How does a search engine bot navigate past a .PDF link?
-
We have a large number of product pages that contain links to a .pdf of the technical specs for that product. These are all set up to open in a new window when the end user clicks.
If these pages are being crawled, and a bot follows the link for the .pdf, is there any way for that bot to continue to crawl the site, or does it get stuck on that dangling page because it doesn't contain any links back to the site (it's a .pdf) and the "back" button doesn't work because the page opened in a new window?
If this situation effectively stops the bot in its tracks and it can't crawl any further, what's the best way to fix this?
1. Add a rel="nofollow" attribute
2. Don't open the link in a new window so the back button remains finctional
3. Both 1 and 2
or
4. Create specs on the page instead of relying on a .pdf
Here's an example page: http://www.ccisolutions.com/StoreFront/product/mackie-cfx12-mkii-compact-mixer - The technical spec .pdf is located under the "Downloads" tab [the content is all on one page in the source code - the tabs are just a design element]
Thoughts and suggestions would be greatly appreciated.
Dana
-
Thanks very much Christopher. This is an excellent explanation. What do you think of Charlie and EGOL's suggestions regarding making sure that there are links embedded in these PDFs pointing either back to the product page or even to the home page?
In your opinion, is this something worth doing? If so, why?
-
Hi Dana,
" ... you are right, one of the fundamental questions I still have is how does a bot behave when it finds an orphaned page like one of these? Does it just revert back to the sitemap and move one? Does it automatically go back to the last non-dead end page and move on from there? What does it do?"
Bots are not really like a single spider that has to crawl around the web that can get trapped when entering an orphaned page with no back-button. When a bot enters a site, it creates a list of all the internal pages that are linked from the home page. Then it visits each page on that list and keeps adding more linked pages to that list. Each time it adds more pages to the list, it only adds new unique pages and does not add duplicates. It also keeps track of which pages it has already visited. When all the pages have been visited once, and no new pages are discovered that are not already on the list, all of the pages have been crawled.
Best,
Christopher -
Hi Don,
Thanks so much for responding and while the answers I have received so far did give me some direction, you are right, one of the fundamental questions I still have is how does a bot behave when it finds an orphaned page like one of these? Does it just revert back to the sitemap and move one? Does it automatically go back to the last non-dead end page and move on from there? What does it do?
Thanks for chiming in. I'd love it if someone more familiar with how a bot actually crawls links like this on a page would jump in with an answer.
Dana
-
Thanks Charlie. I think this is a good suggestion. I work 9-6 too, and just so happen to be the in-house SEO strategist, so this stuff is what I'm there to do. I don't mind the mundane aspects of SEO because the payoff is usually pretty rewarding! Now I know what I'm doing on Monday (on top of a dozen other things!)
Thanks again!
-
I would spend the time needed to do an assessment of these pages.
** how many of them have external links
** how many of them pull traffic from search or other sites
** how many of them are currently useful (are people looking at them)
I would delete (and redirect the URL) of any page that answers "no" to the three items above. These are "dead weight" on your site.
Also, if these are .pdfs of print ads then they might simply be images in a pdf. (test this by searching for an exact phrase from one of them in quotes and include site:yourdomain.com in the query. Keep in mind that google can read the text in some images embedded in pdfs.
I had a lot of pdfs with images on one of my site and got hit with a panda problem. I think that Google thought that the .pdfs were thin content. So I used rel=canonical to assign them to the most relevant page using .htaccess. The panda problem was solved after a couple of months.
Also, keep in mind that .pdfs can be used for conversions. You can embed "add to cart" buttons and links into them and they will function just as on a web page.
If any of these pdfs are pulling in tons I traffic I would figure out how I can put the pdf to better use or create webpage (and redirect the pdf) to best monetize/convert or whatever you business goals dictate.
-
Can a bot navigate via a back button?
I don't think so. They can follow links but they can't "click".
-
Hi Dana
I think your question has been dodged a tad. I ways lead to understand that a .pdf or any page that opens in a new tab and does not link back to the original site, (dangling page) is not a problem. The reason being is that crawlers don't really care how a page is opened. Because the crawler will fork at every link and crawls each new page/link from each fork, when it finds a orphan or dangling page it just stops. This of course is not an issue since if the crawler has forked at each link.
So the question is how a SE treats .pdf's rather how does it treat orphan page. Maybe somebody who works with crawlers can confirm or educate us both on they work.
