How to Convert a PDF to Text: 6 Methods That Actually Work

You have a PDF and you need the words out of it as editable text. The method that works depends on one thing: whether the file already holds a real text layer or is just a picture of text. Get that wrong and you will spend twenty minutes fighting a tool that was never going to succeed.

This guide walks through six ways to pull text from a PDF, when each one works, and where each one breaks. Start by identifying which kind of PDF you have, because that single check decides everything that follows.

First, figure out which kind of PDF you have

There are two kinds of PDF, and they behave completely differently.

Text-based (digital) PDFs were created by software: a Word export, a Google Docs download, an invoice generated by an accounting app. The characters live inside the file as actual text. You can select them, copy them, and search them.

Scanned (image) PDFs are photographs of a page. A scanner, a phone camera, or a fax turned paper into an image and wrapped it in a PDF. There is no text inside, only pixels that look like text to your eyes.

Here is a five-second test. Open the PDF and try to select a single line with your cursor, or press Ctrl+F (Cmd+F on Mac) and search for a word you can see on the page. If the text highlights or the search finds it, you have a text-based PDF and Methods 1, 3, 4, and 5 will work fast. If nothing selects and search finds nothing, you have a scanned PDF, and you need OCR (Methods 2, 4, 5, or 6).

Method 1: Copy and paste from a text-based PDF

For a digital PDF, the fastest route needs no tool at all.

1. Open the PDF in any reader (your browser works).
2. Press Ctrl+A / Cmd+A to select everything on the page, or click and drag to grab a section.
3. Copy with Ctrl+C / Cmd+C.
4. Paste into a text editor, Word, or Google Docs with Ctrl+V / Cmd+V.

When it works: Short documents, single-column layouts, and any time you just need the raw words.

Where it breaks: Copy-paste does nothing on a scanned PDF because there is no text to copy. It also scrambles the reading order on multi-column pages, so a two-column research paper often pastes as a jumble where the left and right columns interleave line by line. Tables usually collapse into a run-on string with no cell boundaries. If you hit any of that, move to a converter that rebuilds structure.

Method 2: OCR for scanned or image PDFs

Optical character recognition (OCR) looks at the image of a page and works out which shapes are letters, then writes them out as real, editable text. This is the only thing that turns a scanned PDF into something you can copy.

OCR quality depends heavily on the scan. You get clean results from a sharp, straight, high-contrast scan at 300 DPI or higher. You get errors from low-resolution scans (below about 150 DPI), skewed pages, faint print, background noise, and handwriting. Common mistakes include swapping "rn" for "m," "0" for "O," and dropping accented characters in other languages.

You have OCR built into several tools you may already own: Google Docs (Method 4), Adobe Acrobat (Method 5), and dedicated online OCR services (Method 3). An AI reader like PdfGPT also runs OCR automatically when you upload a scan, which matters if your goal is to read or question the document rather than reformat it. PdfGPT reads scanned PDFs with OCR in about 10 languages and then lets you pull answers straight from the recognized text.

One rule for any OCR job: proofread the output. OCR is never 100% accurate, so check names, numbers, dates, and anything you will act on.

Method 3: Online PDF-to-text converters

Web converters like Smallpdf, iLovePDF, PDF2Go, and Adobe's online tools let you drop in a file and download text without installing anything. Most run OCR on scanned files and export to TXT, DOCX, or a searchable PDF.

Typical steps:

1. Open the converter site.
2. Upload your PDF or drag it into the drop zone.
3. Pick your output (TXT for plain text, DOCX to keep some layout).
4. If the file is scanned, turn on OCR and choose the document language.
5. Convert, then download.

When it works: A one-off conversion when you do not want to install software.

Where it breaks: Free tiers cap file size, page count, or the number of daily conversions, and many watermark or limit larger jobs. The bigger issue is privacy. You are uploading your document to a third-party server, so never run confidential contracts, medical records, or anything sensitive through a random free converter. Read the retention policy first, or use a local tool instead.

Method 4: Google Docs (free OCR for most people)

Google Docs runs OCR at no cost, which makes it the go-to free option for a scanned PDF.

1. Upload the PDF to Google Drive.
2. Right-click the file in Drive.
3. Choose "Open with" then "Google Docs."
4. Wait while Google processes the file. It opens a new Doc with a copy of the original image at the top and the extracted, editable text below it.
5. Delete the image, clean up the text, then use File > Download to save as DOCX, TXT, or PDF.

When it works: Printed text in a clean scan, one language, simple layout.

Where it breaks: Google Docs strips most formatting during OCR, so columns, tables, and fancy fonts rarely survive. It also handles complex layouts poorly and works best on straightforward pages. Check the current file-size and page limits before you rely on it for a long document, since Google caps what it will process per file. Always compare the output against the original.

Method 5: Adobe Acrobat export

If you have Adobe Acrobat Pro (not the free Reader), it gives you the most control over layout and runs OCR automatically on scans.

