Welcome to the Q&A Forum

I wouldn't worry about it too much from an SEO perspective. You could set up filters in Google Analytics to get rid of the noise so you can see your metrics without these referrals in the mix.

If it was just an hour I don't think you have anything to worry about. In the future you may want to look into serving a 500 code instead when the site goes down.

Companies that spend more time building their brand are going to be trusted more than companies that spend most of their time on SEO. Companies that have well-known brands and high-traffic websites with hundreds of thousands + quality external links, an aged domain, lots of brand-related signals (e.g. searches for their brand) are going to be given the benefit of a doubt.

The rest of us are S.O.L. until we can build that brand presence. You have to toe the line.

Yes you can learn from the big brand sites, especially when it comes to things like upselling, cross-selling, brand building, email marketing... but having worked with a lot of big brands I can tell you that their SEOs don't typically know anymore than others. They also have to go through a maze of bureaucracy to get anything implemented. As such, I don't pay much attention to what they are doing SEO-wise. If something works for them it probably has more to do with the power of their brand and domain than it does with a particular tactic. Emulating them with regard to SEO is typically useless, at best, and often dangerous.

Adrian,

You are on the right track and are trying your best to advise this client with SEO best practices. Pat yourself on the back for that one.

I've had this similar situation a few times and instead of using an image I used either an iFrame or a script. An image might work as well. This is not spammy or blackhat or anything like that. It is quite common for site-wide text like terms of service, privacy policy, FAQ, shipping policy, etc... to be shown on every page of a site in this manner. It is especially common on eCommerce sites with regard to shipping, return policy... No matter what just make sure it is all viewable from the major mobile devices.

An easy way to test things out is to just choose a few location pages, remove the boilerplate copy (replace it using the method of your choice), write your unique content and give it a few weeks/months to see how it competes in the SERPs. For better or worse this should give you the answer.

In the long-run, however, these doorway pages you described are obviously a bad idea so they should have a plan in place on what to do if when those pages fall from the rankings.

As I said, that is what I would recommend doing. Zappos is not, and it could easily be due to limitations with their eCommerce or fulfillment systems since each color is probably a different sku. It could just as easily be due to the ability of these pages to rank better for each color, in which case they have an advantage over most other competitors because they can get away with it, as you have noticed.

I disagree and would advise against this. First of all, these are not the same pages and so it is an improper use of the rel canonical tag, which means - at best - Google could just choose to ignore this at any time and the problem would still be there.

Second, if he has a link-based penalty (looking at his link profile for that page this is very possible) placing a rel canonical tag on the two ranking pages that points to the penalized/filtered page could cause those two to stop ranking as well, leaving him with no traffic at all for that brand - a bit worse than sending users to the wrong page.

Jonathan,

I had a look a the external links to that page and out of 7 links, according to OSE,  6 of them have that exact anchor text "Bit O Honey" with the other one being "Bit O Honey for Visit here".

One of them is on a list of "great links" for an android app website that is HEAVILY spammed with really bad links, such as "female escort services" in various countries.

Another is on a list of links at the bottom of a totally irrelevant post that is also irrelevant to the blog it's on. The blog is about hotels and travel. The post is about cooking chicken with wine. The link is about "Bit O Honey" candy, and it is found also near links for window blinds and ice cream.

The other links, with the exception of one that is debatable, are all along these same lines in terms of quality. If you were to rel canonical (essentially a redirect as far as search engines are concerned) the other two pages to this one you would be putting those pages at risk too.

My advice is to get rid of these links. If you can't get rid of them, disavow them. Wait a few weeks and if you don't come back into the rankings for that page file a reinclusion request. And of course build more high-quality links to that page if you can.

Last but not least, fire your link builder.

Do you have any synonyms of those keywords on the page?

Do you have any anchor text with that keyword, or close synonyms, out there linking to the page?

Are these mobile users?

If you could share the page that might help us diagnose it as well, but one guess could be that these are mostly mobile searchers since iPhone users browse with Safari. Mobile rankings often differ from web rankings, and factoring in geographic variances it could result in most coming from a specific city.

