Lot's of good advice here, so I'll just weigh in with my two cents...

Instead of redirecting all your files to /index.php, why not rewrite those in .htaccess to redirect back to the original (without the /index.php)

This has the dual effect of preserving your link equity to those original urls, and there's a slight correlation between shorter URLs and higher rankings (in part possibly because shorter URLs have a higher click-thru rate)

Regardless, I suspect a perfect storm of factors contributed to your rankings, as you stated yourself:

1. Site was down when on old US host for a minimum of 3 hours one day and perhaps longer

2. Changed from US host to host based in Spain

3. Analytics stopped recording data for 3 days and site was down briefly after the change of host

4. All original URLs now have /index.php at the end

I purposely left out a Google Algorithm change, because of Occam's Razor - the simplest explanation is usually correct, and an algorythm change would be too much coincidence.

As Robert said, make sure you're targeting the right country in Google Webmaster. Other than that, I would try very hard to return all URLs, hosting and settings back to their original state before all these changes.

There's little or no evidence that changing the IP by itself will effect on your rankings or traffic, unless you move to a really slow server or a blacklisted neighborhood - but it doesn't seem like something you need to worry about.

Hopefully the speed improvements will be a big boost to your visitors, which may indirectly effect your SEO in a positive way!

Hi Rhys,

Taking a look at your site, your links all seemed natural and within reason. ( I did think the homepage was a little light on content - mostly just navigation and quick links to products. But that's another conversation.

"Best practice" is to consolidate your non-important links into a format that makes sense and is human friendly. Rand wrote a post about footer links awhile ago that still works today:

http://www.seomoz.org/blog/footer-link-optimization-for-search-engines-user-experience

I wrote about this in another Q&A thread a short time back.

Today, you don't hear much about PageRank sculpting. Most SEOs don't bother with it, partly because of it's decreased effectiveness, but also in part because there are more effective ways of controlling the influence of links.

...Link "equity" or PageRank, (or MozRank), is only one small factor in the overall value of a link. Anchor text, position on the page, and a host of other factors all influence how much influence any given link can wield. Here's a good introduction on the subject (again from Rand)

If you "no-follow" your important contact pages (about us, etc) Google may have trouble finding and crawling those pages. Because these are both valuable pieces of content and trust signals for your site, this probably isn't the outcome you want.

To summarize: Adding nofollow in your case doesn't make sense. It really only makes sense in a very few cases, and isn't as effective in controlling ranking signals as many people would like to believe.

Hope this helps. Best of luck with your SEO!

It's likely that if you are getting a "0" that indicates no data available.

Mozscape will only report the status code for pages it has actually crawled in the last index. If it hasn't crawled the page recently, it doesn't report the status code. (it still has link metrics for that URL because it can see and calculate all the incoming links to that page without crawling it)

If these are backlinks to a particular site, you should be able to get the same data using Open Site Explorer, and running a custom report to export the large amount of data.

Another option, if you actually want to crawl all of those links and return a status code is Screaming Frog. 

The Frog has both a paid and free version, but it's one of the most useful paid tools I have ever used.Hope this helps! Cheers.

Howdy,

Depends on your definition of self-referencing links, as there are several degrees.

1. Internal links, pointing from one page to another
2. On page anchor links, simply pointing to another part of the same page, usually for navigation purposes
3. Truly self-referencing, where a link simply points to it's own page in an attempt to trick a search engine.

Normal internal links (1) are an important part of any website and give search engines important clues about relevance, site structure, and how to flow signals like PageRank. Internal links are very important for SEO, but also the most easy to abuse, so you can get yourself in trouble if you overdo it with aggressive over linking or over-optimized anchor text. And keep in mind that while internal links are important, they don't pack near the punch as authoritative external links.

Internal anchors (2) such as http://example.com#football are also important for navigation, and Google does use them often for clues about content and structure. You'll often see internal anchors show up in search results, especially for sites like Wikipedia.

Finally, I don't see any value in the truly self-referencing link (3). A page that points to itself seems like a mistake, and using it to game a search engine is most likely a bad idea.

As for authoritative sources, you could look at the 2011 Ranking Factors, which shows that the # of internal links on the page is fairly correlated with high rankings. (and notice the anchor in that link, as well

Hope this helps! Best of luck with your SEO!

Hi there,

You've got the right idea, but let me suggest another tactic.

It's true that search functions can generate 1000's of urls that all tend to look like one another. Google suggests that you keep search result pages non-indexed, as these pages offer very little value and create tons of duplicate content.

http://www.seomoz.org/learn-seo/duplicate-content

Here's one way to handle your situation:

1. Put a meta "noindex,follow" tag in your search pages header, like this:

This tells search engines not to index the page, but allows them to follow the links on the page and flow link juice.

2. Hopefully you have a good site architecture and ways for search engines to discover your content. After step one, you can put a directive in your robots.txt file to block that directory from being crawled.

Something like:

User-agent: *
Disallow: /search/

Which blocks anything in the search directory.

3. Find out if search engines have already indexed a lot of your search pages by performing a site: search in Google, like so:

site:yourdomain.com/search

If you find pages in Google's index that shouldn't be there, you can use Google Webmasters URL removal tool to take these out of the index. You can remove the entire search directory with a single request.

http://support.google.com/webmasters/bin/answer.py?hl=en&answer=1663427

This is a powerful and sometimes dangerous tool, so be careful!

4. Finally, if you'd like to add "nofollow" to your search results pages, this should be fine, but only after you've completed the steps above.

Keep in mind, this is only one possible solution. If you have significant link juice flowing through your search results, this strategy may not be the best. But in general, you want to keep search results out of Google's index, so I'm comfortable recommending this strategy for 90% of all cases.

Hope this helps! Best of luck with your SEO.

Hi Todd,

Oh no, looks like all your canonicals are pointing towards the homepage....

<link rel="<a class="attribute-value">canonical</a>" href="[http://www.cherrycreekspine.com<](view-source:http://www.cherrycreekspine.com%3C/)"/>

Also looks like there's an extra carrot "<" at the end of the URL. Looks like it's coming from the wonky code.

Regardless, this isn't what you want. This basically tells search engines that every page is a canonical of the homepage - and that all these other pages aren't important on their own. It's likely search engines will start to drop these pages out of their index unless this tag is removed immediately. Reminds me of Dr. Pete's canonical nightmares.

In short... remove that canonical ASAP. It's probably better to have some duplicate content than a sitewide canonical that points to your homepage.

Have you tried the Joomla Canonicalisation Plugin? Haven't tried it myself, but it might be smidgen easier than trying to code the php yourself. You can find it here: http://extensions.joomla.org/extensions/site-management/seo-a-metadata/url-canonicalization-/5355

My guess is you can completely remove whatever manual canonical code you wrote, and the plugin will take care of the rest.

Remember, when the code is working properly, each page with point to itself (without extra parameters and so on) the way a proper canonical should.

Best of luck!

Hi Mikkel,

Like Chris, I spidered your site and couldn't find any links to /index.php files, which probably indicates one of two things:

1. You've fixed the problem - Yay!
2. Or Google is finding those links from external sources
3. Google found those links at one time in the past, and is still trying to crawl them.

In the Crawl Errors report in Google Webmaster Tools, if you click on the link of each 404, there's often a "linked from" source where you can see where Google discovered the broken link. This is really helpful in rooting out the cause.

Regardless, I'm going to go with #1 and optimistically believe that you were able to fix the problem.