How do I extract data from a scanned PDF using the PDF Extract API?

The PDF Extract API can work with both native (digital) and scanned PDFs. It uses Sensei, which is Adobe's AI technology, to understand the content. You just send your scanned PDF to the API, and it will process it and return structured data, even if the original was a scan.

What is the difference between PDF Extract API and Document Generation API?

Feature Purpose Direction. PDF Extract API Turns PDFs into structured data (JSON, Markdown) PDF -> Data Document Generation API Creates PDFs or Word documents from data and templates Data -> PDF/Word

Why should I use the PDF Extract API instead of just copying and pasting text from a PDF?

Using the PDF Extract API is much better for large amounts of data. Copying and pasting is slow and can lose the structure of tables or reading order. The API automatically gives you structured data like JSON, which is easy for other computer programs to use and analyze.

If my PDF has tables, how does the PDF Extract API handle them?

The PDF Extract API is very good at identifying tables within your PDF. When you extract data, it can convert these tables into formats like CSV or XLSX. This means you get your table data in a spreadsheet format, which is easy to open and work with in programs like Microsoft Excel.

What kind of output can I expect from the PDF Extract API?

Structured JSON or Markdown for text and general content. CSV or XLSX files for tables. PNG images for any pictures found in the PDF.

Extract data with custom parsing rules in Adobe Acrobat — How-to & AI prompts

AI prompts

3 ways to ask · copy any one

AExecute — “help me do it”Show me how to extract [data type] from PDF files using...…+

Show me how to extract [data type, e.g. invoice numbers] from PDF files using custom parsing rules with the Adobe PDF Services API in [language].

when the reply comes backWant to automate this for multiple files?

BImprove — “do it better”How can I improve my parsing rules to get more accurate...…+

How can I improve my parsing rules to get more accurate results when extracting [data type] from PDFs using the Adobe PDF Services API?

when the reply comes backNeed help setting up API authentication?

CDecide — “help me choose”Should I use custom parsing rules or regular extraction for...…+

Should I use custom parsing rules or regular extraction for getting [data type] from PDFs with the Adobe PDF Services API? What are the pros and cons?

when the reply comes backAsk about handling tricky PDF layouts.

How to do it

the tool · the steps · what to avoid

1
Open the PDF file containing the data you want to extract in Adobe Acrobat.
2
Go to the Tools pane and select Export PDF.
3
Choose a structured export format such as Spreadsheet or XML to facilitate custom parsing.
RecommendationSelecting a structured format makes it easier to apply custom logic to the exported data.
4
Export the file and save it to your local system.
5
Write a custom script or program (using Python, JavaScript, etc.) to parse the exported file according to your specific data extraction rules.
Best practiceUsing Python libraries like pandas or openpyxl can simplify parsing and transforming complex data structures from exported spreadsheets.
6
Run your script to extract and process the desired data from the exported file.

In the appWith the API

After exporting your PDF to a structured format, you can use your own code to extract exactly the data you need.

This runs in the Acrobat app - there is no separate API for this task.

Glossary

words on this page

Custom Parsing RulesSpecial instructions you set to pick out certain data from a PDF.ExampleYou set custom parsing rules to automatically pull out invoice numbers and dates from scanned bills.

EndpointA web address your code sends requests to for a specific job.ExampleThe app sends your payment details to a secure endpoint on the server to process the transaction.

AuthenticationHow you prove who you are when using an online service.ExampleEntering your password and a code from your phone provides authentication to access your bank account.

The real tasks

the one, listed here

Convert emailed bills into a standard, editable copy so data can be entered into our ledger.

FAQ

about this task

You create a set of instructions (usually in JSON) that tells the API what patterns or areas to look for in your PDF, then send these rules along with your API request. This helps the API know exactly what data you want to pull out.

Regular extraction tries to get all text or tables from a PDF, but custom parsing lets you tell the API exactly what to find, like only dates or certain numbers. This makes your data cleaner and saves time.

Yes, but the PDF must go through OCR (Optical Character Recognition) first, so the text can be read by the API. After OCR, your custom rules can find the data you need.

Using the API lets you automate data extraction for many files at once, or build the process into your own app or workflow. It saves time and removes manual work.

Check if your rules match the way the data looks in the PDF. Sometimes, patterns need to be more exact, or you may need to adjust for different layouts. Testing with sample files helps improve your results.

Extract data with custom parsing rules

When to use it

AI prompts

How to do it

Glossary

The real tasks

Who does this

FAQ

Sources