I personally block search engines from PDFs.
If someone clicks through from the SERPS to a PDF, they are usually on a dead end page. Having read the PDF, where do they go? If the PDF has a link within it to the site it came from then humans (and search engines) can follow that link, but that is a very poor substitute for the website menu and banner with all the other stuff you want a visitor to see and get them clicking around your site.
If you're offering a PDF as a download, then I'd make it non-spiderable and have a spiderable HTML version on your site. To my mind, PDFS and Word documents are for taking away from the web with you.
|