How to Extract Data from a PDF Using PDF.co Document Parser

Open Document Parser

1. Log in to your PDF.co account.
2. Go to the API Tools tab.
3. Select Document Parser Manager.

Create a New Template

1. Click New Template.
2. This opens the Document Parser Template Editor.

Load a Test PDF or Image

1. In the Template Editor, click Load Test PDF or Image.
2. Upload your source PDF file.
3. Click Add Object
4. Select Add Field under the Rectangle selection option.

Drag a Rectangle to Extract Data

1. Drag the rectangle to the location on the PDF where the target keyword/data appears.
2. Configure/select the keywords you want to extract.
3. Run the template to preview the output.

Review the Extracted Data

You will now see the extracted data in the output preview.

If the results look correct:

• Click Save Template and Return to save the template for future use.

Open API Tester

From your PDF.co dashboard, click API Tester.

Set Up the API Request Tester

1. In the left-hand menu, select the Document Parser endpoint
2. Click Try it

1. Enter:
   • The input file URL (must be publicly accessible)
   • Your template ID
2. Run the request

Run the Request and View Results

Once executed successfully, the API response will return extracted keyword data.

The output JSON will appear on the right side of the tester panel.

Example JSON Output

You will receive structured extracted keyword-based results in JSON format, based on your saved template.

{
	"body": {
		"objects": [
			{
				"name": "Invoice Number",
				"objectType": "field",
				"value": "123456",
				"pageIndex": 0,
				"rectangle": [
					520.5036,
					114.311028,
					50.0916519,
					14.1283293
				]
			},
			{
				"name": "Invoice Date",
				"objectType": "field",
				"value": "01/01/2016",
				"pageIndex": 0,
				"rectangle": [
					503.8064,
					131.650345,
					68.0732651,
					14.1283293
				]
			}
		],
		"elapsed": 0.0157471,
		"templateName": "Test_Template",
		"templateVersion": "4",
		"timestamp": "2026-01-18T11:43:33"
	},
	"pageCount": 1,
	"error": false,
	"status": 200,
	"name": "view.json",
	"credits": 42,
	"remainingCredits": 156721,
	"duration": 4635
}