Hello Anastas,

I agree that you should block the search folder from being indexed. I'm going to assume that nobody is linking to your search pages and that you have other paths (e.g. SEO-friendly navigation, sitemaps...) for search engines to use to access your products).

I don't understand why you have formatted the disallow statement that way, however. Unless I'm missing something (and could be since I don't know what your site is) you only need to do this:

Disallow: /product/search*

And of course after doing this you should test it in GWT to make sure that A: You are blocking the pages you want to block, such as search pages with lots of parameters, and B: You are NOT blocking other pages you don't want to block, such as product pages. Here is more info on where to find the testing tool in GWT if you don't know: http://productforums.google.com/forum/#!topic/webmasters/tbikAxJiIZ4

Let us know how it goes. Good luck.

Michael is correct as it pertains to multiple links going to different pages, but in terms of multiple links going to the same page I'm pretty sure only the first one will count. I think this is true whether the links are internal or external. See references below...

http://moz.com/blog/results-of-google-experimentation-only-the-first-anchor-text-counts
Moz Experiment Results by Rand

http://www.seo-scientist.com/first-link-counted-rebunked.html
"OK, so to me this is a pretty solid data supporting Michael’s and Rand’s claims. The fact that the SERPs reacted to my changes in the order of links back and forth is kind of hard to debunk."

http://www.addictedtoseo.com/exceptions-from-the-first-link-counts-rule.html
I like this post because, like with the SEO Scientist test above, it highlights the fact that nofollowing will not help. This is why internal pagerank sculpting with nofollow links doesn't work.

I'm not sure if it still works, but you used to be able to get around this by linking with hash tags that jump to named anchors on the same page. So www.mysite.com/url1 and www.mysite.com/url1#top would be considered two links and would both pass some pagerank.

Two people have already advised you not to use that software. Consider this the third. I seriously doubt anyone who knows much about SEO these days (as opposed to someone who stopped doing it in 1999) is going to tell you that using this software is a good idea. With that I think you have your answer.

First, I wouldn't use https unless you really need that page to be secure, such as forms, shopping carts, or any page where someone inputs personal information, or where their personal information (or others') is displayed.

Second, in cases where you really need to use https then yes, I would 301 redirect the http version to the https version of the page.

You may also consider using rel canonical tags to help search engines figure out which version is the one you want them to use.

Lastly, if you choose to go with http on some of those pages just make sure you update all of the internal links so they point to the right URL instead of relying on the redirects to do this for you.

You mentioned that this content was on your "main" pages. If that is so, I wouldn't redirect those pages to a subdomain since I am assuming they still have other, more user-friendly, content on them? Perhaps some examples would help.

However, I don't see any problem with using Confluence to put your documentation / help content onto a subdomain. I will let someone else speak to the SEO capabilities of various Confluence ad-ons though, as I have no experience there.

Just make sure the content no longer appears on your "main" pages once you move it over to the subdomain. If that was the only content on the page, then yes you should redirect it. If there was other important content on the page, just remove the documentation content and leave the rest.

I hope I have understood your question accurately.

Why don't you choose one (e.g. A) and use the rel canonical tag on the others so the A page benefits from any links going into them? This would take care of your duplicate content and duplicate title tag issues at the same time. While "technically speaking" the "B" page isn't really a non-canonical version of the "A" page, I think the tag would work well in this situation since nearly ALL of the content is duplicated.

Hello Mathew,

I did a site:domain.com search and do still see some of the old URLs indexed so I checked the URLs using an HTTP header status code checker and they are returning the correct 301 response. I also checked the the rel canonical tag on the new URLs and they do reference themselves, not the old URLs. Therefore I see no reason to be concerned about this issue. It takes time for Google to revisit those old URLs, see the redirect, and update their index. In time the old URLs will drop off and any links going into them should begin counting toward the pagerank of your new URLs.

HOW.Ever...

You have dozens of geotargeted doorway pages that Google probably doesn't like, or that at least violate their guidelines. If there was an office in each location it would be the right thing to do, as you would include the geo-specific address and phone number. Since every page has the same phone number and presumably there is only one office, you are running into the same problem many other "local" businesses have had to deal with over the years. Unfortunately, there still isn't any real solution and you will have real trouble ranking in the local/maps area on Google.

