SEO and PDFs

[MayaLocke]

I read something that there are tools that convert other formats to pdfs but they simply save the whole thing as an image, if you use an application that does that search engines will have a really hard time indexing the content.

[deepsand]

I've noticed here a tendency by many to conflate the roles of a crawler/robot/spider with that of an indexing engine.

The former merely retrieves data; it makes no decisions regarding the use of such, beyond that of seeking within that data evidence of additional resources to be retrieved.

That data is then stored & later analyzed by the indexing engine.

While this may to some seem to be a small & inconsequential distinction, it is in fact one with an important difference, in that there are different directives for controlling the behavior of each, and thus the end results that publicly listed.

Here, for example, the true issue is, not how crawlers/robots/spiders handle PDF files - they're just another type of data set to be retrieved - but rather how the indexing engines handle such.

[deepsand]

Originally Posted by MayaLocke

I read something that there are tools that convert other formats to pdfs but they simply save the whole thing as an image,

Then the result would not be a true PDF file, regardless of the file extension!

[kgun]

I don't know how advanced 2009's SE Bots are, but it may be more difficult to collect and rank data in a PDF document.

Assumption: Data is not converted to any other format.

[deepsand]

Presumably all indexing engines should be sufficiently robust so as to be able to accurately read any number of widely used standard file formats, including PDFs.

[kgun]

In an ideal world, I agree that native SE's should.

[weegillis]

Premise: PDF format is viable for SEM

I'm only speculating here, but it would seem that it is viable, provided there is some HTML reference that points to the document. There is little doubt in the ability of SEs to spider and index PDF content (embedded images I'm not sure of) so why wouldn't it be a viable content wrapper for targeted keywords?

The essential aspects of optimization would be typical of those for any HTML document. One would need to pay close attention to size and load times, though, owing that most PDFs can run into dozens of kilobytes per page in the document. Links can bloat the file, too, so they would need to be kept to a minimum.

Would the real test be a pure PDF site? A simple redirect could have the PDF load as the home page and link to other PDFs instead of html. I know it makes no sense, but it would just be an experiment, right?

[The Leader]

I notice alot of PDF files come up high in searchs so if Content is King then you would have to believe that google can read them!

[texxs]

Originally Posted by kgun

Agree. Se my above example.


IMO, Because it is portable, platform independent. That was the main reason, at least as far as I understand, that Adobe made Acrobat reader and the professional version that can make PDF documents.

I create them very fast in MS Word. When the Word document is finished, I click the PDF button and a document of 50 pages is created in a few minutes.


My bolding. Another example of WWW rumor or fact?

1- Won't MS Word , WordPerfect and OpenOffice run on any computer as well? Or MS shunning linux and apple again? It took them a while but they did come out w/ Odffice for the mac. as far as linux fgoes, that's the big drawback with linux, no one write software for them except server software mostly... Of course if the doc doesn't need locked then there the ultimate cross platform doc: HTML!
Cross Platform compatibility would be a great reason . . .

2- Check out Google desktop TOS, it's scary! and that's just the stuff they admit too. Seriously, they store a list of every document, movie, file, program on your computer, will give this list ot anyone they want and worse . . .

I can't believe it's even legal.
Us Americans are SO easy to take advantage of . . .

[texxs]

Originally Posted by weegillis

Premise: PDF format is viable for SEM

. . . so why wouldn't it be a viable content wrapper for targeted keywords?

Because of the useability issues, you know, it's a dead end navigationally (did I just invent a word?) and pdf's crash so many browsers, or at least "freeze them up" for an extended period of time. On the best computers they still break the 4 second rule"

Never, break the 4 second rule . . .

[Dubbya]

You can avoid the temporary browser hang or prevent the browser from crashing by implementing a file download script instead of simply using a link to the file.

In PHP:
Make link prompt visitor to download .PDF, .DOC, or other file.

In ASP:
ASP 101 - ASP.NET version of "Download" Sample

.02

[texxs]

I tried this once. It does help.

But . . .

I started getting support requests saying they couldn't find the file, etc . . .

Also the disc based adobe reader causes lots of problems with computers too, not as many as the browser version though...

[kronikmedia]

Search engines have evolved with time. Today search engines like Google have no problems what so ever in reading PDF files. In fact I have received good ranking for many of my PDF files. Strangely enough I have never even optimised the PDF's specifically. I have focused on general site wide search engine optimisation for my company's website and the pdf documents on the website have been indexed automatically.

[kgun]

Originally Posted by texxs

1- Won't MS Word , WordPerfect and OpenOffice run on any computer as well?

Yes they do, but as far as I understand, a Portable Document Format (PDF) document made on a Pc can automatically be viewed on a Mac.

