Extract Data
Browser NodeTurn a list on a page into clean, structured records
Overview
The Extract Data node reads repeating items on a page — product cards, table rows, search results — and turns each one into a tidy record with the fields you name. Point it at the container that wraps one item, list the pieces you want from inside each item, and it hands you back an array you can loop over, save to a sheet, or send to another app.
One container, many records
Container selectorA selector that matches ONE element per record — like each product card or table row. Leave it blank to treat the whole page as a single record.
article.product_podFields to extractOne entry per column you want: { "name": "title", "selector": "h3", "type": "text" }. Selectors are searched inside each container. Use "type": "attribute" with "attribute": "href" to grab a link instead of text.
[ { "name": "title", "selector": "h3", "type": "text" } ]Multiple recordsOn = one record per matching element (an array you can loop over). Off = a single record from the first match.
OnOutput variableThe name your list is saved under. Use this exact name as the input for an "Apply to Each" loop.
recordsUsage Example
article.product_pod[
{ "name": "title", "selector": "h3", "type": "text" },
{ "name": "url", "selector": "a", "type": "attribute", "attribute": "href" }
]Best Practices
Tip: Loop over the results
records) straight into an "Apply to Each" loop to do something with every item — add each to a spreadsheet, send an email, and so on.Tip: Keep field selectors simple
h3 or .price usually work best.