What to do about this is beyond the scope of this question, but if you're going to work with another SEO on this I'd recommend one who has experience with service-oriented business with multiple locations. This page would be a good place to start, and I have pre-filtered it to show only "local search" experts.

Good luck!

Did you disavow the remaining spammy links that you were unable to get rid of?

Also if you have another manual penalty (either site-wide or page-specific) could have a "time out" limit on it, which hasn't expired yet - as Tom alluded to above. See the attached video.

ES01L4xjSXE

I agree with Federico that you probably don't need to have every page be secure. Perhaps you should consider making the http://www. version your canonical default instead?

It is fine to have multiple versions of a site in different countries. Some of the biggest brands in the world do this. There are "right" and "wrong" ways to go about it, but if I had a ccTLD for the UK and lots of UK customers I wouldn't send them to my US site, regardless of whether I had a /uk/ folder or not.

Hello James,

Having a ccTLD definitely provides some ranking benefit in the target country. If the .ca domain is doing well for them I don't think I would consolidate it. Contrary to the other response, it is perfectly fine to have a separate ccTLD/domain for other countries.

However you might want to look into geotargeting each domain in Webmaster Tools (setting the .com version to US geotargeting may take care of the 2nd place brand search issue).

Of course this is just my opinion based on my own experiences. I will leave the question open for awhile so you can get more input from others.

Hello Rishad,

Even though this page says "Customizable meta data (title, description, keywords) they must not mean on every page because on this page a representative from Amazon says:

"There may be other solutions, but we don't currently have anything built into the Webstore which would allow you to modify the Product Detail page title. If you like, I can request the ability to the modify the Product Detail Page title creation as a Feature Request."

Later in the thread he continues with "I added the ability to modify the auto-generated page titles to our feature request list. Thanks for the great suggestion."

However that was a year ago and apparently they haven't gotten to it yet, which isn't surprising since I doubt Amazon wants other domains to compete with them at the product level, even if the owners ARE paying them for the privileged to compete.

Sorry.

I like Magento, and also Volusion if you are comfortable with being hosted on a Windows server. Though I have never used BigCommerce for one of my sites, I plan on checking them out in the near future because a few sites I've done audits for were on that platform and, at least from an SEO standpoint, they looked pretty solid. I have no idea what modules or plugins they were using though.

I think people may be misunderstanding your question and thinking you want to make a feed of your content on your own sites. You're wanting to create a feed of content you've written and published on several different article directories / databases so I think Yahoo Pipes will be a good bet: http://pipes.yahoo.com/pipes/ .

I should mention though that the reason these aren't indexed is probably more to do with the quality of article directories and the content found on them. Google doesn't tend to think very highly of them and I personally wouldn't build links using this type of site these days. Just my two cents.

Good luck!

First, it looks like Google is considering "oak furniture" synonymous with "oak beds" since they are bolding the word "furniture" even though i searched for "oak beds". This is pretty common, but that would answer your questions about anchor text and keyword use.

I see the H1 tag on the page. It isn't too terribly keyword stuffed compared to a lot of sites out there. It isn't hidden. As you can see, Google crawls that bit of text just fine:
http://webcache.googleusercontent.com/search?q=cache:Xim_4vDniwMJ:www.oakfurniturewarehouse.co.uk/+oak+beds&cd=2&hl=en&ct=clnk&gl=uk

I agree that their link profile does not seem good enough to merit the #2 position for such a competitive phrase, but links are not the only thing that matter. Searching on the Google.co.uk site means their ccTLD is going to help them over John Lewis and Dunelm Mill for that particular ranking factor. They may have really good user experience metrics, which we can't really tell much about. Above all, the site is very relevant since it has "oak furniture" in the domain and is specific to "oak" furniture, whereas some of the other sites sell lots of other types of furniture. I notice the #1 site is also oak-specific, has oak furniture in the domain, and is on a .co.uk ccTLD.

