How to Turn a Web Page or HTML Into a PDF With Selectable Text

Every PDF page is a canvas that can hold two very different things. A text-based PDF stores the actual characters as text objects with an embedded or referenced font, positioned on the page. The reader knows that the shape in the top corner is the letter A. An image-based PDF stores a single flattened picture of the page: a grid of colored dots that happens to resemble words. The reader has no idea any letters exist.

On screen at 100 percent zoom, the two can be indistinguishable. The difference only shows up when you interact with the file. In a text PDF you can drag-select a paragraph, hit Ctrl+F to jump to a word, and copy a quote straight out. In an image PDF, your cursor slides across the page selecting nothing, search returns no results, and copy gives you an empty clipboard.

There is also a size and sharpness gap. A text-based PDF of a long article is often around 100 to 300 KB, because text and font data are compact. An image-based capture of that same page can run 2 to 6 MB, because it is storing every pixel. Zoom past roughly 150 percent and the image PDF starts to look soft and jagged, while text stays razor sharp at any magnification because it is redrawn, not enlarged.

You do not need any special software to check a PDF you already have. Open it in any viewer and run one of these quick tests. If a test works, you have real text. If it does nothing, you have an image.

The fastest single check is search: press Ctrl+F and type a word you can clearly see on the page. If the viewer highlights it, the text is real. If it insists there are no matches for a word that is obviously right there, the page is an image.

The same logic applies before you commit to a conversion method. If a tool or workflow ever shows you a preview and you cannot select the words in it, whatever it exports will not be searchable either.

Text-based is the right default for almost everything, but the image kind is not a mistake to always avoid. It has genuine uses. The question is whether you care more about reusing the content or about freezing exactly how it looked.

Choose based on what you will do with the file next, not on which button is easier to press.

The Print to PDF feature built into Chrome, Edge, Firefox, and Safari is the most reliable way to get a text-based PDF, and it is free and already installed. It works because the browser is a full rendering engine: it lays the page out and writes the characters into the PDF as real text, fonts and all. It does not take a screenshot.

Open the page you want, press Ctrl+P (Cmd+P on a Mac), and in the Destination or Printer dropdown choose Save as PDF (in some browsers it is labelled Microsoft Print to PDF or Export to PDF). Set the paper size and margins, then save. The result is searchable and, in Chrome and Edge, keeps your hyperlinks clickable.

One setting trips almost everyone up. By default, browsers strip background colors and background images to save ink, so a page with a dark header or colored panels can print as plain white. In the print dialog, expand More settings and turn on Background graphics (Safari calls it Print backgrounds). Also preview every page: fixed headers, sticky bars, and cookie banners sometimes overlap content or repeat, and it is easier to close them before printing than to fix the PDF afterward.

Print to PDF is perfect when you are looking at the page in a browser. But sometimes you have raw HTML as text, such as an email template, a generated invoice, or a code snippet, and no live page to open. Or you want to convert a URL in bulk without clicking through each one. That is where an online HTML-to-PDF converter earns its place: you paste the markup or the address and it renders a proper text-based PDF for you.

The important thing is to pick a converter that renders the HTML rather than one that screenshots it. A renderer produces selectable text; a screenshot service produces the image kind. If the tool offers an image export as well, that is a feature, not a flaw, as long as the PDF path gives you real text. You can convert HTML to a selectable-text PDF online and confirm the result with the three-second search test before you rely on it.

Two practical cautions when converting a live URL rather than pasted HTML. Content that sits behind a login, a paywall, or a cookie wall may not be reachable by the converter, so it captures the gate instead of the article. And pages that build themselves with JavaScript after loading, or that lazy-load images as you scroll, can come out with blank spaces if the converter grabs the page before everything has settled.

If you are generating PDFs from a server or a build step, the same principle holds: use a real rendering engine and you get text-based output for free. All three common options below produce selectable text, not screenshots.

Headless Chrome is the simplest one-liner. Running chrome --headless --print-to-pdf=out.pdf followed by a URL prints the rendered page to a text-based PDF, the same engine as the browser dialog. Puppeteer wraps this in Node with page.pdf(), which returns a text PDF and gives you control over paper size, margins, and headers. The one flag people forget is printBackground, which defaults to false, so pass printBackground: true to keep background colors and images. wkhtmltopdf is a long-standing standalone binary that converts an HTML file or URL directly; it uses an older WebKit engine, so very modern CSS such as flexbox or newer grid features may render differently, but it still emits real text.

For JavaScript-heavy pages, wait for the content to finish before you capture. In Puppeteer, load the page with a waitUntil setting like networkidle0 and, if a specific element appears late, wait for that selector too. That single step prevents most of the blank-section and missing-image problems that make people wrongly blame the converter.

Most HTML-to-PDF complaints trace back to a handful of predictable issues, and each has a clean fix. Working through this list will resolve the large majority of ugly or broken exports.

If your file ends up unexpectedly large or refuses to become searchable no matter what you try, that is the tell-tale sign it went through an image-based path somewhere. Switch to a rendering method, or if you must keep an image PDF, you can still compress the resulting PDF to bring the size down before sending it.