To export as plain text or Word:

1. Open the PDF in Acrobat.
2. Open the "Export a PDF" tool (or File > Export To).
3. Choose your format. Pick "Text (Plain)" for a clean TXT file, or Microsoft Word for a DOCX that keeps more of the layout.
4. If the PDF is scanned, Acrobat runs text recognition on its own before exporting.
5. Name the file and save.

Before exporting to Word, you can choose whether to keep flowing text or preserve the page layout, which changes how well tables and columns hold together.

When it works: You need the closest match to the original formatting, you handle sensitive files and want to keep them off third-party servers, or you convert PDFs often.

Where it breaks: Acrobat Pro is a paid subscription. The free Adobe Reader can view PDFs but cannot export to editable text without a paid plan or a Pro trial. Even Acrobat struggles with borderless tables and heavily designed pages.

Method 6: Use an AI tool to extract and structure the text

Sometimes you do not want a converted file at all. You want the information inside the PDF: the key figures from a 90-page report, the payment terms buried in a contract, the method section of a research paper. Copying every word and reading it yourself wastes the time you were trying to save.

An AI document reader handles that job. Upload the PDF, and the tool runs OCR on scanned pages, reads the text, and answers questions about it directly. This is where PdfGPT fits. It reads scanned and digital PDFs, and runs OCR in about 10 languages, keeps tables and code blocks intact while it reads, and gives page-cited answers so you can trace any claim back to its source page. You can summarize a document, ask follow-up questions, or extract specific sections without wrestling with formatting.

Be clear about what this is and is not. PdfGPT reads, summarizes, extracts, and answers questions about your PDF. It does not export a formatted DOCX or a spreadsheet of your tables. If your real goal is to understand or query the document, this saves the most time. If you specifically need an editable Word or TXT file, use Method 3, 4, or 5. You can also start by summarizing the PDF to decide whether the full conversion is even worth doing.

Why tables and columns come out mangled

Every method on this page can wreck a table, and it helps to know why so you set the right expectations.

A PDF stores where each character sits on the page, not that a group of characters belongs to a table cell. Plain copy-paste and basic converters read left to right, top to bottom, so a two-column layout gets stitched into one scrambled column and table cells run together. OCR adds a second problem: it has to guess the grid from visual cues, and borderless tables (common on invoices and financial statements) remove the ruled lines that most tools depend on. Merged cells and cells that span rows break the logic further.

Formatting metadata makes it worse. When OCR reads a scan, details like bold, italic, and exact font size are generally gone for good, because that information never existed as data in an image.

Three ways to reduce the damage:

• For digital PDFs, export to DOCX rather than TXT so the tool tries to rebuild table structure.
• For scans with tables, use a tool built for table extraction rather than a generic converter.
• Always keep the original PDF open next to your output and fix the cells by hand.

Which method should you use?

Match the method to your file and your goal.

• Text-based PDF, quick job: Copy and paste (Method 1). Free, instant, no tool.
• Text-based PDF, want to keep layout: Adobe Acrobat export to Word (Method 5), or a good online converter (Method 3).
• Scanned PDF, free: Google Docs (Method 4). Zero cost, decent on clean printed scans.
• Scanned PDF, best formatting and privacy: Adobe Acrobat Pro (Method 5), because it runs locally and gives you layout control.
• Scanned PDF, other languages: A converter or AI reader with multi-language OCR. PdfGPT covers about 10.
• Sensitive document: Keep it local. Use Acrobat or offline software, not a free web uploader.
• You want answers, not a file: An AI reader like PdfGPT (Method 6). Skip the conversion and question the document directly.

One more rule for every path: proofread. No method reaches perfect accuracy on scans, so review the output before you trust it.

Frequently asked questions

How do I know if my PDF is scanned or text-based?

Open the file and try to select a line of text with your cursor, or search for a visible word with Ctrl+F / Cmd+F. If the text highlights or the search finds it, the PDF is text-based. If nothing selects, it is a scanned image and you need OCR.

Can I convert a PDF to text for free?

Yes. For a text-based PDF, copy and paste costs nothing. For a scanned PDF, Google Docs runs OCR free of charge. Free online converters also work but often cap file size, page count, or daily use, and they upload your file to a third-party server.

Why is my converted text full of errors and scrambled tables?

Two reasons. OCR misreads low-quality, skewed, or low-resolution scans, so aim for a straight 300 DPI scan. And most tools read a page top to bottom, which scrambles multi-column layouts and tables because a PDF does not record which characters belong to which cell. Export to DOCX instead of TXT to keep more structure, and proofread the result.

How do I convert a scanned PDF in a language other than English?

Use a tool with multi-language OCR and select the correct document language before you run it, because that setting sharply improves accuracy on accented characters. Google Docs, Adobe Acrobat, and AI readers like PdfGPT all handle multiple languages; PdfGPT covers about 10.

Can PdfGPT convert my PDF to a Word or TXT file?

No. PdfGPT reads, summarizes, extracts, and answers questions about your PDF, including scanned files it processes with OCR. It does not export a formatted Word or TXT file. Use it when you want to understand or query the document. For an editable file, use Google Docs, Adobe Acrobat, or an online converter.