The truth is there is no way of knowing what exactly they are doing to rank above another site since there are so many variables, both known and unknown. However, from what I could see there isn't anything overtly spammy or blackhat going on, and the fact that they rank well for that phrase is certainly not outside of the realm of possibility.

You can give the affiliates code with nofollow in there but if they remove it you have no control over that. However, 99.99% of the time they'll leave it in there. Affiliates don't want to help you outrank them anyway.

As Wesley said above, have them use rel nofollow on the links and you should be fine.

That's a good rule of thumb from Brett Collins, though I typically make the alt attribute (technically it's an attribute to an image "tag" rather than a tag in and of itself, but it's just a semantics thing) descriptive of the image instead of the post - often it's the same thing either way. I know some people, myself included much of the time, who don't even use the title attribute unless it is within the href tag of a linking image. I know others who use it as the actual title of the image (e.g. funnydog.jpg would be "funny dog" and cats-are-evil.png would be "cats are evil"). I do typically make the alt and title attributes of an image two different things. I disable the creation of image attachment pages in Wordpress. When inserting individual images, ensure that the Link is set to File Url. When inserting a gallery, ensure that "Link thumbnails to" is set to Image File. You can also find plugins that will redirect the attachment pages. I typically block those in the robots.txt file too.

In the iDevAffilliate control panel you should be able to set the "default incoming traffic" URL and the "alternate incoming traffic" URLs to whatever you want. This would be a good opportunity to set up your affiliate program so ALL affiliate links go through a redirect on another domain or subdomain. For instance, a premium URL shortening service or a redirect script on another domain you own, or even one installed on a subdomain for that site.

This way you can kill off all of those links at once by removing the redirects so they don't affect the final destination page should you need to clean up your link profile. Whereas, if you had those links go straight to the destination page you'd have to ask the affiliates to remove each and every link, and/or use the disavow tool on them. To put it another way...

If spammy affiliate #123 links to www.YourSite.com/Product?affiliate-id=123 you would have to either get them to take the link down or use the Google Webmaster Central "disavow" tool (which is spotty, at best). Multiply this by hundreds or thousands of affiliates and it becomes a royal pain.

However, if spammy affiliate #123  links to http://adsever.yoursite.com/link-id=1 or www.URLShorten.er/asdf, which then redirects to www.YourSite.com/Product then you can simply remove the redirect on your end without ever having to deal with the affiliate, who will then be linking to a 404 page on a different domain, which is of little concern to you.

Hello BeytzNet,

It is not uncommon at all for ecommerce sites to have product variants like this, each with their own SKU. They are, after all, two different products. If someone ordered one color and got the other they would be upset. If someone searched Google Shopping for Gray Nike Shoes and ended up on a page for Pink Nike Shoes it would not be a good experience for them.

Yes, a better way to do this would be to have unique on-page content for each variant of this shoe, or even to have one page that allows the user to choose their color from a drop-down list (oh wait, Zappos does that too...) so the page isn't optimal, but it is unlikely that Google would see this as something worth applying a penalty for. They would more likely just decide to rank only one version. Rather than being sneaky, it is probably just a scalability problem.

With that said, I know lots of lesser-known brands and websites that have been hit hard by Panda for similar "scalability problems". The fact that big, well-known brands can get away with a lot more is something that has been going on for a long time and isn't about to change any time soon. So to answer your question "how do they get away with it" - They get away with it by being a huge, well-known brand. It sucks, but that apparently provides a better user experience for Google searchers. I don't think there is any malicious purpose to that (e.g. Adsense revenue, helping Google partner sites...), rather it has to do with the way we, as searchers, react to branding by clicking on the results we are already familiar with and buying from sites we already trust.

If I were to handle the same situation I'd probably choose a canonical version and redirect the other pages to it since writing unique copy for each color shoe wouldn't be scaleable for a site that size. Of course you would lose some ability to rank for color-specific searches, but you could minimize that by listing the colors out in title or on-page content while allowing the user to select the color from a drop-down.

Cody that is not accurate. Only one of the pages references ...10~2 as the canonical URL. The other ones uses <link rel="canonical" href="/nike-dart-10~1" />.