Now a word document made in 2009 on a Mac can be opened and read on a Pc. It was not that easy when the first versions of Acrobat Reader were introduced back in the Stone age.

[weegillis]

Originally Posted by Dubbya

...implementing a file download script...

Thanks for that, Dubbya.

[Peter]

PDF files can be viewed on any OS with a PDF viewer. You dont even need to use acrobat if you dont like it.

Where a lot of people get "freezes" in Acrobat is that the latest versions of the Adobe reader have become bloatware and package all sorts of crap you havent asked for.

If on a corporate it may be worthwhile to install the development kit and strip all the add-ons, including the awful auto-update routine. There are new features that are only compatible with the latest version. Acrobat has ballooned from 4MB to 41Mb, on the back of pointless trash like auto-download resume.

Therefore using the basic features, optimising images and text, all the usual tricks for making a good webpage, PDF's do work.

They can be printed in multiple sizes, and orientations to suit worldwide printing standards.

So there are significant advantages in PDF's for producing non-corruptable, well designed documents, and you dont have to go anywhere near Adobe to create them.

Whereas for SEO, the HTML pages give many more options of promotion, for a catalogue or library, PDF's can produce far better marketing print materials.

I really dont see why there is such strong objection to the format when used properly.

[Clint1]

There's no question that G can read the text in a PDF file. If you do a search and a PDF file shows up, then click on a "View as HTML" link next to the G result PDF file, the search words are highlighted (as long as they don't pull that "search terms appear only in links pointing to the page" crap and the text is actually in the file), and that is, IF it's actually text in the PDF file and it's not an image as someone pointed out. Also, it would not be likely that PDF files would ever show in the SERPS if G couldn't parse and recognize text in them.

There are .pdf PDF files that have no text in them. Technically these still are PDF files, due to the extension and since Adobe (or similar reader) opens them. I see them all the time in manufacturer datasheets and brochures; they "look like" text, but the entire data is some kind of "scan" of text for lack of a better term. I don't know if it's actually some kind of scan, because the quality & clarity can be outstanding......I guess it still could be if it's a very high quality scan, and at high dpi, then reduced in size.(?) Also, a PDF file doesn't always have to be text, it can be just graphics (line graphs, charts, etc), and they are still a PDF file.

[Clint1]

Originally Posted by kronikmedia

Search engines have evolved with time. Today search engines like Google have no problems what so ever in reading PDF files. In fact I have received good ranking for many of my PDF files. Strangely enough I have never even optimised the PDF's specifically. I have focused on general site wide search engine optimisation for my company's website and the pdf documents on the website have been indexed automatically.

Speaking of optimization, I always do that. I use something called PDF Info Changer where I can take any existing PDF file, change it to the appropriate title, description, subject, etc., which helps loads in the SERP's. You may be surprised at how many PDF files have totally irrelevant useless titles.

I also use the PDF editor part of Acrobat Exchange to add links in the PDF file, before I upload it. I'll create links from existing appropriate potential link text and have those links go to appropriate pages on my site, and also add new text to the bottom of the files like "Back to [whatever blah blah descriptive link text here]" for both the products' category page, and to my homepage.

[weegillis]

A little background on PostScript, the descriptive language precursor to PDF:

Link: PostScript - Wikipedia, the free encyclopedia

Which will lead to (among many other topics),

Link: Portable Document Format

addendum:

Now that PDF is an open standard, several software manufacturers have jumped on board, some with significant contributions (though I doubt much innovation, yet), including the well touted, Foxit:

Link: Foxit Software

[deepsand]

Originally Posted by Clint1

There's no question that G can read the text in a PDF file. If you do a search and a PDF file shows up, then click on a "View as HTML" link next to the G result PDF file, the search words are highlighted (as long as they don't pull that "search terms appear only in links pointing to the page" crap and the text is actually in the file), and that is, IF it's actually text in the PDF file and it's not an image as someone pointed out. Also, it would not be likely that PDF files would ever show in the SERPS if G couldn't parse and recognize text in them.

There are .pdf PDF files that have no text in them. Technically these still are PDF files, due to the extension and since Adobe (or similar reader) opens them. I see them all the time in manufacturer datasheets and brochures; they "look like" text, but the entire data is some kind of "scan" of text for lack of a better term. I don't know if it's actually some kind of scan, because the quality & clarity can be outstanding......I guess it still could be if it's a very high quality scan, and at high dpi, then reduced in size.(?) Also, a PDF file doesn't always have to be text, it can be just graphics (line graphs, charts, etc), and they are still a PDF file.

Text can be optionally stored in a PDF file as either true text or as a graphic.

While some scanners provide for such to be under the user's control, many (most?) do not, but simply store everything as a graphic. And, of those that do provide for the choice, it's not always clear to and easily understood and used by the user, with the result that most use the default setting to store as graphics.