Manufacturing client. Every product has a spec sheet PDF, plus manuals, safety sheets, old catalogues. Around 2,000 PDFs total.
GSC shows most of them indexed. For a lot of part number searches the PDF ranks instead of the product page, which means people land on a file with no navigation, no price, no contact button.
What I’m considering:
– X-Robots-Tag noindex on spec sheets and manuals via server config
– canonical header on each PDF pointing to its product page
– keep old catalogues indexed since they get some traffic for discontinued parts
I’ve read that the canonical header on PDFs works, but I’ve also read Google often ignores it. Has anyone used it and seen the HTML page take over the ranking? Or is noindex the only reliable route?
.