Hello Bryan,

Every page on the 9 old domains should redirect to the same page on on the new domain (.com/301#1/page1 and .com/301#1/page2) instead of redirecting them all to a single page.

Yes, the anchor text should carry through and it is worth being concerned about. This is a tough choice to make and I'm afraid I can't say for sure what would happen. I will leave this open as a discussion in case some other folks have experienced this exact scenario.

If it were me I would do one domain at a time, leaving a few weeks or months between them just to be sure. I would also make sure the link profiles for the redirecting domains were sparkle-clean.

Hello BJS,

The parameters look OK. Keep in mind that adjusting parameter handling in GWT is more of a band-aid than a cure. There are lots of other ways to handle faceted navigation, including rel canonical, robots meta tags, robots.txt and others. Read this article if you haven't already:

http://moz.com/blog/building-faceted-navigation-that-doesnt-suck

And here's a Whiteboard Friday on the topic: "http://moz.com/blog/whiteboard-friday-faceted-navigation

Keep a close eye on rankings once you've implemented those parameter handling rules. Be sure to annotate your analytics account with the date on which this was done for possible future diagnostics. See: http://www.google.com/analytics/features/annotations.html

Hello Erik,

I did notice two beginning and ending body tags in the source code. Search for this in View Source:

Hello

That would mean the only content on the page is "Hello" if Google was to be paying attention to that body tag instead of the other one.

I seriously doubt www Vs non-www would keep your site from being indexed. This is one of those really minor issues that Google does a really great job of handling 99% of the time.

I'm inclined to agree with Jesse that the site has been removed from the index due to the spammy links: http://www.opensiteexplorer.org/links?site=camilojosevergara.com

That is a bigger topic than we can cover here but you can start by looking into how to clean up your link profile, disavow links and file a re-inclusion request.

Assuming the site allows it and the links are followable I see no reason why it wouldn't be OK to link back to the clients' site from their answer, especially if the link goes to a page relevant to the answer (i.e. an FAQ page, help files, etc..).

Generally speaking, I think it's a good idea to move from EMD to a branded domain, especially if you can take the link profile with you. However, I'm not so sure about 301 redirecting 50 of them. Google supposedly treats mass-redirects to the home page as "soft 404s" and they may treat mass redirects to a single /EMD page the same. If you are going to replicate the site on the branded domain and redirect each page to its specific counterpart on the branded domain then you will probably be fine. If the goal is merely to redirect every page on a domain to a single page on another domain you may run into problems because all of that pagerank may not transfer over. Doing that 50 times could even bring on an unnatural link penalty for the branded domain.

If you could share some specifics the community may be able to help you device the best plan of action. Would each EMD redirect to a new directory (ie .com/edm1, .com/emd2) or would they all redirect to a single directory (i.e. .com/EMDs)? Would each page on the EMDs redirect to its specific counterpart on the branded domain, or would the entire EMD redirect to a single page on the branded domain? How many links do each of the EMDs have? Are we talking dozens, hundreds or thousands? What are the quality of those links? Out of the 41 that are NOT ranking well, is there any reason to suspect that they have a link-related penalty?

Hello Jesse,

I doubt 350 visits a day would slow down the site unless it is hosted on a shared server with a much-less-than-stellar hosting company. The high bounce rate on a few terms shouldn't affect rankings for the entire site, but would only affect those particular terms. If the bounce rate was that high on the majority of terms I'd be more concerned.

However, I do think you are correct to be suspicious of hacking. Have you tried looking at Google's cache of those pages, viewing the source of the cache, and searching for the phrases you are concerned with? Have you received any messages from Google either in the SERPs or in GWT related to the site being hacked?

Is it possible for you to share the site so we can take a look at it? This question is very difficult to answer without being able to diagnose the issue, as it could be due to many different reasons, including internal / external links, on-page copy, rel canonical tags, robots meta tags, robots.txt file, URL structure, title tags, anchor text... etc.

Hello Ilan,

Mozscape doesn't show any external links to either of the pages. The white gold page has far more internal links, according to Mozscape, with the anchor text "Gold Category" - though I don't see those links on the site now. Was something recently changed in regard to that?

