Import and flatten structure
For AI/LLM processing
JSON maintains structure perfectly for AI analysis:
Feed product catalogs to ChatGPT/Claude for analysis.
Structure customer reviews for sentiment analysis.
Organize competitor data for market insights.
Format data for AI-powered recommendations.
The structured format helps AI models understand relationships between data points better than flat CSV files.
Common data transformations
Since all values are strings, you'll often need to clean them:
Strategies for lage JSON files
If the JSON file generated is too large for your use case here are some solutions.
Export list tabs individually
Manageable file sizes
Easier to process
Can target specific data
Filter before export
Limit date ranges
Filter by status
Hide historical data
Integrations
Export to AWS S3 for cloud processing
Use Browse AI API
Process via webhooks instead
Common data transformations
Since all values are strings, you'll likely need to clean or transform the data.
Parse numeric values
// Remove currency symbols and convert
const price = parseFloat(item.Price.replace(/[$,]/g, ''));
const rate = parseFloat(item.APR.replace('%', ''));
Handle dates
// Browse AI dates are already ISO format
const date = new Date(item["Extract Date"]);
Clean ratings
// Extract numeric rating
const rating = parseFloat(item.Rating.match(/[\d.]+/)[0]);