Don
-
Many thanks to both you and EGOL for excellent answers!
-
Thanks EGOL. Yes, many of these .pdfs could be and are referenced by other sites. Given that there's no link from the .pdf back to our site, we really are missing out on a huge opportunity. I thought this might be the case as I pondered the whole concept of "dangling links" that was discussed in a SEOMoz blog post this week.
I agree about the last point regarding opening in a new window being more of a usability issue than a problem for SEO. I agree with you completely that opening in the same window is way better for the end user.
Can a bot navigate via a back button?
Thanks very much to both you and Charlie for your excellent answers!
-
lol, thank heaven's they aren't spammy. However, they aren't particularly helfpul either. You see, about 3,000 of them are old .pdf versions of print advertising campaigns, going back as far as 2005. They contain obsolete pricing, products, etc. Unfortunately, instead of archiving them off the server, they've been continuously archived in a sub-directory of our main website.
Nearly all of it is indexed. It seems to me the best thing to do for these is to include a statement that the content is an old advertisement and include a llnk to our current "special offers" page.
What do you think of that as a strategy for at least giving engines and humans a means to navigate to someplace current on the site?
-
I see 6000 pdfs as an amazing opportunity. Get links on those pages and it will funnel a lot of power through your site.
If that was my site, we would be on that job immediately. Could be a huge gain for some easy work.
-
Go back and rework our .pdfs so they at least contain a link back to the homepage?
Yes! Absolutely! And, link them to other relevant pages. If these are reference documents they could be pulling in a lot of links and traffic from other sites.
In addition toAs well as configure the hyperlinks so they open in the same window instead of a new one?
In my opinion, this is not an SEO issue. This is a usability issue. I would have them open in the same window so the back button is available.
-
Thank you Charlie. In our case, our .pdfs contain no links in them at all. There is nothing for a bot (or a human) that will navigate them out of the .pdf....not even the back button.
Considering that, and EGOL's response below, would the best course of action be to include, at the very least, an active link back to our homepage from all of our .pdf files?
We have as many as 6,000 .pdfs.
Thanks,
Dana
-
Thanks EGOL,
Yes, I understand well that .pdf documents can be indexed. That's not my concern. My concern is that a bot that navigates to one of our many .pdf tech specs documents, which, incidentally, contains no outbound links to anything, would then become trapped and not be able to continue crawling the site. This is particularly true because we have them set up to open in a new window. In the example above, sure, there's a text reference back to the site "www.kingdom.com" - but it isn't a link in the .pdf. There are no links, in any of our .pdfs.
So, what is the best way to deal with this? Go back and rework our .pdfs so they at least contain a link back to the homepage? In addition toAs well as configure the hyperlinks so they open in the same window instead of a new one?
-
.pdf documents are crawled by bots and they accumulate pagerank just like .html pages.
You can include links in them to other documents on the web and bots will crawl those links and pagerank will flow through them.
.pdf documents can be given a "title tag" equivalent by opening their properties and giving the document a title. This title will display in the SERPs. .pfd documents can be hard to beat in the SERPs if they are optimized and have links from a competitive number of other web documents.
Lots of document formats behave this way. Excel, PowerPoint, Word for example.
In my opinion, .pdf documents can trigger a Panda problem for your site if you have a lot of them with trivial or duplicate content (as in print versions of web documents). They can be given rel=canonical through .htaccess to solve the Panda problem but Google often takes a long long time (sometimes months) to recognize the canonical and use that instruction.
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
GWT shows 38 external links from 8 domains to this PDF - But it shows no links and no authority in OSE
Hi All, I found one other discussion about the subject of PDFs and passing of PageRank here: http://moz.com/community/q/will-a-pdf-pass-pagerank But this thread didn't answer my question so am posting it here. This PDF: http://www.ccisolutions.com/jsp/pdf/YAM-EMX_SERIES.PDF is reported by GWT to have 38 links coming from 8 unique domains. I checked the domains and some of them are high-quality relevant sites. Here's the list: Domains and Number of Links
Technical SEO | | danatanseo
prodiscjockeyequipment.com 9
decaturilmetalbuildings.com 9
timberlinesteelbuildings.com 6
jaymixer.com 4
panelsteelbuilding.com 4
steelbuildingsguide.net 3
freedocumentsearch.com 2
freedocument.net 1 However, when I plug the URL for this PDF into OSE, it reports no links and a Page Authority if only "1". This is not a new page. This is a really old page. In addition to that, when I check the PageRank of this URL, the PageRank is "nil" - not even "0" - I'm currently working on adding links back to our main site from within our PDFs, but I'm not sure how worthwhile this is if the PDFs aren't being allocated any authority from the pages already linking to them. Thoughts? Comments? Suggestions? Thanks all!0 -
Link Anchor Text
When we have run a Open Site Explorer analysis on our own site, it says that for all our internal links the Link Anchor Text is 'Help with logging in' I am a bit confused as to why it shows that. That text does appear in the header of the page, but is not the first piece of text. Why is it happening on our site?