I also noticed the navigation on the left links to a non-canonical version of the white gold page, and the canonical version of the yellow gold page. It is true that you have a rel canonical tag on both pages, but it might help to link directly to the canonical version from the sidebar navigation.

If you link from the white gold page to the yellow gold page body area (I always recommend having a short snippet of useful intro copy on category pages instead of just listing products) that would probably do the trick. Try something like this:

These 14-carat white gold necklaces make great gifts for your loved ones. Choose a style below and personalize it to the name of your choice! For a more classic gold look see our line of yellow gold name necklaces too.

This does three things...
#1. First and foremost it is useful. I was confused at first about whether I had to look for the name I wanted or if I would type in the name on checkout. Seeing only a few names to choose from, I may have just left assuming that you didn't have the name I wanted. The copy above makes it clear that I can customize for the name of my choice.

#2. It gives you an opportunity to use the keywords you are trying to rank for in a non-spammy way by describing the page. More is NOT better on category pages. Keep it short, simple and useful.

#3. You are linking to the page you want to rank for a keyword from the page Google "thinks" should rank for that keyword, which should be a big signal to Google that the yellow gold page should be ranking. In this case I would not link back to white gold from yellow gold in this way, as it could negate the strategy. You are already linking to both pages in the sidebar navigation anyway.

Good luck!

It's always good to see vendors helping out on here. Thanks Nathan!

The product I checked (see above) had rel canonical tags that used the first category (though not the second / sub-category). That is different than what you listed as "default" above.

Having the default rel canonical tag be .com/product-name-UniqueID/ is fine too. The important thing here is that you should be consistent. Also realize that using the rel canonical tag like this is sort of a temporary band-aid. Ideally you would be linking to the canonical URL and displaying the canonical URL instead of linking to and displaying the non-canonical URL on the site and relying on the rel canonical tag to "fix" it.

That should be fine for now. At some point in the future you probably want to make .com/product-name-UniqueID the version that gets linked to from elsewhere on the site (such as category pages) and for all other versions of that URL to 301 redirect to it.

If you ONLY sell in the other country you can switch over to the ccTLD and redirect the old site to the new one.

If you sell in the US AND the other country you can keep both sites up and avoid potential issues with duplicate content by using the strategies mentioned above.

Other members have given good advice on nofollowing those links. I just wanted to answer your other question: It would be more likely that the Penguin algorithm update would affect you in regard to this specific issue. If you truly were affected by Panda it probably has something to do with the quality of your site, specifically the content, as opposed to off-site factors like followable affiliate links.

You can use the product name and remove it if the technical sheets start to cannibalize rankings for the product name, which is highly doubtful. If you think that "Download PRODNAME technical sheet" is the best anchor text to describe to users what they're going to get when they click that link then use it.

Hello Mr. Vintage Heirloom,

Takeshi has some great points about priorities, and avoiding keyword stuffing your URLs. I might add, however, that putting category directories in product URLs has two major disadvantages that, in my opinion, usually outweigh the advantage of having those keywords in the URL.

#1 - If the product exists in multiple categories you risk having more than one URL for the product. This can be mitigated with redirects or rel canonical tags, but is still a pain. Here's an example:
http://www.vintageheirloom.com/vintage-chanel-bags/red-2.55-classic-double-flap-bag-1362483150
http://www.vintageheirloom.com/vintage-chanel-bags/2.55-bags/red-2.55-classic-double-flap-bag-1362483150
The bag's URL can be accessed from at least two different URLs (some products could have many more versions) and the canonical tag says that the shorter of the two URLs above is the canonical version. If that is the case the keywords in your /2.55-bags/ directory are useless as a ranking factor for that particular product page since that directory is not in the rel canonical tag.

Yes you can avoid the issues presented by multiple URL versions of the page, and some sites only ever put a product into a single category. However, that does nothing to account for this...

#2 - The deeper your category structure goes the further away from the root your product pages are. I have seen product pages five or six folders deep across entire eCommerce sites because of this. While I don't think the entire site architecture should be completely flat (some sort of taxonomy in the URLs is logical and useful) you don't want your most important pages to be several folders deep either.

