Switching from ranked PDF to normal page ...

[TheHecksler]

Hi folks,

I have a project which is ranking well for specific product-keywords. The problem is, that the ranking pages are PDF files of those products (datasheets). But i do have a normal (html) pages for each product as well. Each of these pages has grade a for the specific keyword and is well crawled, etc. How can I urge Google to "switch" from these PDF files to the specific html pages. I was trying this by testing it robots.txt to disallow to index the PDF file (I tested it with one for first). I would add rel=nofollow to the link to the PDF on the specifi product html page (?) ...

Any ideas?

Seb

[TheHecksler]

Thank you Samuel,

thats what I am afraid of from the beginning. The quality of the HTML Page need to be increased. My initial question should be mentione as a second "not so nice way"

Increasing the page quality to run out the PDFs is one of the top things on relaunching the site.

Thanks again

Seb.

[SamuelScott]

The PDF rel=canonical question: This Moz post gives advanced directions on HTTP headers (for using rel=canonical in non-HTML files) and cited this and this information from Google.

However, I would add one other issue now that I understand more about your situation. Your PDF files are ranking over the HTML landing pages likely because the PDF "pages" contain more information and are probably better "quality." It's a real possibility that the HTML pages will not rank as highly as the PDF pages currently do even after doing all of this.

Landing pages rarely rank high in organic search results because they are thin -- such pages are usually used to convert advertising and paid traffic. Good, informational pages are what rank highly in organic search. So, if you want HTML pages to rank like the PDF files, I'd make those HTML pages as "quality" as the PDF ones. I hope that makes sense!

[TheHecksler]

Hi Samuel,

thanx for your respond

No, its not a doubble content Problem. I just prepared landing pages for these specific product keywords - but Google finds the PDFs somehow more appropiate o.O. Yes I want the HTML pages to appear. I am wondering that Google indicates the PDF files as a very well optimized resources ...

I´ve done step 2 by using the robots.txt.

Step 3 would mean that peoples couldnt download the PDF file then - so I would not 301 redirect it.

Quick question to step 1: Do you know any way how to add a rel=canonical tag to a PDF file? Im am not that familiar with creating PDF files.

Thanx.

Seb

[SamuelScott]

The first thing to understand is that you have a duplicate content issue. If you have a PDF and HTML page of the same content, then Google will likely show only one of them in search results -- as you are probably seeing. Therefore, you need to decide which set of pages you want to appear in the SERPs. I presume that you want the HTML pages to appear.

So, I would do all of the following:

1. Add a rel=canonical tag on each PDF page that points to the corresponding HTML page so that Google knows that the HTML page is the "main" one for indexing purposes

2. Add a "no-index" tag to each PDF page (or, you can list each PDF page in your robots.txt file) so that Google knows to remove the PDF pages from the index

3. If the PDF pages serve no purpose at all (to humans or to search engines), then you may just want to 301 redirect each PDF page to the corresponding HTML one. If you want to keep the PDF pages themselves visible on your website for any reason, this can be ignored.

Good luck! Let me know if you have any questions.