# Convert Word to Searchable PDF Without Losing the Text

- Author: Schema Law (https://www.schemalaw.com/)
- Published: 2026-04-06T14:28:51Z
- Updated: 2026-10-11T19:22:20Z
- Category: Searchable PDFs
- Canonical: https://www.schemalaw.com/blog/convert-word-to-searchable-pdf

> Convert Word to searchable PDF by saving as PDF, not printing and scanning. What Word's export keeps, where pictures break search, and a short test.

To convert Word to a searchable PDF, save the document as a PDF from Word itself. The export writes the typed characters into the file, so the PDF is searchable with no OCR. Search fails only where the document holds pictures of text, such as a scanned signature page or a pasted exhibit.

## Is a PDF saved from Word already searchable?

Yes. A PDF saved from Word stores the characters you typed, with their fonts and positions, so a PDF reader can search and copy them at once. No recognition step is involved, which means no recognition errors. The text in the PDF is the text in the document.

Microsoft's instructions for [saving or converting to PDF in the Office desktop apps](https://support.microsoft.com/en-us/office/save-or-convert-to-pdf-or-xps-in-office-desktop-apps-d85416c5-7d77-4fd6-a216-6f4bf7c7c110) give the Windows steps for Word: choose **File**, then **Save As** or **Save a Copy**, and "In the drop-down list, select PDF." That is the whole conversion.

Before the PDF goes out, search the pleading caption, then search for answer and for counterclaim if this document responds to another. Select a contract clause the file quotes and compare the selection with the page. A pasted exhibit that will not select is still a picture. Saving from Word does not put text behind that part.

One option on that page can undo it for part of a document. Under **Options**, Microsoft describes a setting called "Bitmap text when fonts may not be embedded" this way: "If fonts can't be embedded into the document, the PDF uses bitmap images of the text so that the PDF looks the same as the original document." Text turned into bitmap images is a picture and cannot be searched. If a pleading uses an unusual licensed font, search for a word set in that font after you save.

## Save as PDF versus print to PDF: which keeps a Word document's text?

Save as PDF is the path Microsoft documents, and it has options that printing lacks. Printing to a PDF printer sends the pages through a print driver, and what the driver keeps depends on the driver. Use Save As, and treat any printed PDF as untested until you have searched it.

The Save As options Microsoft lists for Word include several that matter in a legal file:

| Option in Word's PDF export | What Microsoft says it does |
| --- | --- |
| Create bookmarks using | Creates "a set of bookmarks in the PDF file" from headings or Word bookmarks |
| Document properties | Includes the document's properties in the PDF |
| Document structure tags for accessibility | Makes the document "easier for screen-reading software to read" |
| Document showing markup | Includes tracked changes in the PDF |
| ISO 19005-1 compliant (PDF/A) | Outputs "version 1.7 PDF, which is an archiving standard" |

Two of those cut both ways. Leave "Document showing markup" off unless you mean to send tracked changes, and decide on purpose whether document properties should travel with the file.

Microsoft's own page shows the gap between the two paths. For Word on the web, it says to choose **Print** and then **Save as PDF**, which "will generate a basic PDF that preserves the layout and format of your original document", and it adds: "If you need more control over the options for your PDF, such as adding bookmarks, use the desktop app to convert your doc to a PDF." The same page notes a known issue in Word for Mac, where with "Best for printing" selected "the hyperlinks may not convert correctly."

The worst path is printing the document on paper and scanning it. That turns every character into a picture. The post on native and scanned PDFs sets out [what printing and scanning a native PDF throws away](https://www.schemalaw.com/blog/native-pdf-vs-scanned-pdf#printing-and-scanning-a-native-pdf-what-the-habit-throws-away).

## Why are the signature page and pasted exhibits in a Word PDF not searchable?

Those pages are pictures. A scanned signature page inserted into the document, a screenshot of an email, or a photographed exhibit pasted onto a page is an image in Word and stays an image in the PDF. The export carries over typed text. It does not read pictures.