I always recommend going with this:
site.com/products/product-name/
Or in your case:

site.com/products/product-name-uniqueID/

Putting the products into the /products/ directory is that level of useful taxonomy I mentioned above. This allows you, for instance, to do a search on Google like (site:domain.com inurl:products) to see how many of your product pages are indexed. The same type of logic is useful when segmenting analytics reports or WMT exports in Excel, among other uses.

Then you don't have to worry about keyword stuffing due to keywords already contained in the category directory portion of the URL.

This is just one person's opinion though. Some may disagree. I just don't find keywords in the URL to be all that important these days compared to other things. It has been spammed to death and thus the importance attributed to that factor has been steadily declining over the years, at least to my observation.

Regarding 301 redirects, they don't really cost you any appreciable amount of pagerank. It truly is negligible as long as you're not going through several redirect hops at once. The key is to make up your mind about your URLs with an eye to the future scalability and useability of the site - and stick with it. One round of redirects will temporarily set you back in the SERPs, but you should bounce back within a couple of weeks (good time of year to do them!) if done correctly.

Good luck!

Grumpy C,

Though this is new to you I can assure you that it is a VERY common issue we find in the SEO world. In other words, get used to it.

A cross-domain rel canonical tag should fix you right up, but in the long run I'd look into 301 redirecting or just removing those other domains:

http://support.google.com/webmasters/bin/answer.py?hl=en&answer=139394

http://www.seomoz.org/blog/rel-confused-answers-to-your-rel-canonical-questions

We used to call these "mirrored domains" which may help you when searching for more information on them. One thing I find useful in locating mirrors is called a "Reverse IP Lookup". You can find free tools all over the place for this. Here's one: http://www.yougetsignal.com/tools/web-sites-on-web-server/ . Before you get all freaked out when using this tool remember that shared hosting is very common so having multiple sites on the same IP is fine. What is not fine is if there are several of the SAME sites out there, as you are now dealing with.

Good luck!

Assuming I understand your specific questions correctly...

Having a /computer-repair/ directory with a /laptop-repair/ subdirectory should not cause you any problems in itself. However, if all of the content in /laptop-repair/* also appears in /computer-repair/* you could have a duplicate content problem to deal with. From your response to Thomas' extremely-detailed and helpful - though somewhat off-topic - comment I assume you already know that though.

I would not put the location in the directory, such as /virus-removal-wilmington-nc/. Whether or not you are moving makes no difference. I might make an exception there if you have multiple locations, but even then I think it would be a bad idea to have, for example:

/virus-removal-wilmington-nc/laptop-repair/

/virus-removal-charlotte-nc/laptop-repair/

...unless you had some very specific things to say about each location in regard to laptop repair that could not be expressed on the same page. These would be dangerously close to "doorway pages" which is common in any multi-location industry, and the bane of any SEO's existance when trying to clean up the penalties and filters that so often follow such a taxonomy.

Keep it simple. Having the location in the URL like that is a tiny, wee little factor that gets an inordinate amount of attention from SEOs, in my opinion. You are of course free to disagree with this, as you did when someone had a similar opinion in the other thread. In that case, let me answer your other question from that thread:

"If I include my current location (Wilmington NC) in the destination (New Domain) URL string for any given 301 redirect from my existing website to the new website and then physically move to another city 3 months later is this setting myself up for a BIG Failure? Seeing how I have no idea of how this technically works with GOOGLE as far as how long this (migration process) takes to Fully complete where the OLD domain completely drops off and everything is Fully passed over to the New Domain in terms leaving the 301 Re directs in place on the Old Domain Server. How long does this process usually take with GOOGLE?"

In my experience having done several dozen site migrations on sites of many different types and sizes, it should take Google no more than a few weeks to have their index completely updated for a site the size of yours. This is assuming that they are given the appropriate 301 response codes and are crawling the site regularly (i.e. no existing penalties or major algorithmic filters are in place that would limit crawling).

