PDF catalogues and datasheets in a relaunch: keep, convert or redirect
PDFs often rank, collect links and get forwarded between buyers, and relaunches routinely break them. Decide file by file whether to keep, convert or retire each one, then redirect every old PDF URL that anybody still uses.
In one sentence
A PDF migration is the part of a relaunch that inventories every PDF on the old site, decides whether each one stays a file, becomes a web page or is retired, and redirects the old file addresses accordingly.
Most manufacturer websites hold more product information in PDFs than in pages: catalogues, datasheets, test certificates, installation manuals and price lists. A relaunch usually replaces the pages carefully and forgets the PDFs, and they are often the files that rank, collect links and get forwarded between buyers.
This guide covers what to do with each PDF during a relaunch: keep it, turn it into a page, or retire it, and how to redirect the old file so nothing that buyers rely on breaks. It sits alongside the redirect map step in the website relaunch runbook.
Do PDFs matter for search and AI answers?
Yes. Google indexes PDF files and lists them among the file types it can index. A datasheet named after a grade or a standard can rank for that exact search, sometimes above the product page. Buyers also save PDFs, paste their links into emails and keep them on shared drives. An old PDF URL can keep receiving visits for years after the page that linked to it has gone.
The weakness of a PDF is everything around it. It has no site navigation, no enquiry form and usually no clear route back to the company's site. A buyer who lands on a datasheet from search reads it and leaves. The same specification on a product page puts a quote button next to it.
How do you find every PDF before the relaunch?
Do not rely on the CMS media library. PDFs get uploaded to odd folders, linked from old news posts and shared directly. Build the list from four sources:
- A crawl of the current site, filtered to files ending in
.pdf. - The server's file system or hosting file manager, searched for
*.pdfunder the web root. - Google Search Console, Performance report, filtered by page URLs containing
.pdf, over the last twelve months. This shows which files receive search clicks. - Server logs or analytics, for PDFs that receive direct visits from links in emails and bookmarks.
Record for each file: its URL, what it contains, search clicks, other visits, and whether other sites link to it. That last column comes from any backlink tool, or from the Links report in Search Console.
Should each PDF be kept, converted or retired?
Decide file by file. The decision depends on what the document is and whether anybody uses it.
| Document | Usual decision, and why |
|---|---|
| Product datasheet | Convert to a product page, and keep a PDF download on that page. The specification belongs in HTML where it can be read, quoted and linked. Engineers still want the PDF to file. |
| Full catalogue | Split into product and range pages, and keep the catalogue as a download. One long PDF answers no single question well. Pages answer one each. |
| Test certificate or approval letter | Keep as PDF, and link it from the product page it covers. The document itself is the evidence. Do not retype it. |
| Installation or operating manual | Keep as PDF, and add a short HTML summary page if it receives searches. Customers need the full document. The summary page catches the search. |
| Old price list or discontinued product sheet | Retire. Redirect it if it has visits or links, otherwise return 410. Old prices in circulation cause disputes and confuse AI answers about what you sell. |
How do you turn a datasheet into a product page that works?
Rebuild the content as text. A scanned catalogue has no text layer at all. To a search engine it is a picture, and the specification inside it might as well not exist. Even a text PDF often holds its tables as positioned fragments that lose their structure when extracted.
- Put every specification in an HTML table with units in the header row, for example "Tensile strength (MPa)". Do not use an image of the table.
- Name the standard each product meets in the text, in the form buyers type it, such as IS 2062 E250 or ASTM A36.
- Keep the product name, grade codes and part numbers identical to the PDF, so a buyer searching the code from an old document finds the new page.
- Offer the PDF as a download from the page, with a link label that says what it is: "Datasheet, PDF, 380 KB".
- Do not put datasheets behind a form. A buyer who has to give his phone number to read a specification will often read a competitor's instead.
The guide to launching a new product line covers what else a product page should carry.
How should old PDF URLs be redirected?
Treat every PDF with visits or links as a row in the redirect map, like a page. The new destination depends on the decision you made:
- Kept at a new address: a permanent 301 from the old PDF URL to the new PDF URL.
- Converted to a page: a 301 from the old PDF to the new product page. A buyer following an old link then lands where he can enquire.
- Retired with no replacement: return 410 Gone, or 404. Do not redirect to the homepage. Google's site move guidance warns that redirecting many old URLs to one irrelevant page, such as the home page, may be treated as a soft 404, and the buyer learns nothing either.
On Apache, a single PDF redirect in .htaccess looks like this:
Redirect 301 /images/pdf/ERC-catalogue-2019.pdf https://www.example.com/products/elastic-rail-clips/
Test the PDF rows with the same script you use for page redirects. They are the rows most often left out.
What if the PDF and the page say the same thing?
When you keep a PDF download that duplicates a product page, tell search engines which version to show. A PDF cannot carry a canonical tag in its own content, but Google accepts a rel="canonical" HTTP header for non-HTML files such as PDFs. On Apache with mod_headers enabled:
<Files "elastic-rail-clip-datasheet.pdf">
Header set Link "<https://www.example.com/products/elastic-rail-clips/>; rel=\"canonical\""
</Files>
To keep a PDF out of search results altogether, an internal price list for instance, send an X-Robots-Tag: noindex header, which Google documents in its robots meta tag guide. Do not block the file in robots.txt as well, or Google cannot fetch it to see the noindex.
How should PDF files be named on the new site?
Give each file a stable, descriptive name and keep it. elastic-rail-clip-datasheet.pdf can be updated in place for years. ERC_final_v3_new(2).pdf gets replaced by v4, and every link to v3 breaks. If you must keep revisions, put the revision date inside the document and on the page, not in the file name.
Set the title in each PDF's document properties too. Some search results and browser tabs show it, and "Microsoft Word - Document1" does not inspire confidence in a specification.
What should you check after switchover?
In the first week, request every old PDF URL from your list and confirm each one returns the redirect or status you planned. Watch Search Console's Page indexing report for PDFs reported as not found. In analytics, record PDF downloads as an event so you can see which documents buyers still want. When a download keeps getting heavy use, the product page probably still lacks something the PDF has.
Questions people ask
Does Google index PDF catalogues and datasheets?
Google indexes PDF catalogues and datasheets, and lists PDF among the file types Google Search can index. A datasheet PDF named after a grade or a standard can rank for that exact search, sometimes above the product page.
Should an old datasheet PDF be redirected to the new product page after a relaunch?
An old datasheet PDF should be redirected to the new product page after a relaunch when the datasheet has been turned into that page and the old PDF URL still receives visits or links. A 301 redirect from the old PDF to the product page sends buyers to a page where they can also enquire.
How do you set a canonical URL on a PDF file?
A canonical URL on a PDF file is set with an HTTP Link header containing rel="canonical" and pointing at the preferred page, because a PDF cannot carry a canonical tag in its content. Google supports the rel="canonical" HTTP header for non-HTML files such as PDFs.
Should a manufacturer put datasheets behind an enquiry form?
A manufacturer should generally not put datasheets behind an enquiry form. Buyers use datasheets to shortlist suppliers, and a form in front of the specification sends many of those buyers to a competitor whose datasheet is open.
Check your own site before or after a relaunch
A plain-English self-check for manufacturers: what a buyer, Google and an AI assistant see on your site today.
What readers usually ask next
Product Launch Blog is an EbizIndia publication. This article does not pitch anything; the disclosure sits here instead, and in the footer, on every page. On this topic, EbizIndia's own page: Website transformation for manufacturers: scope and steps.