Concerned
-
I am a little confused, what made you lose your spot 5 ranking? Did you move your page to a .pdf? how is the .pdf relevant to your keyword? Or was one page ranking in 5 and now its gone. However, you found your .pdf in spot 68 for that phrase?
Did you change anything on that page that was ranking or on your site? Usually something causes a loss in rankings. Esp. when you go from spot 5 to nowhere to be found. Have you checked your Google webmaster tool there may be a message there.
-
Hi Gary Do you have a rel="canonical" on the .ie version of the site that points to the .co.uk pages? Basically tells the bot that this site is a direct copy and the .co.uk is the one to crawl. It may be that because it is duplicated across 2 different domains you are getting penalised for it. More about rel="canonical" here:- http://www.seomoz.org/blog/canonical-url-tag-the-most-important-advancement-in-seo-practices-since-sitemaps Also a WBF about cross domain canonical links:- http://www.seomoz.org/blog/cross-domain-canonical-the-new-301-whiteboard-friday Hope this helps
-
Hi Donnie,
Thanks for your reply.
Yes, it was ranked no.5 and now it's gone and replaced with the .pdf in pos.68 for that phrase.
I have re-written this page and pages related to this page. What I have seen in Analytics when I have made these changes, is that bounce rate has improved from 70% to 30%, Avg time on site has increase by 2 minutes and page views has also increased, so from the user experience it has worked as I imagined, with Google not.
I have checked Google webmaster tools, no messages.
-
Maybe the .pdf was always there just unnoticed?
Perhaps its something you changes on the page: Did you run an SEOmoz onsite optimization test?
What did you change on the page? Also, did you change any internal links pointing at that page? if its none of these factors, it can also be an external linking factor.
-
Yes, maybe the .pdf was always there.
All optimization tests have been done, this was all done before pages went live.
Changes were made first & foremost for the user, and from the results I gave you, this is clearly proven a success.
It was the main body of the text and structure that was changed, header tags etc all remained the same.
I checked Bing, and the URL in question is still on page 1 for the keyword.
-
I don't want to use a rel='canonical" as I want the .ie website to rank well for all keywords in Google.ie and at the moment this seems to be the case.
-
Hmm, your link juice will flow through with a canonical code. However, I don't think this is the problem, in your case. I would experiment try adding the code and see if your results are back up in a few days... If not take the code down.
-
Hi Donnie,
If I use a rel='canonical' on the .ie webpage, is this not telling Google that you do not want this page to rank?
-
Unfortunately, while you should be able to theoretically target .co.uk and .ie separately, Google can screw it up on occasion and treat them as duplicates. If you're seeing the copy bring up the .ie site on Google.co.uk, that's definitely a possibility. You could try the new hreflang approach - see this Google resource:
http://support.google.com/webmasters/bin/answer.py?hl=en&answer=189077
It's basically for regional content where the language is the same (there are other variants, but that's a big one), since Google knows they don't always get it right.
It is also possible that the .co.uk page has been penalized and other content is just being brought in to fill the spot - since the PDF is at #68, that's also possible. Have you done any recent link-building pushes to this particular page?
-
No you use robot.txt to restrict pages from pages. Rel-canonical passes link juice. However, I would also look into what Dr.Peter is suggesting.
-
This can get tricky - rel-canonical passes link juice, but it could also prevent the .ie pages from ranking. Google is a bit inconsistent with this internationally, sometimes, a non-canonical version will still rank, if it's more relevant to the country/language of the query, but I'd hate to trust that.
-
Hi Peter,
The .ie website is not shown in the google.co.uk for the target keyphrase, however what I did in google.co.uk was I pasted the first paragraph of the page which was ranked on page 1 for that target keyphrase and it's the .ie website that appears, .co.uk website is not where to be seen.
I have been doing some link building, however nothing excessive, and on authority websites, industry specific, I just don't feel it could be this so the only thing left is that this webpage has been penalized for duplicate content even though the .co.uk page has been indexed before the .ie webpage.
The strange thing is, I am still ranking really well, top 5 for about 30 or so keywords, very competitive keywords at that, so why would Google just be penalizing that specific webpage in question and not others, arrrrrrggggghhhhh, this is really getting to me.
Do you recommend that I place this code on the .ie webpage:
Pointing to the .co.uk website? -
The two sites should point at each other and use the region codes, so...
(1) The English site should have this tag:
(2) The Irish site should have this tag:
That way, whichever site Google hits, they're aware of the other site(s).
-
Ah! fantastic.
Have you tried this before? Do you recommend putting this across the whole site?
Another thing I noticed is that when I paste in a first paragraph from a co.uk webpage into Google.co.uk it's the .ie webpage that appears, however on another webpage on the .co.uk website it's the .co.uk webpage that appears in Google.co.uk, hope that makes sense? what I would say is that the page in question that is not ranked, if I paste the URL into Google.co.uk it still appears.
-
No - I'll be perfectly honest: I don't do a ton of international. The international SEOs I trust seem to think positively about the new tags, but we don't have a ton of data. The upside is that they're relatively easy to implement and they don't carry any real risk. The worst that happens is that it doesn't work.
My gut reaction is that there's regional confusion and Google is having a tough time reconciling duplicates. That's more in line with the inconsistent ranking you describe than a full-blown penalty would be.
-
Thanks Peter, you have been a great help so far.
I will make these changes and let you know how I get on.
-
Just one more question
Example
If this is on a particular product page does it have to be :
-
Argh - I'm sorry, yes. The hreflang="" code is the same, but the URL is the cross-language version of that URL. As long as the URL structure stays the same, this shouldn't be too hard, but if you use different structures, it could be a pain. I'm editing my previous reply.
-
So just to clarify, the homepage of the website would show: And product page would be:
-
Sorry Peter, my previous reply looks strange, was using the iPad, not sure what happened, anyway, what I meant to say was:
So just to clarify, the homepage of the website would show:
And product page would be:
-
1710.0k
-
02101
-
0368
-
09287
-
05195
-
04344
-
03723
-
261.3k