How to convert PDF to HTML for reading in a browser

PDF to HTML produces one .html file with the text of every page and links to each page. "Embed page pictures" adds an image of every page inside the file but makes it much larger. A scan without OCR shows "(empty page)".

Sometimes a document is easier to use in a browser than in a PDF viewer: you can find a word with Ctrl+F, copy a paragraph, or send a file that opens on any device. The PDF to HTML tool does this: it takes the text out of a PDF and lays it out page by page in a single HTML file.

It helps to know in advance what you will get. The output is a copy for reading and searching, not a designed web page. Below is what goes into the file, how the tool's two settings work, what to do with scans, and how to check the result quickly.

What the output looks like

The result is a single `.html` file, not a ZIP archive with a folder of images. By default it is named `converted_` plus the name of the source PDF; you can set a different name in the "Output file name" field. Styles and pictures (when enabled) live inside the file itself, so it opens offline, can be attached to an email or copied to a USB stick without losing anything.

Inside the file, every PDF page becomes a separate block with a heading such as "Page 1", "Page 2" and so on. The page text is shown in a monospaced font and keeps the line breaks and indentation of the original. The "Page N" labels and other service notes are written in the language the site is open in.

First check that the PDF contains text

The tool does not recognize letters in images; it uses the PDF's existing text layer. Checking for one is easy: open the PDF, select a sentence and paste it into any text editor. If you get normal words, the text layer is there. If nothing can be selected, or the paste is a string of random characters, the HTML will look the same.

The typical problem case is a scan or a photo of a document. Such a PDF converts without an error, but every block shows "(empty page)" instead of text. You have two options:

• Run OCR PDF first: it adds an invisible text layer to the scan. Download the recognized PDF and upload that file to PDF to HTML. The recognized text keeps its line breaks but not the alignment: columns and table cells end up separated by single spaces. Recognition is covered in detail in how to recognize text in a scanned PDF.
• If you do not need the text and only want to read the scan in a browser, turn on "Embed page pictures". The pages will appear as images, but the file becomes much heavier and searching for words will not work.

Step-by-step conversion

1. Open PDF to HTML and drag the PDF into the upload area, or click "or choose on your device". One file is processed per run. 2. In the "Parameters" block, type a name for the result in "Output file name" if you need one. 3. Decide whether you need page pictures: that is what "Embed page pictures" controls (off by default). 4. Keep or clear "Include page navigation" (on by default). 5. Click "Process file", wait for it to finish and click "Download".

If the PDF requires a password to open, processing stops with a message about the password. Remove the protection first; if you know the password, how to remove a PDF password explains how. Limits on file size and page count are shown on the tool page itself.

Text-only output: "Embed page pictures" off

This is the default, and it is enough for most tasks: the file is light, and the text can be searched, selected and copied. Do not expect the original page design to be rebuilt, though. The text is laid out roughly where it stood on the sheet, which means:

• two columns run side by side on the same lines: a line of the left column, then, after a run of spaces, a line of the right one;
• table columns are held together only by spaces; this is not a real HTML table;
• running headers, footers and page numbers stay inside the text of each page.

On a wide screen this layout reads reasonably well. On a phone, long lines wrap and the columns can get mixed up. If a document is set in two columns and people will read it on phones, check one such page beforehand.

Page pictures: when to turn on "Embed page pictures"

When the option is on, an image of each page in PNG format appears above that page's text, embedded directly in the HTML. The text below the picture stays, so searching and copying still work wherever the PDF has a text layer.

This is useful when the meaning depends on how the page looks: a wiring diagram in a manual, a drawing, a stamp and signature, formulas, a complex table. For example, in an equipment manual the diagram on page two does not appear at all in the text-only version; you only get its captions.

The price is file size. Every picture adds data, and the file ends up much larger than the text-only version. A long document with pictures may exceed an email attachment limit or open slowly on a low-end phone. The picture is just an image: the diagram does not become editable web graphics.

Page navigation

"Include page navigation" adds a bar of links at the top of the file: "Page 1", "Page 2" and so on. The bar is pinned: it stays at the top of the screen while you scroll, so any page is one click away.

The links lead to pages, not chapters; the source document's table of contents is not rebuilt. In a document with hundreds of pages the bar grows to several lines and takes up a noticeable part of a phone screen. In that case, clear "Include page navigation" and use the browser's search to find your place.

Checking the result and common problems

Double-click the downloaded file; it opens in your default browser. Then:

• press Ctrl+F (Cmd+F on a Mac) and search for a familiar word from the middle of the document;
• select and copy a paragraph, then paste it into an editor;
• scroll through a page with a table or columns;
• if navigation is on, follow the links to the beginning, middle and end of the document.

What usually goes wrong and how to spot it:

• "(empty page)" on all or nearly all pages: the PDF is a scan without a text layer and needs OCR;
• "(empty page)" only on some pages: those are either genuinely blank sheets or inserted images, such as a scanned appendix;
• words from two columns are mixed together: a side effect of the layout; compare with the page picture or with the PDF;
• the file is far too large: convert again with "Embed page pictures" turned off.

If you only need the text without markup, PDF to Text is more convenient: it saves a plain `.txt` file.

Before publishing on a website

The converted copy is not a finished web page. Its headings are "Page N" rather than the document's sections, tables are held together by spaces, and the page pictures are also labelled "Page N" with no description of their content. The language declared in the page code matches the language of the site you converted on, not the language of the document.

To publish, move the text into your site's editor, add real headings, rebuild the tables, describe the images and check how it looks on a small screen. Keep the source PDF for comparison, and give the HTML a clear name so the recipient does not mistake it for a new edition of the document.