Technical SEO | | MattAshby
Why do I not see this on other sites?
What affect does this have on our ranking?
What's the best fix? Example page that we ran on Open Site Explorer: www.rightboat.com/search?manufacturer=Beneteau&model=Antares+9.800 -
Bing search results
Hi I am getting little or no traffic via Bing. Why would this be? all my traffic is from Google and social media. my site is www.cocoonfxmedia.co.uk
Technical SEO | | Cocoonfxmedia0 -
Do search engines treat 307 redirects differently from 302 redirects?
We will need to send our users to an alternate version of our homepage for a few hours for a certain event. The SEO task at hand is to minimize the chance of the special homepage getting crawled and cached in the search engines in place of our normal homepage. (This has happened in the past so the concern is not imaginary.) Among other options, 302 and 307 redirects are being discussed. IE, redirecting www.domain.com to www.domain.com/specialpage. Having used 302s and 301s in the past, I am well aware of how search engines treat them. A 302 effectively says "Hey, Google! Please get rid of the old content on www.domain.com and replace it with the content on /specialpage!" Which is exactly what we don't want. My question is: do the search engines handle 307s any differently? I am hearing that the 307 does NOT result in the content of the second page being cached with the first URL. But I don't see that in the definition below (from w3.org). Then again, why differentiate it from the 302? 307 Temporary Redirect The requested resource resides temporarily under a different URI. Since the redirection MAY be altered on occasion, the client SHOULD continue to use the Request-URI for future requests. This response is only cacheable if indicated by a Cache-Control or Expires header field. The temporary URI SHOULD be given by the Location field in the response. Unless the request method was HEAD, the entity of the response SHOULD contain a short hypertext note with a hyperlink to the new URI(s) , since many pre-HTTP/1.1 user agents do not understand the 307 status. Therefore, the note SHOULD contain the information necessary for a user to repeat the original request on the new URI. If the 307 status code is received in response to a request other than GET or HEAD, the user agent MUST NOT automatically redirect the request unless it can be confirmed by the user, since this might change the conditions under which the request was issued.
Technical SEO | | CarsProduction0 -
Affiliate links
Is there a best practice for linking out to affiliates URLs post panda? I know some believe it can be a factor.
Technical SEO | | PeterM220 -
4XX Broken Links
I am attempting to fix the issues SEOmoz found when crawling my site. I have a list of 4XX errors that I am attempting to fix. Basically I know one option is to redirect them to another page, but I would like to have the option to remove the links completely. The only problem is I can not find where the links are located. Does SEOmoz provide where on my site these broken links are? Or do they only provide the url that is linked to?
Technical SEO | | ClaytonKendall0 -
Are multiple links devalued on the same domain?
I'm in negotiations to get links placed on a popular blog with good stats. I'm allowed to pick older posts on the site, and I get to pick the anchor text. Is it best practice to diversify the links by having different keywords pointing to different pages or am I better off pointing as many links as I can at one page (varying anchor text)? Also, is it best to pick a more recent blog post, or is it ok to pick one from say, 2009?
Technical SEO | | MichaelWeisbaum0 -
Amazon s3 links
Hello, Dose anyone has experience with Amazon s3 ? Hosting images, site elements and static files (pdf etc) on Amazon S3 bucket can influence the link structure / juice ? (by having external links to the amazon s3 bucket / account). If you host all the site images and site resources with Amazon S3 will all those "links" count as outgoing external links ? Is it worst then having everything linking internally ? Dose anyone knows if that can hurt you ? Personally I was not able to test it and I am sure I can't test it in the near future. Hope someone has and I can skip some steps with this. Thanks !
Technical SEO | | eyepaq1