A brief with a typed body and a scanned exhibit behind it therefore produces a mixed PDF: search works through the argument and stops at the exhibit. The same happens when a signed page is scanned and swapped in for the unsigned one.

Find the picture pages before the file goes out:

1. In the PDF, drag across a line on a suspect page. Typed text highlights as you drag. A picture highlights as one block or not at all.
2. Search for a word you can read on that page.
3. List the page numbers where the search finds nothing.

Then decide. If the exhibit exists as a typed file, insert the text version or attach its native PDF. If it exists only on paper, run OCR on those pages. The guide to [making picture text inside a typed PDF searchable](https://www.schemalaw.com/blog/how-to-make-pdf-text-searchable) covers that mixed case in full.

## PowerPoint slides and spreadsheets saved as searchable PDFs

PowerPoint and Excel save to PDF the same way Word does, and typed text in slides and cells stays searchable. The exceptions match Word's: a chart pasted as a picture, a screenshot on a slide, or a scanned page dropped into a deck is an image and will not be found by search.

Microsoft's page gives the path for each program. In PowerPoint, select **File**, then **Export**, and choose PDF as the type. In Excel, select **File**, then **Save As** or **Save a copy**, and pick PDF in the type list. Both dialogs offer "Standard (publishing online and printing)" or "Minimum size (publishing online)".

Three checks specific to these files:

- **Spreadsheets:** Microsoft warns that "the internal links within the spreadsheet will get lost during the conversion so the links will need to be verified." Also confirm which sheets were exported and that wide columns were not cut off at the page edge, because text that falls outside the print area is not in the PDF.
- **Slides:** Microsoft says the PowerPoint dialog's **Options** button is where you "set the page to be printed" and "choose whether markup should be printed". If the deck has speaker notes or reviewer comments, search the PDF for a phrase from them to learn whether they went into the file.
- **Both:** numbers in a chart image are pixels. If a damages chart must be searchable, include the table behind it.

## Checking a PDF made from Word before it leaves the office

Open the PDF itself, not the Word file, and run four checks: search a late page, search a footnote, test the pictures, and look at what else traveled with the text.

- **Late page.** Search for a word from the final page of the body.
- **Footnote and header.** Search for a word from a footnote and for the case number in the caption.
- **Signature page and exhibits.** Search for a word on each inserted page, or note that those pages are images.
- **Unusual fonts.** Search for a word set in any special font.
- **Markup.** Confirm no tracked changes or comments are showing.
- **Properties.** Open the document properties and read the title and author fields.
- **Bookmarks.** If your court or recipient expects them, confirm they exist and work.

A fictional example: an associate saves a 22-page motion for Halvorsen Dairy Cooperative as a PDF. A search for "Halvorsen" finds 31 hits through page 19 and none on pages 20 to 22. Those three pages are a scanned declaration signature page and a two-page pasted invoice. She runs OCR on those pages alone and searches again.

If you want the picture pages read, Schema Law accepts a PDF, PNG or JPEG of up to 20 MB and exports a searchable PDF; the [export formats section](https://www.schemalaw.com/#formats) lists what each format holds. Keep the PDF saved from Word as your master for the typed pages, since its text is exact.

Be careful with later steps. Stamping, flattening, printing to PDF again or passing the file through a portal can remove text that was there when you checked. The post on [why a PDF is no longer searchable](https://www.schemalaw.com/blog/pdf-no-longer-searchable) lists the steps that strip it. Court rules on text-searchable filings vary, so read your own court's requirements; the post on [searchable PDFs for court filing](https://www.schemalaw.com/blog/searchable-pdf-for-court-filing) explains what to look for.

## Try it on a real document

Upload a scanned PDF and export it as Word, a searchable PDF or plain text on the free plan. Your documents are never used to train an AI model. [Start free](https://www.schemalaw.com/api/auth/google)