Switching your location like this is just one of several reasons I wouldn't put the location in the URL as proposed, so yes it is shooting yourself in the foot a little bit, but nothing that you won't be able to overcome. They will index the new URL, complete with the new location, and ranking calculations will be updated accordingly.

Regarding your other question toward the end of that thread, yes you can change the old htaccess file on the other server later, but there is no guarantee that Google will visit those old URLs again unless you have some external links pointing to them. If they came once, saw the redirect to the new URL, and there is no access to the old URL they may not see if you update the old URL's header response to point to a "new new" URL. If that makes sense. You could initiate the crawl by linking to the old URL from somewhere, such as a sitemap - but I'd advise making a "permanent" redirect permanent. In other words, I'd try to get it right the first time. Keeping the location out of that directory name would allow you to do that, unless I'm missing something.

Lastly, I disagree that domain.com/topic-in-charlotte-nc/sub-topic/ is "less" spammy than domain.com/topic/sub-topic/ but we all have a right to our opinions.

Good luck!

Just following-up to make sure you have your personalization turned off if you are checking your rankings manually. See Takeshi's answer below.

From Google's perspective: "If those pages are 'non-important' why are you linking to them from the navigation on every page?"

I didn't see any right-hand navigation, but I did see navigation on the left side. I do not see an issue with it, however, as this is an important way to get people and search engine spiders to category and eventually product pages. Perhaps you are thinking too hard about this so I will throw my opinion in along with the three other experienced SEOs above who all I think agree on this:

Do not nofollow your navigation links. Do not get rid of them. Do not worry about losing "link juice" into those pages.

If you truly think a page is unimportant, such as a category that doesn't sell many products, consider removing it from the navigation. Here's an example of how that might be done in this case:

I see the fly-out navigation to sub-categories includes all sub-categories for each main category. You could choose the top five sub-cats and then include a "See more categories" type of link instead of showing them all. This could be useful on a long list of sub-cats like the one under "Power Tool Accessories". You could limit the navigation to just two categories deep instead of including the third fly-out navigation level (e.g. Power Tool Accessories --> Bit Sets --> Socket Sets).

If the argument is that those third level categories are useful to visitors, thus you want to leave them there, then you have answered your question. Leave them. Don't worry about page-rank. If they are important enough for visitors to want them there and to use them, they are important enough to have their small share of pagerank.

If you choose to ignore all of this advice and want them to be there, but not pass pagerank, you can change the way your navigation operates by using a script so there aren't actually any href tags on the page for the navigation. All of the links would be controlled in the script, which would be placed in an external .js file, which could be inside a folder that is blocked in the robots.txt file. I don't recommend it, but that's one way to do it.

Honestly, most people just ignore it. But I'll leave the discussion open for awhile in case someone has some other solution/s to the issue and found them worth pursuing.

I can't imagine a modern eCommerce platform creating a new product page URL every time you added more stock for an existing product. Perhaps this is being done incorrectly? Does it have something to do with how the SKU or product ID is entered, or some other user error? If not, I'd suggest changing platforms.

In the meantime I guess rel canonicaling to one of them would work, but what happens when someone buys from that page/product ID? Is there only ever 1 in stock and, if so, does the URL go away/404/redirect when that happens? I think rel canonical would be a band-aid in this case.

Let us know how it works out. Which eCommerce platform is it?

Hello Shanaki,

I checked the http header response code for that URL and it returns a 301 redirect status code. In other words, the setting recommended above did work, but perhaps the SEOMoz crawler hasn't seen it yet.

SEOMoz does not crawl the web and return reports in real time. You may have to wait for the Mozscape index to refresh.

Hey, anyone remember when Google used to "dance" like that? Fun stuff. Popcorn-eating entertainment back then.

Paddy,

Thanks for clarifying. My mum is indeed very nice.

If we replace "my mom's blog" with "a slightly less trusted site, but still very similar in scale, topic, age, etc..." then you can see where it gets more difficult to base the decision on which has been crawled first.

But yes, I agree that being the first page crawled and indexed with that content must be a huge (though I think trust-based metrics would be more important) factor in the decision by Google's ranking algorithms.