PDF files behind corporate firewall

How to get G to find them?

[adfree]

Create additional "outside-firewall" depository and let G crawl it (how to invite?)?

How can we still attract new sign-up's for the site then?

Thanks for any hint, Jens

[mcavic]

Create additional "outside-firewall" depository

Yep, put them up on a web server that's accessible, then create a nice human-readable index that links to all the files. All you probably need is one inbound link to get Googlebot to visit.

new sign-up's for the site

Not sure what you mean.

[adfree]

Thanks mcavic, will propose that.

Signup's: all PDF's reflect deep technical expertise about special procedures within specialty chemicals. Obviously they are worth a lot of tradition, history, R&D etc.

If given away for free (which we do today via DMS or static links from our web site) we wanted at least some customer retention and have them sign-up to our site as a member (gaining additional access, benefits etc.).

In case of the discussed PDF depository solution how could we attach a signup procedure to the event of downloading a doc?

Thanks again, Jens

[TallTroll]

Don't publish full docs on the public side, just precis versions, and/or snippets of the full text. That's enough for someone to judge whether they want to go as far as a full sign up, AND is lots of extra content. Those new PDFs can contain links to all sorts of useful places...

[mcavic]

snippets of the full text

True true.

[nileshkurhade]

How about cloaking a bit like this :
if http_referer="googlebot"
then => let the pages be seen
else
goto => members registration

[takagi]

How about cloaking a bit like this :
if http_referer="googlebot"
then => let the pages be seen
else
goto => members registration

If you let Googlebot access these files, then Google will create HTML versions of the text in the PDF for the SERPs. Not a good idea unless there is some graphic information needed from the PDF files.

[adfree]

We'd prefer have the docs untouched, unrenedered. A pure backdoor access via linkage to all files might do the job just fine.

If we were to password-protect the files and combine the login with subscription info, could G still spider them?

[mcavic]

If we were to password-protect the files

Googlebot wouldn't know the password, and thus wouldn't be able to read the files. But, you could protect the PDFs, and let Google just spider your index of them (with descriptions or snippets).

In your index, then, all of the links would trigger a login/signup page, after which the user would be redirected to the actual document.

[BigDave]

How about cloaking a bit like this :
if http_referer="googlebot"
then => let the pages be seen
else
goto => members registration

I don't know of anyonw that will register on pages when they find those pages from the SERPs. They just hit the back button.

And if I am really annoyed by it, I will fill out the "Help us improve" link on google pointing that site out.

This is an example of cloaking that damages google's SERPs. They do not want this. The users do not want this. Don't do this.

[adfree]

Thanks bunches all. I think we will go with the clean, homemade index version offering snippets or abstracts as SE food and hit the login screen when opening the docs.

This should serve the purpose fine and would not upset anyone, we are global market leader in our industry segment (1.7b revenue), our clients and othr stakeholers will know us anyway and be glad to leave their quick registration data.

Your suggestions made sense one again and closed a couple of doors of suspicious thoughts, we'll do just fine with your help again, many thanks, Jens