PDF SEO: How to Optimize PDF Files for Google and AI Assistants

Quick quiz

Is SEO right for you, and how should you start?

Question 1 of 425%

Do people search Google for what you sell?

Yes, I think soNo, it is new or very nicheI don't know

PDF SEO is the set of steps that make a PDF file findable in Google and readable by AI assistants: real text, a clear title, structure, links and a sound decision about which version (PDF or HTML) should rank. Google lists the PDF format among the file types it can index, so a white paper, a price list or a user manual can bring traffic. It can also compete with your own pages, or stay invisible because of a scan or a missing title. This guide covers how Google processes PDFs, a 9-step optimization checklist, the cases where an HTML page works better, and how to measure the results.

Does Google index PDF files?

Yes. Google has indexed PDFs since 2001. In its FAQ on PDFs in Google search results, Google states that it indexes the text of PDFs in any language, that it may run OCR on text embedded as images, and that the rule of thumb is simple: if you can copy and paste the text, Google can index it. Password-protected or encrypted files are excluded.

The same FAQ answers the other questions SEOs usually ask:

• Ranking: PDFs "rank similarly to other webpages", based on their content and on the pages that link to them.
• Links: links inside a PDF are treated like HTML links and can pass PageRank. You cannot mark them nofollow.
• Title in the results: Google uses the Title metadata of the file and the anchor text of links pointing to it.
• Images: at the time of the FAQ, images inside PDFs were not indexed; Google recommended HTML pages for images you want in image search.

Size is rarely a problem: according to the Googlebot documentation (updated February 2026), Googlebot crawls the first 2MB of an HTML file but the first 64MB of a PDF file. In the results, a PDF shows a "PDF" label next to its title. To check whether a document is in the index, search site:yourdomain.com filetype:pdf or inspect its URL in Search Console. For pages that refuse to get in, the general method is in our guide to Google indexing.

PDF or HTML page: which is better for SEO?

A PDF can rank, but it starts with handicaps that an HTML page does not have:

The conclusion is practical: keep PDFs for documents people download (spec sheets, manuals, forms, reports to print) and publish content you want to rank as web pages.

How to optimize a PDF for SEO in 9 steps

1. Export real text, never a scan

Export from Word, Google Docs or InDesign as a text PDF. If you only have a scan, run OCR (Acrobat's "Scan & OCR" tool, for example) and check the result: Google's OCR is a fallback, and a poor scan gives poor text. Avoid text drawn as vector outlines and decorative fonts.

2. Fill in the Title metadata

This is the PDF equivalent of the title tag. In Acrobat: File > Properties > Description > Title. Write 50 to 60 characters with the main keyword at the start, then in the Initial View tab set "Show" to Document Title, so readers see it too. Without it, Google falls back on the file name or on a line of the document.

3. Give the file a descriptive name

The file name becomes the URL: pdf-seo-checklist.pdf beats final_v3_2.pdf. Use lowercase words separated by hyphens and keep the same URL when you update the file, so links and rankings are preserved.

4. Complete the other properties

Fill in Subject (a one-sentence summary, which may be reused for the snippet), Author and the document language (Advanced > Reading Options in Acrobat). The Keywords field has no known effect on Google; leave it short.

5. Structure the document with tagged headings

A tagged PDF carries a structure (H1, H2, paragraphs, lists, tables) that screen readers, search engines and AI tools use to understand the hierarchy. In Word, tick "Document structure tags for accessibility" when saving as PDF; in Acrobat, use the accessibility tools to add tags automatically, then fix the order. One H1, logical H2s and short sections help both readers and parsers.

6. Add links, in both directions

Inside the PDF, link to the relevant pages of your site with descriptive anchors: those links pass signals, and they bring readers back to your navigation. Outside, link to the PDF from at least one indexable HTML page with an anchor that describes the document. That anchor also feeds the title Google shows.

7. Describe the images

Add alternative text to charts and figures through the tags (Acrobat: set the alternate text on each figure). It helps accessibility and gives search engines and AI tools a text version of the visual. For images you want in Google Images, the rules of image SEO on an HTML page remain the reliable route.

8. Keep the file light

Compress images, remove embedded fonts you do not need and save with Fast Web View (linearized PDF) so the first page displays before the whole file has downloaded. A 2MB brochure opens on a phone; a 40MB one gets abandoned.

9. Make it discoverable

Include PDFs you want indexed in your XML sitemap, do not block them in robots.txt, and check the HTTP headers they are served with: a stray noindex or a wrong canonical in the HTTP response headers cancels all the work above.

How to manage duplicates between a PDF and an HTML page

Many sites publish the same content twice: an article and its PDF version. Google recommends serving a single copy when possible. Otherwise, choose the version that should rank and tell Google with one of these two methods.

Option 1: point the PDF to the HTML page. A PDF cannot contain a link element, so the canonical tag goes in the HTTP header of the file. Google's documentation on consolidating duplicate URLs describes this rel="canonical" HTTP header for non-HTML documents such as PDFs. The header looks like this, with an absolute URL: Link: <https://www.example.com/guide/>; rel="canonical". Rankings and links then consolidate on the HTML page.

Option 2: keep the PDF out of the index. For order forms, internal documents or old versions, send an X-Robots-Tag: noindex header. Google's page on the robots meta tag and X-Robots-Tag gives the server rule; on Apache, for every PDF of the site: <Files ~ "\.pdf$"> Header set X-Robots-Tag "noindex" </Files>. Files already indexed drop out over time; the Removals tool in Search Console hides them faster. See our definition of noindex for the difference with robots.txt.

Gated PDFs (behind a form) are a separate case: Google cannot see them at all. If the content should attract traffic, publish an open summary page and keep the full file behind the form, as described in our guide to gated content.

How do ChatGPT, Gemini and Perplexity read PDFs?

AI assistants that browse the web can open a PDF, but they work on extracted text split into passages. The same weaknesses that bother Google bother them more: a scan without OCR gives nothing, a two-column layout can mix sentences, headers and footers repeated on every page break paragraphs, and a 60-page file without headings is hard to quote precisely. AI crawlers also fetch HTML pages more often than downloadable files.

To be cited, the logic of optimizing content for AI answers applies: put the key content in an HTML page with a direct answer at the top, question-style headings, figures with their source and a table when things compare, then offer the PDF as a download. To see what an AI tool extracts from a page, convert it with our webpage to Markdown tool: if the text comes out clean and in the right order, a language model can quote it.

How to track the performance of your PDFs

A PDF cannot run an analytics tag, so combine three sources:

• Search Console: in the Performance report, filter pages containing ".pdf" to see clicks, impressions, positions and the queries that bring traffic. Our Search Console definition explains the reports.
• GA4: enhanced measurement records a file_download event when a visitor clicks a link to a PDF on your site (it does not see visits that land on the PDF directly from Google).
• Server logs: every request for the file is logged, including those from Googlebot and AI crawlers.

With Sorank, the GEO and SEO dashboard brings your Search Console data into one view, so you can spot a PDF that collects impressions and deserves an HTML page.

When an analysis shows that a topic covered only in a PDF has search demand, the SEO article generator turns the subject into a web page optimized for Google and for AI answers, published on your site in autopilot. Plans are on the pricing page, with a 3-day free trial.

The essentials of PDF SEO

Google indexes and ranks PDFs that contain real text, with a filled Title, a clear file name, tags, links and a reasonable size. That covers documents people download. For content that must bring traffic, leads and AI citations, the HTML page should be the version that ranks, with the PDF as a download linked to it through a canonical header. Start by listing your PDFs that already get impressions in Search Console: they are your quickest wins.